Futhark

Vol 7 2016

International Journal of Runic Studies

Main editorsJames E Knirk and Henrik Williams

Assistant editorMarco Bianchi

copy Contributing authors 2017This work is licensed under a Creative Commons Attribution 40 International License (CC BY 40)

All articles are available free of charge at httpwwwfuthark-journalcom

A printed version of the issue can be ordered throughhttpurnkbseresolveurn=urnnbnseuudiva-309051

Editorial advisory board Michael P Barnes (University College London) Klaus Duumlwel (University of Goumlttingen) Lena Peterson (Uppsala University) Marie Stoklund (National Museum Copenhagen)

Typeset with Linux Libertine by Marco Bianchi

University of OsloUppsala University

ISSN 1892-0950

Published with financial support from the Nordic Publications Committee for Humanist and Social Sciences Periodicals (NOP-HS)

copy Contributing authors 2017This work is licensed under a Creative Commons Attribution 40 International License (CC BY 40)

All articles are available free of charge at httpwwwfuthark-journalcom

A printed version of the issue can be ordered throughhttpurnkbseresolveurn=urnnbnseuudiva-309051

Editorial advisory board Michael P Barnes (University College London) Klaus Duumlwel (University of Goumlttingen) Lena Peterson (Uppsala University) Marie Stoklund (National Museum Copenhagen)

Typeset with Linux Libertine by Marco Bianchi

University of OsloUppsala University

ISSN 1892-0950

Published with financial support from the Nordic Publications Committee for Humanist and Social Sciences Periodicals (NOP-HS)

Contents

Foreword 5

Bernard Mees The Hogganvik Inscription and Early Nordic Memorialisation 7

Wolfgang Beck Die Runeninschrift auf der Guumlrtelschnalle von Pforzen als Zeugnis der germanischen Heldensage 29

Luzius Thoumlny The Chronology of Final Devoicing and the Change of z to ʀ in Proto-Norse 47

Helmer Gustavson Tvaring runristade kopparamuletter fraringn Solberga Koumlpingsvik (Oumll Fv197696A och Oumll Fv197696B) 63

Elena A Melʹnikova A New Runic Inscription from Hagia Sophia Cathedral in Istanbul 101

Jana Kruumlger and Vivian Busch The Metrical Characteristics of Maeshowe Runic Inscription No 20 111

Short noticesJuliana Roost An Inscribed Fibula from Basel-Kleinhuumlningen 127Charlotte Boje Andersen and Lisbeth M Imer Ydby-stenen (DR 149)

genfundet 131Jan Owe Aringsa en mouml i Skaumlnninge (Oumlg 239) 137Magnus Kaumlllstroumlm Till tolkningen av runorna paring ett dryckeskaumlrl fraringn

Lund (DR EM85474A) 143Per Stille Johan Bures runtavla och dess titel 149

ReviewsMartin Findell Runes Reviewed by Mindy MacLeod 155Heikki Oja Riimut Viestejauml viikingeiltauml Reviewed by Kendra Willson 158Wolfgang Krause Schriften zur Runologie und Sprachwissenschaft

Reviewed by Martin Hannes Graf 164Klaus Duumlwel Runica minora Ausgewaumlhlte kleine Schriften zur

Runenkunde Reviewed by Patrik Larsson 170Irene Garciacutea Losquintildeo The Early Runic Inscriptions Their Western

Features Reviewed by Martin Hannes Graf 174Lisbeth M Imer and (photo) Roberto Fortuna Danmarks runesten En

fortelling Reviewed by Anne-Sofie Graumlslund 181

Florian Busch Runenschrift in der Black-Metal-Szene Skripturale Praktiken aus soziolinguistischer Perspektive Reviewed by Martin Findell 186

Contributors 193

The Hogganvik Inscription and Early Nordic Memorialisation

Bernard Mees (RMIT University Melbourne)

AbstractIn 2009 an early runic inscription was discovered on a triangular projecting area that through subsequent excavation was confirmed to be at the lower part of a funerary monument Yet such find reports and commentaries as have appeared to date have tended not to assess the Hogganvik inscription prin-ci pally as a commemorative expression as an example of a broader memo-rial epigraphic tradition Rather than as an epigraphic record of the history of emotions suggestions of magic appear in the main treatments of the remark-able find After all lexically irregular sequences found on other early runic memo rials are often taken as signs they feature a magical aspect Taking the Hoggan vik inscription in its broader linguistic and archaeological context how ever suggests a rather different understanding is to be assumed for the early Norwegian memorial Instead of reflecting magic the less clear sections of the Hoggan vik text can more regularly be understood as abbreviated or other wise obscurely expressed sequences

Keywords Hogganvik runestone (Vest-Agder) runic inscriptions history of emotions onomastics memo ri alisation curses abbreviations

Introduction

The question of what constitutes proper method in the humanities was of particular concern to scholars such as Wilhelm Dilthey In

his Einleitung in die Geisteswissenschaften (1883) Dilthey laid out his own under standing of what social scientists such as Max Weber (1922) would come to call Verstehenthinspmdashthinspinterpretative understanding of humanly derived expressions It remains rare however in runological discourse for inter pretative issues to be treated explicitly even in assessments of

Mees Bernard ldquoThe Hogganvik Inscription and Early Nordic Memorialisationrdquo Futhark International Journal of Runic Studies 7 (2016 publ 2017) 7ndash28

copy 2017 Bernard Mees This is an open-access article distributed under the terms of the CC BY 40 International License

and available free of charge at httpurnkbseresolveurn=urnnbnseuudiva-316585

Florian Busch Runenschrift in der Black-Metal-Szene Skripturale Praktiken aus soziolinguistischer Perspektive Reviewed by Martin Findell 186

Contributors 193

The Hogganvik Inscription and Early Nordic Memorialisation

Bernard Mees (RMIT University Melbourne)

AbstractIn 2009 an early runic inscription was discovered on a triangular projecting area that through subsequent excavation was confirmed to be at the lower part of a funerary monument Yet such find reports and commentaries as have appeared to date have tended not to assess the Hogganvik inscription prin-ci pally as a commemorative expression as an example of a broader memo-rial epigraphic tradition Rather than as an epigraphic record of the history of emotions suggestions of magic appear in the main treatments of the remark-able find After all lexically irregular sequences found on other early runic memo rials are often taken as signs they feature a magical aspect Taking the Hoggan vik inscription in its broader linguistic and archaeological context how ever suggests a rather different understanding is to be assumed for the early Norwegian memorial Instead of reflecting magic the less clear sections of the Hoggan vik text can more regularly be understood as abbreviated or other wise obscurely expressed sequences

Keywords Hogganvik runestone (Vest-Agder) runic inscriptions history of emotions onomastics memo ri alisation curses abbreviations

Introduction

The question of what constitutes proper method in the humanities was of particular concern to scholars such as Wilhelm Dilthey In

his Einleitung in die Geisteswissenschaften (1883) Dilthey laid out his own under standing of what social scientists such as Max Weber (1922) would come to call Verstehenthinspmdashthinspinterpretative understanding of humanly derived expressions It remains rare however in runological discourse for inter pretative issues to be treated explicitly even in assessments of

Mees Bernard ldquoThe Hogganvik Inscription and Early Nordic Memorialisationrdquo Futhark International Journal of Runic Studies 7 (2016 publ 2017) 7ndash28

copy 2017 Bernard Mees This is an open-access article distributed under the terms of the CC BY 40 International License

and available free of charge at httpurnkbseresolveurn=urnnbnseuudiva-316585

The Hogganvik Inscription and Early Nordic Memorialisation

Bernard Mees (RMIT University Melbourne)

AbstractIn 2009 an early runic inscription was discovered on a triangular projecting area that through subsequent excavation was confirmed to be at the lower part of a funerary monument Yet such find reports and commentaries as have appeared to date have tended not to assess the Hogganvik inscription prin-ci pally as a commemorative expression as an example of a broader memo-rial epigraphic tradition Rather than as an epigraphic record of the history of emotions suggestions of magic appear in the main treatments of the remark-able find After all lexically irregular sequences found on other early runic memo rials are often taken as signs they feature a magical aspect Taking the Hoggan vik inscription in its broader linguistic and archaeological context how ever suggests a rather different understanding is to be assumed for the early Norwegian memorial Instead of reflecting magic the less clear sections of the Hoggan vik text can more regularly be understood as abbreviated or other wise obscurely expressed sequences

Keywords Hogganvik runestone (Vest-Agder) runic inscriptions history of emotions onomastics memo ri alisation curses abbreviations

Introduction

The question of what constitutes proper method in the humanities was of particular concern to scholars such as Wilhelm Dilthey In

his Einleitung in die Geisteswissenschaften (1883) Dilthey laid out his own under standing of what social scientists such as Max Weber (1922) would come to call Verstehenthinspmdashthinspinterpretative understanding of humanly derived expressions It remains rare however in runological discourse for inter pretative issues to be treated explicitly even in assessments of

Mees Bernard ldquoThe Hogganvik Inscription and Early Nordic Memorialisationrdquo Futhark International Journal of Runic Studies 7 (2016 publ 2017) 7ndash28

copy 2017 Bernard Mees This is an open-access article distributed under the terms of the CC BY 40 International License

and available free of charge at httpurnkbseresolveurn=urnnbnseuudiva-316585

8 bull Bernard Mees

Futhark 7 (2016)

recently un covered finds such as the older runic memorial text unearthed at Hoggan vik Norway in 2009 The early runic texts are usually not approached in terms of broader devel opments in ancient or medieval histori og raphy such as the ldquoepigraphic habitrdquo of Roman experience first emphasised by Ramsay MacMullen (1982) or the more recent history of emotions approach to early medieval funerary memorials advanced by Barbara Rosen wein (2006 57ndash78) This includes previous interpretations of the Hoggan vik memorial discovered by Henrik and Arnfinn Henrik sen while clearing away stumps on their property in the village of Saringnum-Lunde vik in the Norwegian county of Vest-Agder A reflection of the long-discussed matter of interpretative method however can be seen in discussions of the role that historical imagination plays in runic studies partic ularly as the matter was set out by Ray Page in the first edition of his Introduction to English Runes (1973 13ndash15)

BackgroundIn the first edition of his ldquolittle red bookrdquo Page contrasts the approach to epi graphic interpretation of Karl Schneider (1956) with that of Erik Molt ke (cf Moltke 1985)thinspmdashthinspand even the extremely reactive stance taken by An-ders Baeligk sted (1952) Pagersquos main concern here was inter pretations of runic texts that are overly reliant on magical explanation often with out using any sort of formal substantiation of what magic is and what it may reason-ably be taken to constitute in a runic context (cf Page 1964 = 1995 105ndash25 Niel sen 1985) Runology has long been practised very much by scholars with the op posite approach to what Ulrich von Wilamo witz-Moel len dorf privately derided as ldquoDM-Wissenschaftrdquo (Braun et al 1995 232)

For Wilamowitz classical epigraphy was evidently a pedantic form of scholar ship that was overly obsessed with cataloguing relatively triv ial ex-pressions such as funerary epigraphs (DM or D(is) M(anibus) lsquoto the spirits of the deadrsquo being a common formula in Roman funerary inscrip tions) Such epi graphers were apparently so lacking in intel lec tual ambi tion they never felt able to venture beyond the bounds of their textsthinspmdashthinsptheir work never seemed to allow them to contribute anything of impor tance to broader scholar ship There is likewise usually little engage ment in ldquoscepticalrdquo runol ogy with the key inter pretative issue in historical analysisthinspmdashthinspie how to deal with what the historical philosopher (and epigrapher) R G Col-ling wood (1946) saw as the essentially unempir ical nature of historical under standing In 1967 Page had already written mockingly of ldquothat law of runol ogy which ordains that all unintel ligible inscriptions shall be thought

The Hogganvik Inscription and Early Nordic Memorialisation bull 9

Futhark 7 (2016)

magicalrdquo in reference to the ignotum per ignotius reasoning of Schneider But much of what is presented in the works of Page and his magic-abjuring followers remains limited to matt ers of kinds which saw them fail to de vel-op their scholarship beyond the level of descriptive empiricism

The main source of magical interpretations in runic studies at the time was the German comparativist Wolfgang Krause whose corpus of older runic inscriptions (Krause and Jankuhn 1966) is filled with con-jec ture of Pagersquos ldquoimaginativerdquo type The interpretations favoured by Krause however often make runic epigraphs appear more remarkable than comparable texts found in Roman tradition and suggest a kind of Northern exceptionalism that is asserted but not properly demonstrated The approach of many scholars since the 1970s has been to ally them-selves to the culturist ambitions of Krause but to reject his penchant for inter preting runic texts as magical None theless both of the scholars to have published extended treatments of the Hog gan vik inscription since 2009 can be faulted for ful filling Pagersquos law (the ldquoSecond Law of Runo-dynamicsrdquo according to Page in Duumlwel 1981 18) in their assessments of the early Nor wegian memorial They may also be criticised for failing to develop their treatments of the Hogganvik inscription to the fuller inter-pretative level that Marc Bloch (1954 [1949]) explained properly consti-tutes historical understanding

