5 Drazen Marsic

8/10/2019 5 Drazen Marsic

1/36

111

e subject of this paper are three fragmentary portraitstelae from the Trogir City Museum. e fragment of amonumental stela with portraits of four deceased personsstands out in terms of quality of work and representation.It was discovered in Salona at the very beginning of the19thcentury and once was a part of the I.L. Garagnin col-lection. A fragment of the same monument still exists inthe garden of the Archaeological Museum in Split, whichis confirmed by identical fracture line and representation.Based on the remains of the inscription present on the

fragment from Split, the author establishes that the lowerpart of the same monument, which is now lost, was alsokept in the Garagnin collection and present at the famoussketch of that collection drawn by the painter Ivan Daniloin 1805. e inscriptionis written in sepulchral poetry andit mentions four members of the family of freed slaves At-tius and a girl which was their freed slave. e other twomonuments were discovered in Trogir or immediate vi-cinity of the town. e lower part of a small portrait stelacontains fully preserved, interesting inscription dedicatedto the old woman Marcella, while her portrait is chipped

from the neck upwards. Typological determination of thethird monument is questionable, but it is most likely a

fragment of another monumental stela. ick stone plate,

chipped from all sides represents a man in togated pos-ture. Several individual sections contain discussion on thecircumstances of the find, their state of preservation, ty-

Predmet su rada tri fragmentarna nadgrobna spomeni-ka s portretima pokojnika pohranjena u Muzeju gradaTrogira. Reprezentativnou i kvalitetom izradbe izdvajase fragment monumentalne stele s portretima et vero po-kojnika, neko uvan u zbirci I.L. Garagnina, pronaen uSaloni na samom poetku 19. st. Komad istoga spomenikauva se danas i u vrtu Arheolokog muzeja u Splitu, topotvruje kut loma i identian prikaz. Na temelju osta-tka natpisa na splitskom fragmentu autor zakljuuje daje donji dio istoga spomenika s natpisom, neko takoer

uvan u Garagninovoj zbirci, a danas izgubljen, prikazanna uvenom nacrtu slikara Ivana Danila iz 1805. godine.Natpis je pisan sepulkralnom poezijom, a spominje etve-rolanu osloboeniku obitelj Attius i njihovu osloboe-nicu. Druga dva spomenika potjeu iz samoga Trogira ilinjegove najblie okolice. Na donjoj polovici manje portre-tne stele u cijelosti je sauvan zanimljiv natpis posveenstarici Marceli, dok je njezin portret odbijen od vrata pre-ma gore. Tipoloka je pripadnost treeg spomenika upitna,ali je po svoj prilici i on ostatak monumentalne nadgrobnestele. Na debljoj ploi odlomljenoj sa svih strana prikazanje lik togata. U posebnim se poglavljima raspravlja o okol-nostima nalaza, stanju uuvanosti, tipologiji, portreti-

ma, natpisima i kronologiji triju spomenika.Kljune rijei: Muzej g rada Trogira, nadgrobne stele , por-treti, natpisi, tipologija, kronologija.

Draen MARI

RIMSKE PORTRETNE STELE IZ MUZEJA GRADA TROGIRA

ROMAN PORTRAIT STELAE FROM THE TROGIR CITYMUSEUM

UDK 904(497.5-37 Split):7373.041.5(37)069(497.5 Trogir).51:73Primljeno/Received: 31. 09. 2004.Prihvaeno/Accepted: 15. 10. 2004.

Draen MariSveuilite u Zadru

Obala kralja P. Kreimira IV. br. 2HR 23000 Zadar


2/36

112

Draen MARIRIMSKE PORTRETNE STELE IZ MUZEJA GRADA TROGIRA

Uvod

Muzej grada Trogira jedan je od najmlaih gradskihmuzeja na tlu Hrvatske. Osnovan je godine 1963,slubeno otvoren 1966, a smjeten je u palai Gara-gnin-Fanfogna. Iako se po svom karakteru i koliini

spomenike grae ne moe mjeriti s mnogo stari-jim bratskim muzejima u Dalmaciji, u njegovu selapidariju, otvorenom godine 1987, uvaju mnogi

vrijedni spomenici iz antikoga razdoblja od kojihsu neki gotovo nepoznati iroj znanstvenoj javnosti.Tri takva spomenika predmet su ovoga rada. Sva trinadgrobnoga su karaktera s portretima pokojnikakao jedinom dekoracijom. Dva se spomenika tipo-loki odreuju kao nadgrobne stele s portretima po-kojnika ili, kako se to udomailo u strunoj literatu-ri, kao portretne stele, dok je tipoloka pripadnost

treeg spomenika upitna, ali je po svoj prilici i onostatak monumentalne nadgrobne stele.

Iako su prilino fragmentarne, portretne stele izMuzeja grada Trogira istinski su dragocjene za poz-navanje sepulkralne umjetnosti rimske provincijeDalmacije. Prva stela pripada tipu koji je posvje-doen relativno velikim brojem primjera, ali je saspekta ikonografije i rasporeda figura pokojnika

jedinstvena za cijelo priobalno podruje provincijeDalmacije. Druga pak stela pripada samom koncupoganskoga razdoblja te svojom ikonografijom iepigrafikom ve navjeuje razdoblje kasne antike.