Taking the example of a Roman funerary inscription ldquocarved from a single block made for a single purposerdquo Bloch (pp 119 f) claimed that ldquonothing could be more variegated than the evidences which there await the probing of the scholarrsquos lancetrdquo Bloch admitted that he knew how to read Roman inscriptions but ldquonot how to cross-question themrdquo (Bloch 1954 54) Early runic memorial inscriptions might more profitably be ap proached in terms of Blochrsquos (p 71) ldquostruggle with docu mentsrdquo (p 53) ldquothe prime necessity of well-conducted historical researchrdquo E H Carr (1964 30) took Blochrsquos notion of interpretative struggle further describing a ldquocon tin uous process of interaction between a historian and his factsrdquo Rather than matters magical ancient memorials may be more con vinc ingly exam ined in light of MacMullenrsquos notion of an ldquoepi graph ic habitrdquo and the broader ldquoemotional turnrdquo (Plamper 2010) that has recently emerged elsewhere in historical studies

The Hogganvik inscriptionIn 2009 the first older runic memorial to have been discovered in Norway in over fifty years was announced in the Scandinavian press As head of

10 bull Bernard Mees

Futhark 7 (2016)

the Runic Archives in Oslo James Knirk was duly given access to the find eventually publishing a full report on the memorial and its discovery in the archaeological journal Viking in 2011 The University of Agderrsquos Michael Schulte also produced an analysis of the inscription which ap-peared in a Festschrift for the Dutch Nordicist Arend Quak that year after presenting a more typologically sophisticated study of the runestone text to the Agder Academy of Sciences in 2010 (and cf also Schulte 2013) The inscrip tion from Hogganvik proved an exceptional and exciting find for Nor wegian runology

The Hogganvik stone is 152 cm broad and 145 cm high is a reddish augen gneiss and is rounded at the top where the longest section of the inscription is found Dated to the late Roman Iron Age (ie AD c 150160ndash375400) by Imer (2015 122) on account of the rectangular form of its e-runes the Hogganvik inscription features four lines of text which are read by Knirk (2011 28) from right to left as

kelbathornewassṭainazaaasrpkfaarpaainananabozeknaudigastiz ekerafaz

Much of the Hogganvik text is fairly readily interpretable but not all parts of the inscription made immediate sense to Knirk and Schulte The in scrip tion clearly features an explanation that the stone belonged to a figure called Kelba thornewaz other early runic memorials often featuring the name of the memorialised in a genitival relationship to a suitably funer-ary object description such as stainaz lsquostonersquo The inscription also fea tures two first-person statements ek naudigastiz and ek erafaz with the lat-ter seeming to represent an expected erbaz (cf Old Norse [here after ON] jarfr (ierfr) lsquowolverinersquo) the apparent devoicing perhaps a sign of medial voicing of f (and hence the orthographic equivalence of -b- and -f-) emerging at the phonetic level cf the By stonersquos (-)laif- (KJ 71) vs the Mykle bostad memorialrsquos -[la]ib- (KJ 77 Schulte 2010 59 f) After all the erector of the Jaumlrs berg memorial (KJ 70) similarly describes himself by relating that he is called hellipubazhiteharabanazthinspmdashthinspie with an idio nym fol lowed by what seems best to be understood as an animal cog nomen (erafaz or lsquowolverinersquo at Hogganvik harabanaz or lsquoravenrsquo at Jaumlrs berg) But not much of the rest of the text on the Hogganvik stone could be inter preted precisely by either of the earliest publishers of inter pretations of the find

The Hogganvik Inscription and Early Nordic Memorialisation bull 11

Futhark 7 (2016)

Apart from the most obviously interpretable sections the Hoggan-vik inscription features two seemingly non-lexical sequences as well as a pre po sitional phrase that Knirk and Schulte interpret differently Both also suggest that the (apparently) non-lexical sequences may be magical citing a study by the German runologist Klaus Duumlwel (1988 = 2011) which seeks to explain the seemingly nonsensical early runic letter sequences that appear so commonly on the Old Germanic bracteates in terms of classical forms of letter magic But what a magical alphabetic sequence would mean on a runic memorial is not fully explored by Knirk or Schulte

The name Kelbathornewaz is one of several early runic examples of dithe-matic names in -thornewaz much as the element -gastiz is particularly well attested among the oldest Nordic finds (Peterson 2004) The element -thornewaz is also found as an independent lexical item in early runic inscriptions and is often interpreted literally as lsquoservantrsquo (rather than Krausersquos lsquoliege manrsquo) The first element kelba- is evidently a full-grade variant of the Common Ger manic word for lsquocalfrsquo (cf Old English [hereafter OE] cilfer Old High Ger man [hereafter OHG] chilburra lsquochilver ewe lambrsquo lt kelbizjō-uzjō cf Schulte 2010 52 Knirk 2011 31) A heroic interpretation lsquocalf-thanersquo (ie a name for a young warrior) also makes much better sense than (mere ly) lsquocalf-servantrsquothinspmdashthinspa thornewaz (as someone who lsquoservedrsquo militarily) presumably represented one of the middle-ranking figures (or liege men) who featured in Iron Age armies (cf Mees 2003 59 f Pauli Jensen et al 2003) Naudigastiz is more transparently to be translated literally as lsquoneed-guestrsquothinspmdashthinspie a guest who is (or has been) in needthinspmdashthinspan onomastic indexing of the early Germanic (and indeed Indo-European) tradition of (military) hospitality (Wesseacuten 1927 44 f cf Watkins 1996 246 Schulte 2011 63)

The obviously lexical part of the Hogganvik text for which the proper inter pretation is disputed is the collocation in(n)ana nabōz literally lsquofrom within the naversquo Schulte (2010 56 2013 124) notes that Old Norse nǫf can also refer to the corner of a house (presumably because the junctures of timber were seen to be comparable to that of the nave of a wooden wheel) but in some later dialects (and place-names) the same form can also mean lsquoprotruding rock elevated headland promontory capersquo evidently a devel op ment of the semantic lsquocornerrsquo Bjor vand and Linde man (2007 786 f) argue that two different etymological roots are involved with the mean ing lsquocorner of housersquo semantically unrelated to lsquonaversquo At any rate compar able runic memorials do not feature the preposition in(n)ana nor do they clear ly feature references to headlands wheel naves or the corners of houses

12 bull Bernard Mees

Futhark 7 (2016)

The early runic memorial habit

It is quite clear however even from a superficial survey of the somewhat over forty early runestone inscriptions which have survived that textual formulas and subtypes can be isolated within the overall older funerary genre Memorial epigraphy is typically formulaicthinspmdashthinspfunerary inscriptions usually repre sent the clearest epigraphic testimony for what Weber (1922 12ndash16) characterised as traditional action (cf Mees 2013 327 f) Much like sets of Iron Age grave goods funerary epigraphs typically accord to standard typol ogies and feature recurrent formulas because they are solemn emo tional expressions that are public in character (cf Rosenwein 2006 61) Like Roman memorials they clearly reflect a form of ldquoepigraphic habitrdquo (MacMullen 1982) and often feature little more than indications of who the memorialised was and who set the monument up Since all forms of memo rial epigraphy are stereotypical the disagreement which has arisen in inter pretations of the Hogganvik inscription may perhaps be resolved by a fuller consideration of the fundamental formulaic characteristics of the older runic memorial habit

The Hogganvik inscription was found near the remains of an Iron Age grave yard or burial site (Gloslashrstad et al 2011) and clearly belongs to the early Nordic genre of runestone texts that characteristically feature a manrsquos name in the genitive and a labelling of the associated grave or memo rial stone The Boslash stonersquos hnabudas hlaiwa lsquoHnabudazrsquos graversquo Sten stadrsquos igijon halaz lsquoIngijorsquos stonersquo and the similar hariẉulfsmiddotstᴀinᴀz lsquoHari wulfzrsquos stonesrsquo on the transitionalyounger Raumlvsal memorial are parade examples of this labelling type (KJ 78 KJ 81 and KJ 80 respectively cf Anton sen 2002 191 and Schulte 2010 49 f 2013 121) In contrast early runic magico-religious inscriptions often feature item descriptions and naming expressions or names but these are never brought together syn tac tically in the (genitival) manner seen at Hogganvik (MacLeod and Mees 2004 2006 71ndash101) Early Nordic runestone texts sometimes also feature what are usually taken to represent locativising prepo sitional phrases (another feature absent from early runic magico-religious texts) but the proper interpretation of such expressions has often been debated The Moumljbro inscription features the form anahahai taken by Krause and also Antonsen as ana ha(n)hai lsquoon a steedrsquo (cf ON hestr lt hanhistaz lsquostallionrsquo)thinspmdashthinspie as a description which reflects the image of a horse and rider that also appears on the stone (KJ 99 Antonsen 1975 no 11) Yet hanhaz is not hanhistazthinspmdashthinspa form hanhaz is not paralleled in Germanic with a meaning lsquosteedrsquo Hence Staffan Fridell (2008 2009)

The Hogganvik Inscription and Early Nordic Memorialisation bull 13

Futhark 7 (2016)

has more recently suggested a toponymic interpretation lsquoat Haringrsquo with hahai instead representing the name of the place where the memorialised warrior died (although the apparent Sanskrit cognate śankuacute- lsquopeg spikersquo and Old Church Slavonic sǫkŭ lsquobranchrsquo suggest that Haring was a u-stem in Ger manic) The Rouml memorial features what seems to have been meant as another prepositional phrase beginning with the preposition ana but with the rest of the line largely unreadable today (KJ 73) Even less clear is the inscription on the Nordhuglo stone that instead finishes simply with the sequence ih which has often been supposed (since Bugge NIaeligR no 49) to represent an abbreviation for in Hugla lsquoin Huglorsquo (KJ 65) although this interpretation is rather speculative The Hogganvik text however reads in(n)ana lsquowithinrsquo not in lsquoinrsquo or ana lsquoat uponrsquo as would be expected in connection with a place-name Schultersquos connection of naboz with a geographical description of the location of the monument lsquowithin the area of the protruding rocksrsquo seems rather unlikely in this light

Indeed the preposition in(n)ana clearly represents a locativising ex pres-sion which etymologically means lsquofrom withinrsquo the suffix -na being direc-tional (cf Latin -nē in supernē lsquofrom aboversquo Goth utana lsquofrom withoutrsquo aftana lsquofrom behindrsquo etc Schmidt 1962 178ndash81 and 183 f) Surely the semantic which fits best with a directional lsquofrom withinrsquo is lsquocorner of a housersquothinspmdashthinspie a formation of a type which is widely paralleled in Old Norse cf ON innan borgar lsquowithin the townrsquo innan hallar lsquowithin the hallrsquo innan veggja lsquowithin the wallsrsquo and (especially) innandura innan gareths innan gaacutetta innanhuacutess and innanstokks all of which signify lsquoat home in doorsrsquo The Hogganvik expression inananaboz can be understood as meto nymic thenthinspmdashthinspie with the literal meaning lsquofrom within the corner of a housersquo indicating lsquoat home within the cornerswalls of his housersquo If the use albeit limited of prepositional phrases in other runestone memorials can be used as a guide then the description inananaboz looks as if it indicates that Kelba thornewaz died at home rather than (more heroically) out of doors No indication comparable to the Moumljbro stonersquos slaginaz lsquoslainrsquo is given but the Moumljbro inscription is unique among the older runic texts by featuring an explicit verbal indication of dying

Non-lexical sequencesThe seemingly non-lexical sequences on the Hogganvik stone are linked by both Knirk and Schulte with magical interpretations of comparable orderings which appear on early runic amulets Nordeacuten (1934 1940) inter-preted several of the early Nordic runestone inscriptions as funer ary curses

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 9

Futhark 7 (2016)

magicalrdquo in reference to the ignotum per ignotius reasoning of Schneider But much of what is presented in the works of Page and his magic-abjuring followers remains limited to matt ers of kinds which saw them fail to de vel-op their scholarship beyond the level of descriptive empiricism

The main source of magical interpretations in runic studies at the time was the German comparativist Wolfgang Krause whose corpus of older runic inscriptions (Krause and Jankuhn 1966) is filled with con-jec ture of Pagersquos ldquoimaginativerdquo type The interpretations favoured by Krause however often make runic epigraphs appear more remarkable than comparable texts found in Roman tradition and suggest a kind of Northern exceptionalism that is asserted but not properly demonstrated The approach of many scholars since the 1970s has been to ally them-selves to the culturist ambitions of Krause but to reject his penchant for inter preting runic texts as magical None theless both of the scholars to have published extended treatments of the Hog gan vik inscription since 2009 can be faulted for ful filling Pagersquos law (the ldquoSecond Law of Runo-dynamicsrdquo according to Page in Duumlwel 1981 18) in their assessments of the early Nor wegian memorial They may also be criticised for failing to develop their treatments of the Hogganvik inscription to the fuller inter-pretative level that Marc Bloch (1954 [1949]) explained properly consti-tutes historical understanding

Taking the example of a Roman funerary inscription ldquocarved from a single block made for a single purposerdquo Bloch (pp 119 f) claimed that ldquonothing could be more variegated than the evidences which there await the probing of the scholarrsquos lancetrdquo Bloch admitted that he knew how to read Roman inscriptions but ldquonot how to cross-question themrdquo (Bloch 1954 54) Early runic memorial inscriptions might more profitably be ap proached in terms of Blochrsquos (p 71) ldquostruggle with docu mentsrdquo (p 53) ldquothe prime necessity of well-conducted historical researchrdquo E H Carr (1964 30) took Blochrsquos notion of interpretative struggle further describing a ldquocon tin uous process of interaction between a historian and his factsrdquo Rather than matters magical ancient memorials may be more con vinc ingly exam ined in light of MacMullenrsquos notion of an ldquoepi graph ic habitrdquo and the broader ldquoemotional turnrdquo (Plamper 2010) that has recently emerged elsewhere in historical studies

The Hogganvik inscriptionIn 2009 the first older runic memorial to have been discovered in Norway in over fifty years was announced in the Scandinavian press As head of

10 bull Bernard Mees

Futhark 7 (2016)

the Runic Archives in Oslo James Knirk was duly given access to the find eventually publishing a full report on the memorial and its discovery in the archaeological journal Viking in 2011 The University of Agderrsquos Michael Schulte also produced an analysis of the inscription which ap-peared in a Festschrift for the Dutch Nordicist Arend Quak that year after presenting a more typologically sophisticated study of the runestone text to the Agder Academy of Sciences in 2010 (and cf also Schulte 2013) The inscrip tion from Hogganvik proved an exceptional and exciting find for Nor wegian runology

The Hogganvik stone is 152 cm broad and 145 cm high is a reddish augen gneiss and is rounded at the top where the longest section of the inscription is found Dated to the late Roman Iron Age (ie AD c 150160ndash375400) by Imer (2015 122) on account of the rectangular form of its e-runes the Hogganvik inscription features four lines of text which are read by Knirk (2011 28) from right to left as

kelbathornewassṭainazaaasrpkfaarpaainananabozeknaudigastiz ekerafaz

Much of the Hogganvik text is fairly readily interpretable but not all parts of the inscription made immediate sense to Knirk and Schulte The in scrip tion clearly features an explanation that the stone belonged to a figure called Kelba thornewaz other early runic memorials often featuring the name of the memorialised in a genitival relationship to a suitably funer-ary object description such as stainaz lsquostonersquo The inscription also fea tures two first-person statements ek naudigastiz and ek erafaz with the lat-ter seeming to represent an expected erbaz (cf Old Norse [here after ON] jarfr (ierfr) lsquowolverinersquo) the apparent devoicing perhaps a sign of medial voicing of f (and hence the orthographic equivalence of -b- and -f-) emerging at the phonetic level cf the By stonersquos (-)laif- (KJ 71) vs the Mykle bostad memorialrsquos -[la]ib- (KJ 77 Schulte 2010 59 f) After all the erector of the Jaumlrs berg memorial (KJ 70) similarly describes himself by relating that he is called hellipubazhiteharabanazthinspmdashthinspie with an idio nym fol lowed by what seems best to be understood as an animal cog nomen (erafaz or lsquowolverinersquo at Hogganvik harabanaz or lsquoravenrsquo at Jaumlrs berg) But not much of the rest of the text on the Hogganvik stone could be inter preted precisely by either of the earliest publishers of inter pretations of the find

The Hogganvik Inscription and Early Nordic Memorialisation bull 11

Futhark 7 (2016)

Apart from the most obviously interpretable sections the Hoggan-vik inscription features two seemingly non-lexical sequences as well as a pre po sitional phrase that Knirk and Schulte interpret differently Both also suggest that the (apparently) non-lexical sequences may be magical citing a study by the German runologist Klaus Duumlwel (1988 = 2011) which seeks to explain the seemingly nonsensical early runic letter sequences that appear so commonly on the Old Germanic bracteates in terms of classical forms of letter magic But what a magical alphabetic sequence would mean on a runic memorial is not fully explored by Knirk or Schulte

The name Kelbathornewaz is one of several early runic examples of dithe-matic names in -thornewaz much as the element -gastiz is particularly well attested among the oldest Nordic finds (Peterson 2004) The element -thornewaz is also found as an independent lexical item in early runic inscriptions and is often interpreted literally as lsquoservantrsquo (rather than Krausersquos lsquoliege manrsquo) The first element kelba- is evidently a full-grade variant of the Common Ger manic word for lsquocalfrsquo (cf Old English [hereafter OE] cilfer Old High Ger man [hereafter OHG] chilburra lsquochilver ewe lambrsquo lt kelbizjō-uzjō cf Schulte 2010 52 Knirk 2011 31) A heroic interpretation lsquocalf-thanersquo (ie a name for a young warrior) also makes much better sense than (mere ly) lsquocalf-servantrsquothinspmdashthinspa thornewaz (as someone who lsquoservedrsquo militarily) presumably represented one of the middle-ranking figures (or liege men) who featured in Iron Age armies (cf Mees 2003 59 f Pauli Jensen et al 2003) Naudigastiz is more transparently to be translated literally as lsquoneed-guestrsquothinspmdashthinspie a guest who is (or has been) in needthinspmdashthinspan onomastic indexing of the early Germanic (and indeed Indo-European) tradition of (military) hospitality (Wesseacuten 1927 44 f cf Watkins 1996 246 Schulte 2011 63)

The obviously lexical part of the Hogganvik text for which the proper inter pretation is disputed is the collocation in(n)ana nabōz literally lsquofrom within the naversquo Schulte (2010 56 2013 124) notes that Old Norse nǫf can also refer to the corner of a house (presumably because the junctures of timber were seen to be comparable to that of the nave of a wooden wheel) but in some later dialects (and place-names) the same form can also mean lsquoprotruding rock elevated headland promontory capersquo evidently a devel op ment of the semantic lsquocornerrsquo Bjor vand and Linde man (2007 786 f) argue that two different etymological roots are involved with the mean ing lsquocorner of housersquo semantically unrelated to lsquonaversquo At any rate compar able runic memorials do not feature the preposition in(n)ana nor do they clear ly feature references to headlands wheel naves or the corners of houses

12 bull Bernard Mees

Futhark 7 (2016)

The early runic memorial habit

It is quite clear however even from a superficial survey of the somewhat over forty early runestone inscriptions which have survived that textual formulas and subtypes can be isolated within the overall older funerary genre Memorial epigraphy is typically formulaicthinspmdashthinspfunerary inscriptions usually repre sent the clearest epigraphic testimony for what Weber (1922 12ndash16) characterised as traditional action (cf Mees 2013 327 f) Much like sets of Iron Age grave goods funerary epigraphs typically accord to standard typol ogies and feature recurrent formulas because they are solemn emo tional expressions that are public in character (cf Rosenwein 2006 61) Like Roman memorials they clearly reflect a form of ldquoepigraphic habitrdquo (MacMullen 1982) and often feature little more than indications of who the memorialised was and who set the monument up Since all forms of memo rial epigraphy are stereotypical the disagreement which has arisen in inter pretations of the Hogganvik inscription may perhaps be resolved by a fuller consideration of the fundamental formulaic characteristics of the older runic memorial habit

The Hogganvik inscription was found near the remains of an Iron Age grave yard or burial site (Gloslashrstad et al 2011) and clearly belongs to the early Nordic genre of runestone texts that characteristically feature a manrsquos name in the genitive and a labelling of the associated grave or memo rial stone The Boslash stonersquos hnabudas hlaiwa lsquoHnabudazrsquos graversquo Sten stadrsquos igijon halaz lsquoIngijorsquos stonersquo and the similar hariẉulfsmiddotstᴀinᴀz lsquoHari wulfzrsquos stonesrsquo on the transitionalyounger Raumlvsal memorial are parade examples of this labelling type (KJ 78 KJ 81 and KJ 80 respectively cf Anton sen 2002 191 and Schulte 2010 49 f 2013 121) In contrast early runic magico-religious inscriptions often feature item descriptions and naming expressions or names but these are never brought together syn tac tically in the (genitival) manner seen at Hogganvik (MacLeod and Mees 2004 2006 71ndash101) Early Nordic runestone texts sometimes also feature what are usually taken to represent locativising prepo sitional phrases (another feature absent from early runic magico-religious texts) but the proper interpretation of such expressions has often been debated The Moumljbro inscription features the form anahahai taken by Krause and also Antonsen as ana ha(n)hai lsquoon a steedrsquo (cf ON hestr lt hanhistaz lsquostallionrsquo)thinspmdashthinspie as a description which reflects the image of a horse and rider that also appears on the stone (KJ 99 Antonsen 1975 no 11) Yet hanhaz is not hanhistazthinspmdashthinspa form hanhaz is not paralleled in Germanic with a meaning lsquosteedrsquo Hence Staffan Fridell (2008 2009)

The Hogganvik Inscription and Early Nordic Memorialisation bull 13

Futhark 7 (2016)

has more recently suggested a toponymic interpretation lsquoat Haringrsquo with hahai instead representing the name of the place where the memorialised warrior died (although the apparent Sanskrit cognate śankuacute- lsquopeg spikersquo and Old Church Slavonic sǫkŭ lsquobranchrsquo suggest that Haring was a u-stem in Ger manic) The Rouml memorial features what seems to have been meant as another prepositional phrase beginning with the preposition ana but with the rest of the line largely unreadable today (KJ 73) Even less clear is the inscription on the Nordhuglo stone that instead finishes simply with the sequence ih which has often been supposed (since Bugge NIaeligR no 49) to represent an abbreviation for in Hugla lsquoin Huglorsquo (KJ 65) although this interpretation is rather speculative The Hogganvik text however reads in(n)ana lsquowithinrsquo not in lsquoinrsquo or ana lsquoat uponrsquo as would be expected in connection with a place-name Schultersquos connection of naboz with a geographical description of the location of the monument lsquowithin the area of the protruding rocksrsquo seems rather unlikely in this light

Indeed the preposition in(n)ana clearly represents a locativising ex pres-sion which etymologically means lsquofrom withinrsquo the suffix -na being direc-tional (cf Latin -nē in supernē lsquofrom aboversquo Goth utana lsquofrom withoutrsquo aftana lsquofrom behindrsquo etc Schmidt 1962 178ndash81 and 183 f) Surely the semantic which fits best with a directional lsquofrom withinrsquo is lsquocorner of a housersquothinspmdashthinspie a formation of a type which is widely paralleled in Old Norse cf ON innan borgar lsquowithin the townrsquo innan hallar lsquowithin the hallrsquo innan veggja lsquowithin the wallsrsquo and (especially) innandura innan gareths innan gaacutetta innanhuacutess and innanstokks all of which signify lsquoat home in doorsrsquo The Hogganvik expression inananaboz can be understood as meto nymic thenthinspmdashthinspie with the literal meaning lsquofrom within the corner of a housersquo indicating lsquoat home within the cornerswalls of his housersquo If the use albeit limited of prepositional phrases in other runestone memorials can be used as a guide then the description inananaboz looks as if it indicates that Kelba thornewaz died at home rather than (more heroically) out of doors No indication comparable to the Moumljbro stonersquos slaginaz lsquoslainrsquo is given but the Moumljbro inscription is unique among the older runic texts by featuring an explicit verbal indication of dying

Non-lexical sequencesThe seemingly non-lexical sequences on the Hogganvik stone are linked by both Knirk and Schulte with magical interpretations of comparable orderings which appear on early runic amulets Nordeacuten (1934 1940) inter-preted several of the early Nordic runestone inscriptions as funer ary curses

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

10 bull Bernard Mees

Futhark 7 (2016)

the Runic Archives in Oslo James Knirk was duly given access to the find eventually publishing a full report on the memorial and its discovery in the archaeological journal Viking in 2011 The University of Agderrsquos Michael Schulte also produced an analysis of the inscription which ap-peared in a Festschrift for the Dutch Nordicist Arend Quak that year after presenting a more typologically sophisticated study of the runestone text to the Agder Academy of Sciences in 2010 (and cf also Schulte 2013) The inscrip tion from Hogganvik proved an exceptional and exciting find for Nor wegian runology

The Hogganvik stone is 152 cm broad and 145 cm high is a reddish augen gneiss and is rounded at the top where the longest section of the inscription is found Dated to the late Roman Iron Age (ie AD c 150160ndash375400) by Imer (2015 122) on account of the rectangular form of its e-runes the Hogganvik inscription features four lines of text which are read by Knirk (2011 28) from right to left as

kelbathornewassṭainazaaasrpkfaarpaainananabozeknaudigastiz ekerafaz

Much of the Hogganvik text is fairly readily interpretable but not all parts of the inscription made immediate sense to Knirk and Schulte The in scrip tion clearly features an explanation that the stone belonged to a figure called Kelba thornewaz other early runic memorials often featuring the name of the memorialised in a genitival relationship to a suitably funer-ary object description such as stainaz lsquostonersquo The inscription also fea tures two first-person statements ek naudigastiz and ek erafaz with the lat-ter seeming to represent an expected erbaz (cf Old Norse [here after ON] jarfr (ierfr) lsquowolverinersquo) the apparent devoicing perhaps a sign of medial voicing of f (and hence the orthographic equivalence of -b- and -f-) emerging at the phonetic level cf the By stonersquos (-)laif- (KJ 71) vs the Mykle bostad memorialrsquos -[la]ib- (KJ 77 Schulte 2010 59 f) After all the erector of the Jaumlrs berg memorial (KJ 70) similarly describes himself by relating that he is called hellipubazhiteharabanazthinspmdashthinspie with an idio nym fol lowed by what seems best to be understood as an animal cog nomen (erafaz or lsquowolverinersquo at Hogganvik harabanaz or lsquoravenrsquo at Jaumlrs berg) But not much of the rest of the text on the Hogganvik stone could be inter preted precisely by either of the earliest publishers of inter pretations of the find

The Hogganvik Inscription and Early Nordic Memorialisation bull 11

Futhark 7 (2016)

Apart from the most obviously interpretable sections the Hoggan-vik inscription features two seemingly non-lexical sequences as well as a pre po sitional phrase that Knirk and Schulte interpret differently Both also suggest that the (apparently) non-lexical sequences may be magical citing a study by the German runologist Klaus Duumlwel (1988 = 2011) which seeks to explain the seemingly nonsensical early runic letter sequences that appear so commonly on the Old Germanic bracteates in terms of classical forms of letter magic But what a magical alphabetic sequence would mean on a runic memorial is not fully explored by Knirk or Schulte

The name Kelbathornewaz is one of several early runic examples of dithe-matic names in -thornewaz much as the element -gastiz is particularly well attested among the oldest Nordic finds (Peterson 2004) The element -thornewaz is also found as an independent lexical item in early runic inscriptions and is often interpreted literally as lsquoservantrsquo (rather than Krausersquos lsquoliege manrsquo) The first element kelba- is evidently a full-grade variant of the Common Ger manic word for lsquocalfrsquo (cf Old English [hereafter OE] cilfer Old High Ger man [hereafter OHG] chilburra lsquochilver ewe lambrsquo lt kelbizjō-uzjō cf Schulte 2010 52 Knirk 2011 31) A heroic interpretation lsquocalf-thanersquo (ie a name for a young warrior) also makes much better sense than (mere ly) lsquocalf-servantrsquothinspmdashthinspa thornewaz (as someone who lsquoservedrsquo militarily) presumably represented one of the middle-ranking figures (or liege men) who featured in Iron Age armies (cf Mees 2003 59 f Pauli Jensen et al 2003) Naudigastiz is more transparently to be translated literally as lsquoneed-guestrsquothinspmdashthinspie a guest who is (or has been) in needthinspmdashthinspan onomastic indexing of the early Germanic (and indeed Indo-European) tradition of (military) hospitality (Wesseacuten 1927 44 f cf Watkins 1996 246 Schulte 2011 63)

The obviously lexical part of the Hogganvik text for which the proper inter pretation is disputed is the collocation in(n)ana nabōz literally lsquofrom within the naversquo Schulte (2010 56 2013 124) notes that Old Norse nǫf can also refer to the corner of a house (presumably because the junctures of timber were seen to be comparable to that of the nave of a wooden wheel) but in some later dialects (and place-names) the same form can also mean lsquoprotruding rock elevated headland promontory capersquo evidently a devel op ment of the semantic lsquocornerrsquo Bjor vand and Linde man (2007 786 f) argue that two different etymological roots are involved with the mean ing lsquocorner of housersquo semantically unrelated to lsquonaversquo At any rate compar able runic memorials do not feature the preposition in(n)ana nor do they clear ly feature references to headlands wheel naves or the corners of houses

12 bull Bernard Mees

Futhark 7 (2016)

The early runic memorial habit

It is quite clear however even from a superficial survey of the somewhat over forty early runestone inscriptions which have survived that textual formulas and subtypes can be isolated within the overall older funerary genre Memorial epigraphy is typically formulaicthinspmdashthinspfunerary inscriptions usually repre sent the clearest epigraphic testimony for what Weber (1922 12ndash16) characterised as traditional action (cf Mees 2013 327 f) Much like sets of Iron Age grave goods funerary epigraphs typically accord to standard typol ogies and feature recurrent formulas because they are solemn emo tional expressions that are public in character (cf Rosenwein 2006 61) Like Roman memorials they clearly reflect a form of ldquoepigraphic habitrdquo (MacMullen 1982) and often feature little more than indications of who the memorialised was and who set the monument up Since all forms of memo rial epigraphy are stereotypical the disagreement which has arisen in inter pretations of the Hogganvik inscription may perhaps be resolved by a fuller consideration of the fundamental formulaic characteristics of the older runic memorial habit

The Hogganvik inscription was found near the remains of an Iron Age grave yard or burial site (Gloslashrstad et al 2011) and clearly belongs to the early Nordic genre of runestone texts that characteristically feature a manrsquos name in the genitive and a labelling of the associated grave or memo rial stone The Boslash stonersquos hnabudas hlaiwa lsquoHnabudazrsquos graversquo Sten stadrsquos igijon halaz lsquoIngijorsquos stonersquo and the similar hariẉulfsmiddotstᴀinᴀz lsquoHari wulfzrsquos stonesrsquo on the transitionalyounger Raumlvsal memorial are parade examples of this labelling type (KJ 78 KJ 81 and KJ 80 respectively cf Anton sen 2002 191 and Schulte 2010 49 f 2013 121) In contrast early runic magico-religious inscriptions often feature item descriptions and naming expressions or names but these are never brought together syn tac tically in the (genitival) manner seen at Hogganvik (MacLeod and Mees 2004 2006 71ndash101) Early Nordic runestone texts sometimes also feature what are usually taken to represent locativising prepo sitional phrases (another feature absent from early runic magico-religious texts) but the proper interpretation of such expressions has often been debated The Moumljbro inscription features the form anahahai taken by Krause and also Antonsen as ana ha(n)hai lsquoon a steedrsquo (cf ON hestr lt hanhistaz lsquostallionrsquo)thinspmdashthinspie as a description which reflects the image of a horse and rider that also appears on the stone (KJ 99 Antonsen 1975 no 11) Yet hanhaz is not hanhistazthinspmdashthinspa form hanhaz is not paralleled in Germanic with a meaning lsquosteedrsquo Hence Staffan Fridell (2008 2009)

The Hogganvik Inscription and Early Nordic Memorialisation bull 13

Futhark 7 (2016)

has more recently suggested a toponymic interpretation lsquoat Haringrsquo with hahai instead representing the name of the place where the memorialised warrior died (although the apparent Sanskrit cognate śankuacute- lsquopeg spikersquo and Old Church Slavonic sǫkŭ lsquobranchrsquo suggest that Haring was a u-stem in Ger manic) The Rouml memorial features what seems to have been meant as another prepositional phrase beginning with the preposition ana but with the rest of the line largely unreadable today (KJ 73) Even less clear is the inscription on the Nordhuglo stone that instead finishes simply with the sequence ih which has often been supposed (since Bugge NIaeligR no 49) to represent an abbreviation for in Hugla lsquoin Huglorsquo (KJ 65) although this interpretation is rather speculative The Hogganvik text however reads in(n)ana lsquowithinrsquo not in lsquoinrsquo or ana lsquoat uponrsquo as would be expected in connection with a place-name Schultersquos connection of naboz with a geographical description of the location of the monument lsquowithin the area of the protruding rocksrsquo seems rather unlikely in this light

Indeed the preposition in(n)ana clearly represents a locativising ex pres-sion which etymologically means lsquofrom withinrsquo the suffix -na being direc-tional (cf Latin -nē in supernē lsquofrom aboversquo Goth utana lsquofrom withoutrsquo aftana lsquofrom behindrsquo etc Schmidt 1962 178ndash81 and 183 f) Surely the semantic which fits best with a directional lsquofrom withinrsquo is lsquocorner of a housersquothinspmdashthinspie a formation of a type which is widely paralleled in Old Norse cf ON innan borgar lsquowithin the townrsquo innan hallar lsquowithin the hallrsquo innan veggja lsquowithin the wallsrsquo and (especially) innandura innan gareths innan gaacutetta innanhuacutess and innanstokks all of which signify lsquoat home in doorsrsquo The Hogganvik expression inananaboz can be understood as meto nymic thenthinspmdashthinspie with the literal meaning lsquofrom within the corner of a housersquo indicating lsquoat home within the cornerswalls of his housersquo If the use albeit limited of prepositional phrases in other runestone memorials can be used as a guide then the description inananaboz looks as if it indicates that Kelba thornewaz died at home rather than (more heroically) out of doors No indication comparable to the Moumljbro stonersquos slaginaz lsquoslainrsquo is given but the Moumljbro inscription is unique among the older runic texts by featuring an explicit verbal indication of dying

Non-lexical sequencesThe seemingly non-lexical sequences on the Hogganvik stone are linked by both Knirk and Schulte with magical interpretations of comparable orderings which appear on early runic amulets Nordeacuten (1934 1940) inter-preted several of the early Nordic runestone inscriptions as funer ary curses

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 11

Futhark 7 (2016)

Apart from the most obviously interpretable sections the Hoggan-vik inscription features two seemingly non-lexical sequences as well as a pre po sitional phrase that Knirk and Schulte interpret differently Both also suggest that the (apparently) non-lexical sequences may be magical citing a study by the German runologist Klaus Duumlwel (1988 = 2011) which seeks to explain the seemingly nonsensical early runic letter sequences that appear so commonly on the Old Germanic bracteates in terms of classical forms of letter magic But what a magical alphabetic sequence would mean on a runic memorial is not fully explored by Knirk or Schulte

The name Kelbathornewaz is one of several early runic examples of dithe-matic names in -thornewaz much as the element -gastiz is particularly well attested among the oldest Nordic finds (Peterson 2004) The element -thornewaz is also found as an independent lexical item in early runic inscriptions and is often interpreted literally as lsquoservantrsquo (rather than Krausersquos lsquoliege manrsquo) The first element kelba- is evidently a full-grade variant of the Common Ger manic word for lsquocalfrsquo (cf Old English [hereafter OE] cilfer Old High Ger man [hereafter OHG] chilburra lsquochilver ewe lambrsquo lt kelbizjō-uzjō cf Schulte 2010 52 Knirk 2011 31) A heroic interpretation lsquocalf-thanersquo (ie a name for a young warrior) also makes much better sense than (mere ly) lsquocalf-servantrsquothinspmdashthinspa thornewaz (as someone who lsquoservedrsquo militarily) presumably represented one of the middle-ranking figures (or liege men) who featured in Iron Age armies (cf Mees 2003 59 f Pauli Jensen et al 2003) Naudigastiz is more transparently to be translated literally as lsquoneed-guestrsquothinspmdashthinspie a guest who is (or has been) in needthinspmdashthinspan onomastic indexing of the early Germanic (and indeed Indo-European) tradition of (military) hospitality (Wesseacuten 1927 44 f cf Watkins 1996 246 Schulte 2011 63)

The obviously lexical part of the Hogganvik text for which the proper inter pretation is disputed is the collocation in(n)ana nabōz literally lsquofrom within the naversquo Schulte (2010 56 2013 124) notes that Old Norse nǫf can also refer to the corner of a house (presumably because the junctures of timber were seen to be comparable to that of the nave of a wooden wheel) but in some later dialects (and place-names) the same form can also mean lsquoprotruding rock elevated headland promontory capersquo evidently a devel op ment of the semantic lsquocornerrsquo Bjor vand and Linde man (2007 786 f) argue that two different etymological roots are involved with the mean ing lsquocorner of housersquo semantically unrelated to lsquonaversquo At any rate compar able runic memorials do not feature the preposition in(n)ana nor do they clear ly feature references to headlands wheel naves or the corners of houses

12 bull Bernard Mees

Futhark 7 (2016)

The early runic memorial habit

It is quite clear however even from a superficial survey of the somewhat over forty early runestone inscriptions which have survived that textual formulas and subtypes can be isolated within the overall older funerary genre Memorial epigraphy is typically formulaicthinspmdashthinspfunerary inscriptions usually repre sent the clearest epigraphic testimony for what Weber (1922 12ndash16) characterised as traditional action (cf Mees 2013 327 f) Much like sets of Iron Age grave goods funerary epigraphs typically accord to standard typol ogies and feature recurrent formulas because they are solemn emo tional expressions that are public in character (cf Rosenwein 2006 61) Like Roman memorials they clearly reflect a form of ldquoepigraphic habitrdquo (MacMullen 1982) and often feature little more than indications of who the memorialised was and who set the monument up Since all forms of memo rial epigraphy are stereotypical the disagreement which has arisen in inter pretations of the Hogganvik inscription may perhaps be resolved by a fuller consideration of the fundamental formulaic characteristics of the older runic memorial habit

The Hogganvik inscription was found near the remains of an Iron Age grave yard or burial site (Gloslashrstad et al 2011) and clearly belongs to the early Nordic genre of runestone texts that characteristically feature a manrsquos name in the genitive and a labelling of the associated grave or memo rial stone The Boslash stonersquos hnabudas hlaiwa lsquoHnabudazrsquos graversquo Sten stadrsquos igijon halaz lsquoIngijorsquos stonersquo and the similar hariẉulfsmiddotstᴀinᴀz lsquoHari wulfzrsquos stonesrsquo on the transitionalyounger Raumlvsal memorial are parade examples of this labelling type (KJ 78 KJ 81 and KJ 80 respectively cf Anton sen 2002 191 and Schulte 2010 49 f 2013 121) In contrast early runic magico-religious inscriptions often feature item descriptions and naming expressions or names but these are never brought together syn tac tically in the (genitival) manner seen at Hogganvik (MacLeod and Mees 2004 2006 71ndash101) Early Nordic runestone texts sometimes also feature what are usually taken to represent locativising prepo sitional phrases (another feature absent from early runic magico-religious texts) but the proper interpretation of such expressions has often been debated The Moumljbro inscription features the form anahahai taken by Krause and also Antonsen as ana ha(n)hai lsquoon a steedrsquo (cf ON hestr lt hanhistaz lsquostallionrsquo)thinspmdashthinspie as a description which reflects the image of a horse and rider that also appears on the stone (KJ 99 Antonsen 1975 no 11) Yet hanhaz is not hanhistazthinspmdashthinspa form hanhaz is not paralleled in Germanic with a meaning lsquosteedrsquo Hence Staffan Fridell (2008 2009)

The Hogganvik Inscription and Early Nordic Memorialisation bull 13

Futhark 7 (2016)

has more recently suggested a toponymic interpretation lsquoat Haringrsquo with hahai instead representing the name of the place where the memorialised warrior died (although the apparent Sanskrit cognate śankuacute- lsquopeg spikersquo and Old Church Slavonic sǫkŭ lsquobranchrsquo suggest that Haring was a u-stem in Ger manic) The Rouml memorial features what seems to have been meant as another prepositional phrase beginning with the preposition ana but with the rest of the line largely unreadable today (KJ 73) Even less clear is the inscription on the Nordhuglo stone that instead finishes simply with the sequence ih which has often been supposed (since Bugge NIaeligR no 49) to represent an abbreviation for in Hugla lsquoin Huglorsquo (KJ 65) although this interpretation is rather speculative The Hogganvik text however reads in(n)ana lsquowithinrsquo not in lsquoinrsquo or ana lsquoat uponrsquo as would be expected in connection with a place-name Schultersquos connection of naboz with a geographical description of the location of the monument lsquowithin the area of the protruding rocksrsquo seems rather unlikely in this light

Indeed the preposition in(n)ana clearly represents a locativising ex pres-sion which etymologically means lsquofrom withinrsquo the suffix -na being direc-tional (cf Latin -nē in supernē lsquofrom aboversquo Goth utana lsquofrom withoutrsquo aftana lsquofrom behindrsquo etc Schmidt 1962 178ndash81 and 183 f) Surely the semantic which fits best with a directional lsquofrom withinrsquo is lsquocorner of a housersquothinspmdashthinspie a formation of a type which is widely paralleled in Old Norse cf ON innan borgar lsquowithin the townrsquo innan hallar lsquowithin the hallrsquo innan veggja lsquowithin the wallsrsquo and (especially) innandura innan gareths innan gaacutetta innanhuacutess and innanstokks all of which signify lsquoat home in doorsrsquo The Hogganvik expression inananaboz can be understood as meto nymic thenthinspmdashthinspie with the literal meaning lsquofrom within the corner of a housersquo indicating lsquoat home within the cornerswalls of his housersquo If the use albeit limited of prepositional phrases in other runestone memorials can be used as a guide then the description inananaboz looks as if it indicates that Kelba thornewaz died at home rather than (more heroically) out of doors No indication comparable to the Moumljbro stonersquos slaginaz lsquoslainrsquo is given but the Moumljbro inscription is unique among the older runic texts by featuring an explicit verbal indication of dying

Non-lexical sequencesThe seemingly non-lexical sequences on the Hogganvik stone are linked by both Knirk and Schulte with magical interpretations of comparable orderings which appear on early runic amulets Nordeacuten (1934 1940) inter-preted several of the early Nordic runestone inscriptions as funer ary curses

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

12 bull Bernard Mees

Futhark 7 (2016)

The early runic memorial habit

It is quite clear however even from a superficial survey of the somewhat over forty early runestone inscriptions which have survived that textual formulas and subtypes can be isolated within the overall older funerary genre Memorial epigraphy is typically formulaicthinspmdashthinspfunerary inscriptions usually repre sent the clearest epigraphic testimony for what Weber (1922 12ndash16) characterised as traditional action (cf Mees 2013 327 f) Much like sets of Iron Age grave goods funerary epigraphs typically accord to standard typol ogies and feature recurrent formulas because they are solemn emo tional expressions that are public in character (cf Rosenwein 2006 61) Like Roman memorials they clearly reflect a form of ldquoepigraphic habitrdquo (MacMullen 1982) and often feature little more than indications of who the memorialised was and who set the monument up Since all forms of memo rial epigraphy are stereotypical the disagreement which has arisen in inter pretations of the Hogganvik inscription may perhaps be resolved by a fuller consideration of the fundamental formulaic characteristics of the older runic memorial habit

The Hogganvik inscription was found near the remains of an Iron Age grave yard or burial site (Gloslashrstad et al 2011) and clearly belongs to the early Nordic genre of runestone texts that characteristically feature a manrsquos name in the genitive and a labelling of the associated grave or memo rial stone The Boslash stonersquos hnabudas hlaiwa lsquoHnabudazrsquos graversquo Sten stadrsquos igijon halaz lsquoIngijorsquos stonersquo and the similar hariẉulfsmiddotstᴀinᴀz lsquoHari wulfzrsquos stonesrsquo on the transitionalyounger Raumlvsal memorial are parade examples of this labelling type (KJ 78 KJ 81 and KJ 80 respectively cf Anton sen 2002 191 and Schulte 2010 49 f 2013 121) In contrast early runic magico-religious inscriptions often feature item descriptions and naming expressions or names but these are never brought together syn tac tically in the (genitival) manner seen at Hogganvik (MacLeod and Mees 2004 2006 71ndash101) Early Nordic runestone texts sometimes also feature what are usually taken to represent locativising prepo sitional phrases (another feature absent from early runic magico-religious texts) but the proper interpretation of such expressions has often been debated The Moumljbro inscription features the form anahahai taken by Krause and also Antonsen as ana ha(n)hai lsquoon a steedrsquo (cf ON hestr lt hanhistaz lsquostallionrsquo)thinspmdashthinspie as a description which reflects the image of a horse and rider that also appears on the stone (KJ 99 Antonsen 1975 no 11) Yet hanhaz is not hanhistazthinspmdashthinspa form hanhaz is not paralleled in Germanic with a meaning lsquosteedrsquo Hence Staffan Fridell (2008 2009)

The Hogganvik Inscription and Early Nordic Memorialisation bull 13

Futhark 7 (2016)

has more recently suggested a toponymic interpretation lsquoat Haringrsquo with hahai instead representing the name of the place where the memorialised warrior died (although the apparent Sanskrit cognate śankuacute- lsquopeg spikersquo and Old Church Slavonic sǫkŭ lsquobranchrsquo suggest that Haring was a u-stem in Ger manic) The Rouml memorial features what seems to have been meant as another prepositional phrase beginning with the preposition ana but with the rest of the line largely unreadable today (KJ 73) Even less clear is the inscription on the Nordhuglo stone that instead finishes simply with the sequence ih which has often been supposed (since Bugge NIaeligR no 49) to represent an abbreviation for in Hugla lsquoin Huglorsquo (KJ 65) although this interpretation is rather speculative The Hogganvik text however reads in(n)ana lsquowithinrsquo not in lsquoinrsquo or ana lsquoat uponrsquo as would be expected in connection with a place-name Schultersquos connection of naboz with a geographical description of the location of the monument lsquowithin the area of the protruding rocksrsquo seems rather unlikely in this light

Indeed the preposition in(n)ana clearly represents a locativising ex pres-sion which etymologically means lsquofrom withinrsquo the suffix -na being direc-tional (cf Latin -nē in supernē lsquofrom aboversquo Goth utana lsquofrom withoutrsquo aftana lsquofrom behindrsquo etc Schmidt 1962 178ndash81 and 183 f) Surely the semantic which fits best with a directional lsquofrom withinrsquo is lsquocorner of a housersquothinspmdashthinspie a formation of a type which is widely paralleled in Old Norse cf ON innan borgar lsquowithin the townrsquo innan hallar lsquowithin the hallrsquo innan veggja lsquowithin the wallsrsquo and (especially) innandura innan gareths innan gaacutetta innanhuacutess and innanstokks all of which signify lsquoat home in doorsrsquo The Hogganvik expression inananaboz can be understood as meto nymic thenthinspmdashthinspie with the literal meaning lsquofrom within the corner of a housersquo indicating lsquoat home within the cornerswalls of his housersquo If the use albeit limited of prepositional phrases in other runestone memorials can be used as a guide then the description inananaboz looks as if it indicates that Kelba thornewaz died at home rather than (more heroically) out of doors No indication comparable to the Moumljbro stonersquos slaginaz lsquoslainrsquo is given but the Moumljbro inscription is unique among the older runic texts by featuring an explicit verbal indication of dying

Non-lexical sequencesThe seemingly non-lexical sequences on the Hogganvik stone are linked by both Knirk and Schulte with magical interpretations of comparable orderings which appear on early runic amulets Nordeacuten (1934 1940) inter-preted several of the early Nordic runestone inscriptions as funer ary curses

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 13

Futhark 7 (2016)

has more recently suggested a toponymic interpretation lsquoat Haringrsquo with hahai instead representing the name of the place where the memorialised warrior died (although the apparent Sanskrit cognate śankuacute- lsquopeg spikersquo and Old Church Slavonic sǫkŭ lsquobranchrsquo suggest that Haring was a u-stem in Ger manic) The Rouml memorial features what seems to have been meant as another prepositional phrase beginning with the preposition ana but with the rest of the line largely unreadable today (KJ 73) Even less clear is the inscription on the Nordhuglo stone that instead finishes simply with the sequence ih which has often been supposed (since Bugge NIaeligR no 49) to represent an abbreviation for in Hugla lsquoin Huglorsquo (KJ 65) although this interpretation is rather speculative The Hogganvik text however reads in(n)ana lsquowithinrsquo not in lsquoinrsquo or ana lsquoat uponrsquo as would be expected in connection with a place-name Schultersquos connection of naboz with a geographical description of the location of the monument lsquowithin the area of the protruding rocksrsquo seems rather unlikely in this light

Indeed the preposition in(n)ana clearly represents a locativising ex pres-sion which etymologically means lsquofrom withinrsquo the suffix -na being direc-tional (cf Latin -nē in supernē lsquofrom aboversquo Goth utana lsquofrom withoutrsquo aftana lsquofrom behindrsquo etc Schmidt 1962 178ndash81 and 183 f) Surely the semantic which fits best with a directional lsquofrom withinrsquo is lsquocorner of a housersquothinspmdashthinspie a formation of a type which is widely paralleled in Old Norse cf ON innan borgar lsquowithin the townrsquo innan hallar lsquowithin the hallrsquo innan veggja lsquowithin the wallsrsquo and (especially) innandura innan gareths innan gaacutetta innanhuacutess and innanstokks all of which signify lsquoat home in doorsrsquo The Hogganvik expression inananaboz can be understood as meto nymic thenthinspmdashthinspie with the literal meaning lsquofrom within the corner of a housersquo indicating lsquoat home within the cornerswalls of his housersquo If the use albeit limited of prepositional phrases in other runestone memorials can be used as a guide then the description inananaboz looks as if it indicates that Kelba thornewaz died at home rather than (more heroically) out of doors No indication comparable to the Moumljbro stonersquos slaginaz lsquoslainrsquo is given but the Moumljbro inscription is unique among the older runic texts by featuring an explicit verbal indication of dying

Non-lexical sequencesThe seemingly non-lexical sequences on the Hogganvik stone are linked by both Knirk and Schulte with magical interpretations of comparable orderings which appear on early runic amulets Nordeacuten (1934 1940) inter-preted several of the early Nordic runestone inscriptions as funer ary curses

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

14 bull Bernard Mees

Futhark 7 (2016)

and many such interpretations duly feature in the corpus edition of older runic texts produced by Krause and Jankuhn Schulte (2013 122) simi larly cites Page (1987 30) who claims that some early rune stone inscrip tions in-clude magical features that seem to have been intended to ldquokeep the grave from desecration or the corpse in the graverdquo much as if the Hoggan vik sequences represent some sort of magical threat arraigned against thieves or haunting Page however was clearly referring to the appearance of the etymologically controversial magico-religious term alu as the sole form recorded on the Elgesem runestone and the palin drome sueus from the Kylver runestone which can both be paralleled by forms that appear in amuletic epigraphy (KJ 57 and KJ 1 respectively) Early runic alu is a relatively common term in bracteate texts and a similar palindrome to the Kyl ver form sueus is reflected in the Roman-letter sequence siususuis which appears on the medallion imitation from Kaumllder (IK 286)

Yet few of the texts of Nordeacutenrsquos grave-haunting type feature non-lexi-cal sequences and nothing directly comparable to the Hogganvik forms aaasrpkf and aarpaa is known from any of the bracteates The first expres-sion is somewhat reminiscent of the sequence authornrkf pre served on the Roskilde bracteate (Hauck and Heizmann 2003) which seems to re pre sent a scrambled form of futhornark with the a moved to where the equivalent char-acter A appears in the Roman abecedarium and the f-rune taking the place where the corresponding letter F comes in the Latin pedagogical ordering A more recent bracteate find from Stavnsager simi larly features a spelling aalul that includes comparable letter doubling but appears to re pre sent an irregular form of the common early runic charm word alu (Ax boe and Imer 2012) Rather than comparing with attested brac teate sequences however Schulte (2010 53thinspf 2013 123) cites the evidence of inscrip tions such as the transitional-runic Aumlllerstad memo rial which ends with a sequence kk︲kiiii︲kkk that is often taken to be magical (KJ 59) Triple repe tition (such as kkk) is often connected with magical and religious expres sions in what West (2007 106) characterises as ldquolitur gical-magicalrdquo iteration Such repetitions are also reflected in some younger runic texts where their context is more clearly magicalthinspmdashthinspeg the three i-runes of the eleventh-century amulet from Sigtuna which has been read as iii isiR thornis isiR lsquoiii ice these icersquo in the inscription (Eriksson and Zett er holm 1933 MacLeod and Mees 2006 118) But Krause (in KJ 59) instead inter preted the final Aumlllerstad sequence as a form of cryptic runes taking kk︲kiiii︲kkk to represent a coded form of the com mon early runic brac teate term alu His derivation however relies on the coding system of so-called is-runes developed for the sixteen-character younger futhark and hence seems quite implausibly anachronistic

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 15

Futhark 7 (2016)

There appear to be four possible interpretations of such expres-sionsthinspmdashthinspeither as abbreviations coded sequences linguistically meaning-less expressions or magical letter orderings Knirk and Schulte assume that the last possibility is the most likely at Hogganvik relying on the ap-proach to such sequences advocated by Duumlwel for the irregular spellings attested on the Migration Age bracteates But the difficult Hog gan vik expressions are only broadly paralleled on the bracteates and the Hoggan-vik inscription shows no other sign that it was intended to be magical In stead it features elements that are not reflected in other early runic magico-religious inscriptions but that are typical of memorial texts The characteristic features of early runic magical inscriptions are names (and longer naming expressions) item descriptions letter sequences (often futhark orderings) ldquocharmrdquo words (such as early runic alu) and magical sym bols (such as swastikas)thinspmdashthinspbut not possessive genitive anthroponyms and locativising prepositional phrases (MacLeod and Mees 2004 2006 71ndash101) The inscription on the early Nordic Kylver stone is generally thought to be magical as it features a futhark row and a tree-like symbol (as well as the palindrome sueus) while the appearance of alu as the sole term on the Elgesem stone is unambiguously cultic or magical The Kylver and Elge sem texts do not feature memorial formulas or vocabulary and are typologically unlike the Hogganvik memorial which is otherwise quite clearly to be associated formulaically with other non-magical runestone inscrip tions The Hogganvik memorial seems to represent a typical com-mem o rative expression supplemented by some non-lexical sequences that may or may not be magical and which from a Roman perspective seem unlikely to represent a magico-religious addition to the early runic memorial habit

Abbreviations and repetitionsThe most orthographically regular manner by which to explain an un pro-nounceable sequence of letterforms is as an abbreviation Standard epi-graphic formulas often appear abbreviated in Latin epigraphy sequences such as Dis Manibus lsquoto the spirits of the deadrsquo sit tibi terra levis lsquomay the earth lie lightly upon yoursquo hoc monumentum heredem non sequitur lsquothis tomb does not pass to the heirrsquo and sua pecunia lsquofrom his own moneyrsquo often appearing as DM STTL HMHNS or SP (Keppie 1991 138 f) Some early Nordic bracteate texts also appear to feature abbreviations such as r(ūnōz) lsquorunesrsquo and f(aihidō) lsquoI drew I decoratedrsquo which are not clearly motivated on magico-religious grounds (KJ 132 and KJ 134) and the

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

16 bull Bernard Mees

Futhark 7 (2016)

sequence ih on the Nordhuglo stone has long been thought to represent a similar abbreviation Imer (2011 22) argues that an abbreviation w(orahtō) or w(ritu) is to be understood on the Garingrdloumlsa fibula (KJ 12) and Schneider (1975 116) similarly suggested than an abbreviation h(aitē) was intended on the Thorsberg shield boss (KJ 21)

Some unpronounceable sequences of runes however can evidently be ascribed to coding rather than abbreviation The representation of the patently magical thornistil mistil kistil (or lsquothistle mistletoe casketrsquo) for-mula as thornmkiiissstttiiilll on the eleventh-century Ledberg stone is the most transparent instance of coding in Northern memorial epigraphy (Moltke 1985 171 MacLeod and Mees 2006 145 f) Similarly the reference to stᴀbᴀ thornria | fff in the transitional inscription on the Gummarp stone (KJ 95) is often taken to indicate a triple repetition of the rune name fehu lsquocattle wealthrsquo (MacLeod and Mees 2006 112) and authors such as Krause put considerable effort into promoting similarly ideographic or coded interpretations of obscure sequences in early runic texts Clear references to fehu however are otherwise unparalleled in early runic epigraphy whereas employments of faihjan lsquodraw decorate colourrsquo are quite common (with an abbreviated form possibly evidenced on the Femoslash brac teate) Hence an interpretation of the Gummarp inscription (which was destroyed in a fire in the eighteenth century and is only known today from illustrations) with more parallels is [afətr] Hathornuwoləfa sat(t)e staba thornri(j)a f(āhda) f(āhda) f(āhda) lsquoSet up in memory of Hathornuwoləfr three staves I drew I drew I drewrsquo Commonly used terms such as fecit lsquomadersquo are routinely abbreviated in Latin epigraphy and are represented in some of the makerrsquos marks which appear on weapons imported into the North during the later Roman Iron Age (Imer 2011) Indeed a double abbreviation FF is preserved in several Roman inscriptions as an abbreviation for fecerunt lsquo(they) madersquo (eg CIL 3 no 4197) and doubling and tripling is regularly used in Roman epigraphy to indicate plurality A memorial inscrip tion from Lyon for instance ends with the sequence PPP CCC SSS AAA DDD which is clearly a tripled abbreviation of the common Latin memo rial formula p(onendum) c(uraverunt) s(ub) a(scia) d(edicaverunt) lsquo(all three) caused (this) to be placed (and) dedicated while it was under constructionrsquo (CIL 13 no 2016) A similar threefold repetition of a form of the fabricatory verb faihjan (lsquothree staves they drewrsquo) would seemthinspmdashthinspfrom a Roman perspectivethinspmdashthinspto be the best paralleled inter pre-tation of the difficult Gummarp sequence

A comparably ldquoscepticalrdquo approach might accordingly be taken to the difficult Aumlllerstad sequence which also has the impression of an abbre vi-

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 17

Futhark 7 (2016)

ation about it although an alternative procedure involving parallels in the non-runic world may be to compare it with the numeric sequences that often appear in Roman textsthinspmdashthinspas indications of time weight distance price or the like (Gordon 1983 44ndash49) With its four i-runes the sequence looks much like it is numeric and Roman memorials sometimes feature temporal expressions such as vixit ann(os) vi m(ensibus) viii d(iebus) xxii lsquolived for six years eight months and twenty-two daysrsquo (Gordon 1983 no 99) Given the origin of the Roman numeral v in a representation of a hand the Aumlllerstad sequence kk︲kiiii︲kkk (where each of the k-runes takes the transitional form laquo) could be interpreted as a similar attempt to indicate three numbers 10 (years) 9 (months and) 15 (days)

Schulte however compares the Hogganvik sequences to the non-lexical form aaaaaaaazzznnnlowastbmuttt on the Lindholmen amulet (KJ 29) an expression which has often been connected with the medieval magical tradition of thornurs thornriacuteu lsquothree thorn-runesrsquo (or thurses) aacutesa aacutetta lsquoeight a-runesrsquo (or AEligsir) and nauethir niacuteu lsquonine n-runesrsquo (or needs) see MacLeod and Mees (2006 121 f) Yet precisely what an inscription on a piece of worked horn or bone might have to do with the early Nordic memorial habit is not made clear by Schulte Alliterating expressions such as thornurs thornriacuteu may instead be related to the younger runic practice of indicating the names of the Old Norse numerals in abbreviated forms ie as e(inn) t(veir) thorn(riacuter) f(joacuterir) f(imm) s(ex) s(jau) aacute(tta) n(iacuteu) t(iacuteu) etc (MacLeod and Mees 2006 147 f) In younger runic employment a thorn-rune (called thornurs) could in this case serve as an abbreviation for thornriacuteu lsquothreersquo an a-rune for aacutetta lsquoeightrsquo and an n-rune for niacuteu lsquoninersquo and a tradition where such alliterating numerical pairs were taken to be magical appears to have developed in medieval times But nothing similar to the younger numeric employment is clearly evidenced in older epigraphy and only three (or perhaps four) n-runes (ie rather than nine) are preserved in the Lindholmen inscription (which Krause nonetheless persisted in interpreting as a long sequence of ideographs) The irregular sequences on the Hogganvik stone show no obvious similarity with the repeated runes found on the Lindholmen amulet which can alternatively be taken as a sort of coded term or name (perhaps Az(i)n[a]mu(n)d(az) if not a reversed form of tumbnaz lsquoof the tiprsquo cf ON tumba lsquotumblersquo OHG zumba lsquostub tip penisrsquo) and appear in a context that clearly supports a magical interpretation (Groslashnvik 1996 70ndash73 MacLeod and Mees 2006 92) A more mundane explanation for the Hog gan vik sequences would seem to be much better paralleled in early runic memorial epigraphy

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

18 bull Bernard Mees

Futhark 7 (2016)

Memorial and magical inscriptions

A rather clearer comparison however may be with the similarly not-immediately interpretable expression that appears on the obverse of the Krog sta stone (KJ 100) It is taken by MacLeod and Mees (2006 110) to be magico-religious yet even this rather minimalist runestone text can be under stood as a member of a well-attested early runic sub-genre of memorial monuments Moreover Imer (2015 154) has dated the Krogsta monument (by means of the shape of its e-rune) to the late Roman Iron Age much like the similarly funerary Hogganvik stonethinspmdashthinspie to the period AD c 150160ndash375400

The most outstanding feature of the Krogsta stone is the picture of a man that it bears the representation having its arms extended in a manner which in both ancient and early Christian contexts is usually regarded as a representation of praying (Klauser 1959 1960) This orans or lsquoone who praysrsquo posture taken by the figure suggests that the person represented at Krog sta was considered very pious (perhaps even a pagan gudja or priest) The picture is also accompanied by two inscriptions one on the front of the stone and one on its back Only the text on the reverse of the stone seems immediately readable however and even it appears to be deficient in one striking manner

The readable Krogsta text is siumlainaz a sequence that has usually been taken as a mistake for the common early runic term stainaz lsquostonersquo with the (comparatively rare) iuml-rune apparently standing erroneously for what was meant to be a t-rune The other Krogsta expression seems quite uninterpretable lexically however reading mwsiumleijlowast a perhaps incomplete sequence It seems that a slightly substandard text on one side of the Krogsta stone has been complemented by an even less regular one on the other

Yet the most characteristic feature of the Krogsta inscription would seem to be its labelling of the associated stone This behaviour suggests (given the memorials which feature texts of the genitive name plus object descrip tion type) that the more obviously lexical Krogsta sequence should be interpreted as lsquo(this pious manrsquos) stonersquothinspmdashthinspie as a mixed pictorialorthographic reflection of the lsquoNNrsquos stonegraversquo formula typical of the early runic memorial habit The orans figure seems to serve as the pictorial equivalent of an anthroponym the labelling of the object representing a kind of deixis (or linguistic ldquopointingrdquo) which (equally) stands in contrast to the usual tendency in more literate cultures for the focus of such labelling to be on naming the owner of a find rather than the associated object

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 19

Futhark 7 (2016)

Typically however the difficult sequence which accompanies the orans representation at Krogsta is treated as completely opaque by scholars Yet the use of the iuml-rune where we would expect a runic t on the more immediately interpretable side suggests that a similar reading should be entertained for the sequence on the orans face of the stone This then as Seebold (1994 79) suggests could be taken as evidence that mwsiumleijlowast was supposed to be understood as featuring a sequence steijlowast a form which he links with an obliquely inflected variant of stainaz Nonetheless if it is to be interpreted in such a manner a spelling steijlowast would appear more likely to represent a development of Proto-Germanic stajanan (lt Indo-European steh2- sth2-eh1thinspithinsp

eothinsp-) lsquostand remainrsquo cf OHG stēn stān Old Saxon stān Latin stō stāre Oscan staiacutet stahiacutent Old Church Slavonic stojǫ stojati and the younger runic staeligndr staeliginn lsquostands the stonersquo formula (Sihler 1995 529 f Kaumlllstroumlm 2007 52) Given the early j-loss in stajanan (Thoacuter halls doacutettir 1993 35 f) steijlowast looks as if it may continue analogical vocalism (cf OHG stēn lt stai-) If so the inscription on the orans side of the Krogsta stone could possibly be read as mw staeligij[u] with mw representing an abbreviationthinspmdashthinspie a typical fabricant (NN statuit or lsquoNN set uprsquo) expression and not a magical sequence of a type not clearly paralleled elsewhere in runic epigraphy

The Krogsta expression mwsiumleijlowast may thus represent an irregularly expressed but typical mundane formation and hence like the sequence kk︲kiiii︲kkk on the Aumlllestad memorial may not be a good example of a magical sequence on an early runic memorial Yet some of the older and transitional runestone inscriptions are undoubtedly magico-religious in character the Elgesem monument being found in a burial mound much as the Kylver stone was found in a grave These texts however clearly accord to the magico-religious genre described by MacLeod and Mees (2004 2006 71ndash101) and not to those usually attested in early runic memorial epigraphy The date at which magico-religious textual features first found their way into the runic memorial habit is unclear As it is possible to interpret both the Krogsta and the Aumlllerstad inscriptions as wholly mundane albeit in part abbreviated or otherwise orthographically irregular memorial expressions a similar opacity might well be present in the Hogganvik inscription

What is clear about the Hogganvik sequences aaasrpkf and aarpaa however is that they are quite different than the Kylver stonersquos palindrome sueus or other kinds of magical letter sequences known from ancient magical sources such as the row of Greek vowels (αηιουω) found on a Jasper amulet of uncertain provenance now in the National Museum in

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

20 bull Bernard Mees

Futhark 7 (2016)

Copen hagen (Mackeprang 1940 94 and fig 16 p 96)1 Nonetheless some letter doubling is evident in the Hogganvik sequence aarpaa in a manner comparable to that which appears in the Stavnsager bracteatersquos aalul and the names uuigaz and ssigaduz on the Vaumlsby and Svarteborg amulets (KJ 128 and KJ 47 ie IK 241 and IK 181 respectively Wagner 2009) Cer-tain kinds of partially and completely non-lexical sequences such as abece-daria and palindromes were used in both classical and early Germanic magic but it is only the letter doubling in the Hogganvik sequences aaasrpkf and aarpaa that makes them seem potentially magical Taken from the perspective of Roman funerary epigraphy however the two sequences seem more likely to represent abbreviated forms than they do magical expressions

The Hogganvik sequencesAs the Roman use of letter repetition suggests the doubling and tripling of runic staves may not be clear evidence of a magical use of early Nordic writing The difficult Hogganvik forms are evidently similar to each other as each features a repetition of a-runes and the sequence rp and the Tune memorial (KJ 72) records a suitably funerary alliterating sequence arbija thinspthinsp arbijano lsquoinheritance thinspthinsp of the inheritorsrsquo (Mees 2015) Yet the sequence rp would be more difficult to account for than aa as an abbreviation as terms beginning with p are typically restricted to loanwords in Proto-Ger manic And while an early North Germanic text comparable to sua pecunia might well have featured a reference to an early form of Old Norse penningar lsquopennies moneyrsquo nothing similar is attested anywhere in runic epigraphy

Nonetheless the triple repetition in aaasrpkf is reminiscent of the ab-bre viation AAA for annorum lsquoyearsrsquo recently found on an amphora from Cologne (AE 2009 no 918b) and the latter part of the difficult Hoggan vik sequence is quite similar to the Roman memorial style d(e) s(ua) p(ecunia) c(uravit) f(aciendum) lsquopaid for it with her own moneyrsquo (CIL 1 no 1688) The other Hogganvik form aarpaa however is also reminiscent of OE earp lsquodark duskyrsquo a scribal or apophonic variant of OE eorp lsquoidemrsquo and ON jarpr lsquoswarthy brownrsquo cf OE Earpwald Eorpwald After all a similar o-grade form is continued by Greek ὀρϕνός lsquodark duskyrsquo and the Illerup shield mount features a semantically comparable anthroponym swarta lsquoBlack onersquo (Moltke 1985 95) Consequently the sequence aarpaa looks

1 My thanks to Peter Penz of the National Museum in Copenhagen for this reference

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 21

Futhark 7 (2016)

as if it may best be presumed to be another anthroponym the fourth to be recognised on the Hogganvik stone

Moreover erpaz lsquodark duskyrsquo has long been thought to represent the colour term from which the early North Germanic animal name erbaz lsquowolverinersquo (and hence Hogganvikrsquos erafaz) derives Their clear morphological relationship suggests that the two colour terms represent the detritus of an older Klugersquos law n-stem erbōn arbnaz (arpthinsppaz) the colour and animal descriptions having the same relationship as Greek ὀρϕνός lsquodark duskyrsquo has to ὄρϕος lsquodusky perchrsquo (Kluge 1884 Wood 1902 71 Krahe and Meid 1969 138 f Kroonen 2011 55ndash84 and 133ndash46) The form aarpaa does not feature an epenthetic vowel comparable to that seen in erafaz but this may merely be a sign that it is the fricative (ie not just the juncture with r) which has occasioned the epenthesis Thus the sequence aarpaa looks as if it may have been related to the cognomen erafaz

Letter doubling appears often enough in older runic texts that it seems to have been used occasionally as an emphasising strategy similar to ligaturing the employment of mirror runes and other comparable forms of orthographic highlightingthinspmdashthinsphence presumably the appearance of letter repetition not just in names but also in more remarkable expressions such as the Lindholmen sequence aaaaaaaazzznnnlowastbmuttt (MacLeod 2006) If not a contraction of a name like Earpwald Arpa however seems to represent a nickname based on the hair colour of the man who bore it the cognomen lsquowolverinersquo presumably (at least in part) being similarly inspired by the physical appearance of Naudigastiz Schulte (2013 122 f) suggests that the designation lsquowolverinersquo may also have indicated Naudigastizrsquos elite social standing as references to fur coats appear in later Germanic anthroponyms (such as ON Biarnheethinn and OHG Mardhetin) But the sequence aarpaa is immediately followed by inananaboz much as if Arpa (rather than Kelbathornewaz) is being described as in(n)ana nabōzthinspmdashthinspie Arpa was the name that Naudigastiz was known by lsquoat home indoorsrsquo Consequently the style erafaz lsquoWolverinersquo may have been Naudigastizrsquos public (and elite) cognomen aarpaa lsquoDark one Dusky onersquo his more colloquial nickname among members of his immediate household

If so the earlier sequence aaasrpkf may also represent another ortho graphically unconventional reference to Naudigastiz The final rune f could be understood as an abbreviated reference to Naudi gastizrsquos function as the inscriber or commissioner of the Hogganvik memorial with aaasrpkf an abbreviated reference to Arpa writing (f(aihidē)) the inscrip tion on Kelbathornewazrsquos memorial stone Indeed aaasrp could well

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

22 bull Bernard Mees

Futhark 7 (2016)

reflect a scrambled or irregular form of arpa as if the sequence were meant to be read partly in a retrograde manner ie as sa aarp k f sā Arp(a) K(elbathornewas) f(aihidē) lsquothe one Arpa son of Kelbathornewaz drewrsquo cf sā thornat bəriutithorn lsquothe one that breaks (this)rsquo in the curse on the Stentoften stone (KJ 96) Yet the medial letters of the most difficult Hoggan vik sequence are not clearly to be associated with an early-runic memorial style commonly attested elsewhere so it may simply be better to leave aaasrpkf uninterpreted (ie as an unexplained sequence) and not to draw any conclusions from the unexpected form at all

In contrast to Roman memorials however a more obvious characteristic of the Hogganvik inscription is that it gives considerably more space to describing the identity of the person who raised the stone than it does the memorialised Some of the early Nordic memorial inscriptions only mention the name of the dead but there are others such as the Nordhuglo text which only seem to mention the name of the commissioner of the memorial Still others spend rather more time describing who the manu-factures or (in the case of the Tune memorial) who the inheritors of the memorialisedrsquos estate were This unexpected feature suggests that a key function of some of the Iron Age monuments was to commemorate the act of raising the memorial as much as it was to celebrate the memory of the deceased Ogam stones in contrast never feature indications of who raised them or who wrote (or benefited from) their early Irish inscriptions and Roman funerary epigraphs although often mentioning the name of the commemorator similarly focus mainly on the name and situation (titles age at death etc) of the memorialised (Keppie 1991 106 f Saller and Shaw 1984 McManus 1997 51)

Yet MacMullen (1982) talks of the Imperial Roman epigraphic habit in terms of its ldquosense of audiencerdquo and Meyer (1996) explains the rapid growth in monumental epigraphy in the Roman provinces (the production of which is often thought to have peaked in the late second century) as due to a desire of the new provincial elites of the Empire to demon-strate their Latinity Speidel (2015 335ndash37) similarly notes the habit of self-representation common in the epigraphs associated with Roman veterans and soldiers particularly in texts which demonstrate the social standing of the commissioners of an inscription It sometimes seems to have been the supplying of the name of the erector of the memorial that was considered most important in older runic tradition as if the practice of raising runestones (rather than testifying to the deeds and social standing of the dead) was seen to be the most significant aspect of early Nordic commemoration The main emotional display represented by the

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 23

Futhark 7 (2016)

Hoggan vik monument seems to have been focussed on the (first-person) com memorator not the (genitival) memorialisedthinspmdashthinspand this inversion of textual focus so different from what typically applies in Roman funerary epigraphy appears to represent evidence for a remarkably self-predicative or agentive aspect to early Scandinavian commemoration

ConclusionTaken from a textual and pragmatic perspective the inscriptions that appear on early Nordic funerary runestones seem quite different in many ways to those from the Viking and later periods where a stronger sense of epigraphic habit had evidently led to the development of a greater level of textual standardisation The formulaic elements that appear in older runic memorials are often very simple possessive and labelling expressions of a kind that suggests deictic oral language of a form reasonably to be expected in an only marginally literate culture Other aspects of the early runic memorial texts typically represent rhetorical expansions whether adding comments to the obvious memorialising context naming the erector of the monument or adding some sort of elaboration to the name of the commissioner or the memorialised Minimalistic from an emotional perspective such extensions can also evidently include expressing part of the text in manners which are difficult to interpret today

In the Hogganvik inscription these extensions include a reference to in(n)ana nabōz a prepositional phrase that seems best to be translated as lsquoat home indoorsrsquo The phrase is immediately preceded by an orthographically unex pected form aarpaa that looks much like the expected o-grade equivalent of the Old English onomastic theme Eorp- Earp- lsquodark duskyrsquo A third even less regular sequence is also represented on the stone but whether it represents an abbreviated coded or even magical sequence remains unclear No clearly magical sequences are found on other early runic memorials however suggesting that an abbreviated or otherwise irregular text was intended by the carver

Where younger memorial texts are often so standardised as to be predictable the older runic memorial inscriptions are often much more interesting and varied Many of the comments and styles found on early runic memorials are repeated often enough that their general typological character can be reconstructed Difficult sequences found in early runic texts are also often thought to be magical But the recent focus on epigraphic habits first articulated by the classicist MacMullen and the history of emotions approach promoted especially by the medievalist

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

24 bull Bernard Mees

Futhark 7 (2016)

Rosen wein asks runologists to look at early memorials primarily in terms of audience genre and emotion and in this way bring runic studies more clearly into accord with the mainstream of contemporary epigraphic historiography

BibliographyAE = LrsquoAnneacutee eacutepigraphique Paris Presses Universitaires de France 1888ndashAntonsen Elmer H 1975 A Concise Grammar of the Older Runic Inscriptions

Sprach strukturen ser A Historische Sprachstrukturen 3 Tuumlbingen Niemeyer―thinsp 2002 Runes and Germanic Linguistics Trends in Linguistics Studies and

Mono graphs 140 Berlin De GruyterAxboe Morten and Lisbeth M Imer 2012 ldquoNye guldbrakteater med runerrdquo

httpruner-moenternatmusdknye-guldbrateater-med-runer dated 4 JulyBaeligksted Anders 1952 Maringlruner og troldruner Runemagiske studier National-

museets skrifter Arkaeligologisk-Historisk Raeligkke 4 Copenhagen GyldendalBjorvand Harald and Frederik Otto Lindeman 2007 Varingre Arveord Etymologisk

ordbok 2nd ed Oslo NovusBloch Marc 1954 [1949] The Historianrsquos Craft Trans Peter Putnam Manchester

Manchester University PressBraun Maximillian William M Calder III and Dietrich Ehlers eds 1995 ldquoLieber

Prinzrdquo Der Briefwechsel zwischen Hermann Diels und Ulrich von Wilamowitz-Moellen dorff (1869ndash1921) Hildesheim Weidmann

Carr Edward H 1964 What is History Harmondsworth PenguinCIL = Corpus Inscriptionum Latinarum Ed Theodor Mommsen et alAcademia

litterarum regiae Borussica (and successor bodies) 17 vols Berlin ReimerDe Gruyter 1863ndash

Collingwood Robin G 1946 The Idea of History Oxford ClarendonDilthey Wilhelm 1883 Einleitung in die Geisteswissenschaften Versuch einer

Grund legung fuumlr das Studium der Gesellschaft und der Geschichte Leipzig Deucker (English translation Introduction to the Human Sciences An Attempt to Lay a Foundation for the Study of Society Trans Ramon J Betanzos Detroit Wayne State University Press 1988)

Duumlwel Klaus 1981 ldquoThe Meldorf Fibula and the Origin of Germanic Writingrdquo Michi gan Germanic Studies 7 1 8ndash14 [with discussion on pp 15ndash18]

―thinsp 1988 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo Fruumlhmittelalterliche Studien 22 70ndash110

―thinsp 2011 ldquoBuchstabenmagie und Alphabetzauber Zu den Inschriften der Gold-brakteaten und ihrer Funktion als Amuletterdquo In Die Goldbrakteaten der Voumllker-wanderungs zeitthinspmdashthinspAuswertung und Neufunde ed Wilhelm Heizmann and Mor-ten Axboe 475ndash524 Ergaumlnzungsbaumlnde zum Reallexikon der Germa nischen Alter tums kunde 40 Berlin De Gruyter

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 25

Futhark 7 (2016)

Eriksson Manne and Delmar O Zetterholm 1933 ldquoEn amulet fraringn Sigtuna Ett tolk ningsfoumlrsoumlkrdquo Fornvaumlnnen 28 129ndash56

Fridell Staffan 2008 ldquoHaring Haringtuna och anahahairdquo Namn och bygd 96 47ndash59 ―thinsp 2009 ldquoHaring Haringtuna and the interpretation of Moumljbro anahahairdquo NOWELE

North-Western European Language Evolution 5657 89ndash106Gloslashrstad Zanette Tsigaridas Jakob Johansson and Frans-Arne Stylegar 2011

ldquoMinne lund og monument Runesteinen paring Hogganvik Mandal Vest-Agderrdquo Viking 74 9ndash24

Gordon Arthur E 1983 Illustrated Introduction to Latin Epigraphy Berkeley Univer sity of California Press

Groslashnvik Ottar 1996 Fra Vimose til Oslashdemotland Nye studier over runeinskrifter fra foslashr kristen tid i Norden Oslo Universitetsforlaget

Hauck Karl and Wilhelm Heizmann 2003 ldquoDer Neufund des Runen-Brakteaten IK 585 Sankt Ibs Vej-C Roskilde (Zur Ikonologie der Goldbrakteaten LXII)rdquo In RunicathinspmdashthinspGermanicathinspmdashthinspMediaevalia ed Wilhelm Heizmann and Astrid van Nahl 243ndash64 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Alter-tums kunde 37 Berlin De Gruyter

IK + no = bracteate or its inscriptions published in Die Goldbrakteaten der Voumllker-wanderungs zeit vols 12ndash32 Ikonographischer Katalog ed Karl Hauck et al 3 vols Muumlnster Mittelalter-Schriften 2412ndash2432 (Muumlnster Fink 1985ndash89)

Imer Lisbeth 2011 ldquoMaturus fecitthinspmdashthinspUnwod Made Runic Inscriptions on Fibulae in the Late Roman Iron Agerdquo Lund Archaeological Review 2011 11ndash27

―thinsp 2015 ldquoJernalderens runeindskrifter i NordenthinspmdashthinspKatalogrdquo Aarboslashger for nordisk oldkyndighed og historie 2014 1ndash347

Kaumlllstroumlm Magnus 2007 ldquoThe Rune-stone Fragment from Finsta in Skederidthinspmdashthinspthe Oldest Rune-stone with Long-branch Runes in the Maumllar Valleyrdquo In Cultural Inter action between East and West Archaeology Artefacts and Human Contacts in Northern Europe ed Ulf Fransson Marie Svedin Sophie Bergerbrant and Fedir Androshchuk 50ndash55 Stockholm Studies in Archaeology 44 Stockholm Stock holm University

Keppie Lawrence 1991 Understanding Roman Inscriptions Baltimore Johns Hop kins University Press

KJ + no = inscription published in Krause and Jankuhn 1966Klauser Theodor 1959 ldquoStudien zur Entstehungsgeschichte der christlichen

Kunst IIrdquo Jahrbuch fuumlr Antike und Christentum 2 115ndash45―thinsp 1960 ldquoStudien zur Entstehungsgeschichte der christlichen Kunst IIIrdquo Jahr-

buch fuumlr Antike und Christentum 3 112ndash33Kluge Friedrich 1884 ldquoDie germanische consonantendehnungrdquo Beitraumlge zur

Geschichte der deutschen Sprache und Literatur 9 149ndash86Knirk James E 2011 ldquoHogganvik-innskriften En hard runologisk noslashttrdquo Viking

74 25ndash39Krahe Hans and Wolfgang Meid 1969 Germanische Sprachwissenschaft vol 3

Wort bildungslehre Sammlung Goumlschen 1218b Berlin De Gruyter

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

26 bull Bernard Mees

Futhark 7 (2016)

Krause Wolfgang and Herbert Jankuhn 1966 Die Runeninschriften in aumllteren Futhark Abhandlungen der Akademie der Wissenschaften in Goumlttingen Philo-logisch-Historische Klasse 3rd ser 66 Goumlttingen Vandenhoeck amp Ruprecht

Kroonen Guus 2011 The Proto-Germanic N-stems A Study in Diachronic Morpho-phonology Leiden Studies in Indo-European 18 Amsterdam Rodopi

Mackeprang Mogens B 1940 ldquoAarslev-fundet Et rigt fynsk Gravudstyr fra 4 Aarh e Krrdquo Fra Nationalmuseets Arbejdsmark 1940 87ndash96

MacLeod Mindy 2006 ldquoLigatures in Early Runic and Roman Inscriptionsrdquo In Runes and Their Secrets Studies in Runology ed Marie Stoklund Michael Lerche Niel sen Bente Holmberg and Gillian Fellows-Jensen 183ndash200 Copenhagen Museum Tusculanum

MacLeod Mindy and Bernard Mees 2004 ldquoOn the t-like Symbols Rune-rows and Other Amuletic Features of the Early Runic Inscriptionsrdquo Interdisciplinary Journal for Germanic Linguistics and Semiotic Analysis 9 249ndash99

―thinsp 2006 Runic Amulets and Magic Objects Woodbridge BoydellMacMullen Ramsay 1982 ldquoThe Epigraphic Habit in the Roman Empirerdquo

American Journal of Philology 103 233ndash46 McManus Damien 1997 A Guide to Ogam Maynooth Monographs 4 Maynooth

An SagartMees Bernard 2003 ldquoRunic erilaRrdquo NOWELE North-Western European Language

Evolution 42 41ndash68―thinsp 2013 ldquolsquoGivingrsquo and lsquoMakingrsquo in Early Runic Epigraphyrdquo Transactions of the

Philological Society 111 326ndash53―thinsp 2015 ldquoFurther Thoughts on the Tune Memorialrdquo Norsk lingvistisk tidsskrift

33 49ndash62Meyer Elizabeth A 1990 ldquoExplaining the Epigraphic Habit in the Roman Empire

The Evidence of Epitaphsrdquo Journal of Roman Studies 80 74ndash96Moltke Erik 1985 Runes and Their Origin Denmark and Elsewhere Rev ed

Trans Peter G Foote Copenhagen National Museum of Denmark NIaeligR = Norges Indskrifter med de aeligldre Runer By Sophus Bugge and Magnus

Olsen 3 vols Norges Indskrifter indtil Reformationen 1st division Christiania Broslashggers 1891ndash1924

Nielsen Karl Martin 1985 ldquoRunen und Magie Ein forschungsgeschichtlicher Uumlber blickrdquo Fruumlhmittelalterliche Studien 19 75ndash97

Nordeacuten Arthur 1934 ldquoFraringn Kivik till Eggjum II Runristningar med gen garingngar-besvaumlrjelserdquo Fornvaumlnnen 29 97ndash117

―thinsp 1940 ldquoTysk runforskning under de sista aringrenrdquo Fornvaumlnnen 35 318ndash32 Page Raymond Ian 1964 ldquoAnglo-Saxon Runes and Magicrdquo Journal of the British

Archaeo logical Association 3rd ser 27 14ndash31―thinsp 1967 Review of Peter Berghahn and Karl Schneider Anglo-friesische Runen-

solidi im Lichte des Neufundes von Schweindorf (Ostfriesland) (Cologne 1967) The Numismatic Chronicle 7 304ndash06

―thinsp 1973 An Introduction to English Runes London Methuen

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

The Hogganvik Inscription and Early Nordic Memorialisation bull 27

Futhark 7 (2016)

―thinsp 1987 Runes Reading the Past London British Museum Press―thinsp 1995 Runes and Runic Inscriptions Collected Essays on Anglo-Saxon and

Viking Runes Ed David Parsons Woodbridge BoydellPauli Jensen Xenia Lars Joslashrgensen and Ulla Lund Hansen 2003 ldquoThe Germanic

ArmythinspmdashthinspWarriors Soldiers and Officersrdquo In The Spoils of Victory The North in the Shadow of the Roman Empire ed Lars Joslashrgensen Birger Storgaard and Lone Gebauer Thomsen 310ndash28 Copenhagen National Museum

Peterson Lena 2004 ldquoReflec tions on the Inscription laguthornewa on Shield-Handle Mount 3 from Illeruprdquo In Namenwelten Orts- und Personennamen in historischer Sicht ed Astrid van Nahl Lennart Elmevik and Stefan Brink 659ndash77 Ergaumlnzungsbaumlnde zum Reallexikon der Germanischen Altertumskunde 44 Berlin De Gruyter

Plamper Jan 2010 ldquoThe History of Emotions An Interview with William Reddy Barbara Rosenwein and Peter Stearnsrdquo History and Theory 49 237ndash65

Rosenwein Barbara H 2006 Emotional Communities in the Early Middle Ages Ithaca Cornell University Press

Saller Richard P and Brent D Shaw 1984 ldquoTombstones and Roman Family Rela-tions in the Principate Civilians Soldiers and Slavesrdquo Journal of Roman Studies 74 124ndash56

Schmidt Gernot 1962 ldquoStudien zum germanischen Adverbrdquo Dissertation Free Uni versity of Berlin

Schneider Karl 1956 Die germanischen Runennamen Versuch einer Gesamt deu-tung Ein Beitrag zur idggerm Kultur- und Religionsgeschichte Meisenheim Hain

―thinsp 1975 Review of Heinz Klingenberg Runenschrift Schriftdenken Runen-inschriften (Heidelberg Winter 1973) Beitraumlge zur Namenforschung ns 10 110ndash17

Schulte Michael 2010 ldquoRunene paring Hogganvik-steinen ved Mandal En runologisk og lingvistisk kommentarrdquo Agder Vitenskapsakademi Aringrbok 2010 48ndash65

―thinsp 2011 ldquoDie sprachliche Deutung der Hogganvik-Inschrift Ergaumlnzungen zum vorlaumlufigen Berichtrdquo Amsterdamer Beitraumlge zur aumllteren Germanistik 67 57ndash68

―thinsp 2013 ldquoThe Norwegian Hogganvik Stone as an Emblem of Social Status and Identityrdquo Journal of the North Atlantic special vol 4 120ndash28

Seebold Elmar 1994 ldquoDie sprachliche Deutung und Einordnung der archaischen Runeninschriftenrdquo In Runische Schriftkultur in kontinental-skandinavischer und -angelsaumlchsischer Wechselbeziehung Internationales Symposium in der Werner-Reimers-Stiftung vom 24ndash27 Juni 1992 in Bad Homburg ed Klaus Duumlwel 56ndash94 Ergaumlnzungs baumlnde zum Reallexikon der Germanischen Altertumskunde 10 Berlin De Gruyter

Sihler Andrew 1995 New Comparative Grammar of Greek and Latin New York Oxford University Press

Speidel Michael A 2015 ldquoThe Roman Armyrdquo The Oxford Handbook of Roman Epigraphy ed Christer Bruun and Jonathan Edmondson 319ndash44 Oxford Oxford University Press

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer

28 bull Bernard Mees

Futhark 7 (2016)

Thoacuterhallsdoacutettir Gudruacuten 1993 ldquoThe Development of Intervocalic j in Proto-Ger-manicrdquo Dissertation Cornell University

Wagner Norbert 2009 ldquoZum ssigaduʀ des Svarteborg-Medaillonsrdquo Beitraumlge zur Namen forschung ns 44 209ndash11

Watkins Calvert 1995 How to Kill a Dragon Aspects of Indo-European Poetics New York Oxford University Press

Weber Max 1922 Wirtschaft und Gesellschaft Vol 3 of Grundriss der Sozial-oumlkonomik Tuumlbingen Mohr

Wesseacuten Elias 1927 Nordiska namnstudier Uppsala universitets aringrsskrift 1927 Filo sofi spraringkvetenskap och historiska vetenskaper 3 Uppsala A-B Lunde-quistska bokhandeln

West Martin L 2007 Indo-European Poetry and Myth Oxford Oxford University Press

Wood Francis A 1902 Color-names and Their Congeners A Semasiological Investi-gation Halle a S Niemeyer