Moe se rei da dvije stele predstavljaju dvije stepe-nice razvoja i produkcije portretnih stela: onu naj-raniju iz razdoblja ranoga Carstva i onu posljednjuiz razdoblja kasnoga Carstva ili Dominata. Treispomenik zbog stupnja sauvanosti nema jednaku

vanost, ali i on se jasno uklapa u dosad steenespoznaje o ovome tipu nadgrobnog spomenika nairemu salonitanskom arealu.

U sljedeih e se nekoliko poglavlja razmatrati svavana pitanja o okolnostima nalaza, uuvanosti,tipologiji, portretima, natpisima i kronologiji triju

spomenika. Radi preglednosti, ali i injenice da se unavedenim poglavljima uvijek pojavljuju pod istimbrojem, najprije se donose osnovni kataloki podaciza svaki od njih.

1. Monumentalna portretna stela obitelji Attius(rekonstrukcija na slici 7)

a. Fragment u Muzeju grada Trogira (Sl. 1):

Mjesto nalaza: Solin 1805. (?) Inventarni broj:3. Dimenzije: visina 74 cm, irina 52 cm, de-bljina nije poznata. Literatura: Babi 1984:149, pl. XVIII, spomenik i slika br. IV (crte

I. Danila). Isti rad objavljen je i na hrvatskomjeziku: Babi 1985: 77, sl. 5, spomenik i slikabr. IV.

pology, portraits, inscriptions and the chronology of thesethree artifacts.

Key words: Trogir City Museum, grave stelae, portraits,inscriptions, typology, chronology.

Introduction

e Trogir City Museum is one of the youngest citymuseums in Croatia. Located in the Garagnin-Fan-fogna Palace, it was established in 1963 and formallyopened in 1966. Although its very character and thequantity of artifacts contained therein cannot meas-ure up to the much older fraternal museums in Dal-matia, its lapidarium, which was opened in 1987, pre-serves many valuable monuments from Classical An-tiquity, some of them almost completely unknown tothe broader scholarly public. ree such monuments

are the subject of this work. All three have a grave-marker character, bearing portraits of the deceasedas the only decoration. Two of the artifacts have beentypologically specified as grave stelae bearing por-traits of the deceased or, using the jargon of schol-arly literature, as portrait stelae, while there is somedoubt as to the typology of the third artifact, but it ismost likely the remains of a monumental grave stela.

Although considerably fragmentary, the portraitstelae from the Trogir City Museum are truly inval-uable to an understanding of sepulchral art in the

Roman province of Dalmatia. e first stela belongsto a type of which there are relatively many exam-ples, although in terms of iconography and arrange-ment of the figure of the deceased it is unique in theentire coastal belt of the province of Dalmatia. esecond stela belongs to the very close of the paganera and its iconography and epigraphy already her-ald the period of Late Antiquity. One can say thatthese two stelae represent two steps in the develop-ment and production of portrait stelae: the earliestfrom the Early Roman Empire and the last from theLater Roman Empire. e third artifact, due to its

physical condition, does not have the same impor-tance, but it clearly corresponds to previously-ac-quired knowledge on this type of grave monumentin the wider Salona environs.

e next several sections shall entail an examinationof all vital matters pertaining to the circumstancesof the find, their state of preservation, typology, por-traits, inscriptions and the chronology of these threeartifacts. In the interests of brevity, but also due tothe fact that they always appear under the samenumber in the sections that follow, the basic cata-

logue data for each of them are provided hereunder.1. Monumental portrait stela of the Attius family

(reconstruction on fig. 7)


3/36

113

Draen MARIROMAN PORTRAIT STELAE FROM THE TROGIR CITY MUSEUM

b. Fragment u Arheolokome muzeju u Splitu(Sl. 3)

Mjesto nalaza: Solin 1884. Inventarni broj: D116. Dimenzije: visina 63 cm, irina 38 cm,debljina 28 cm. Literatura: Rinaldi Tufi 1971:109, br. 34 (bez slike!).

c. Donja polovica stele s gotovo cijelim natpi-snim poljem (Sl. 4)

Mjesto nalaza: Solin 1805. (?) Neko u zbirciI. L. Garagnina, danas izgubljena ili nepozna-te destinacije. Pretpostavljene dimenzije natemelju crtea I. Danila s pridodanim mjeri-lom: visina 107 cm, irina 86,3 cm, debljinanije poznata. Literatura: Pavlovi-Lui 1811:21 i d. (natpis); CIL 3: 2197 (isto); Babi 1984:149, pl. XVIII, spomenik i slika br. XXVI (cr-te I. Danila); Babi 1985: 77, sl. 5, spomenik

i slika br. XXVI (isto).2. Donja polovica nadgrobne stele s portretom sta-

rice Marcele (Sl. 8)

Mjesto nalaza: Trogir ili njegova okolica. (?) In-ventarni broj: A 22 (679). Dimenzije: visina 38cm, irina 28 cm, debljina 7,5 cm. Neobjavljena.

3. Fragment nadgrobne stele (?) s portretom mu-karca (Sl. 9)

Mjesto nalaza: Trogir ili njegova okolica. (?) In-ventarni broj: 667. Dimenzije: visina 72 cm, iri-na 44 cm, debljina 27 cm. Neobjavljen.

1. Okolnosti nalaza

U naslovu rada kao matina ustanova triju spomeni-ka naveden je Muzej grada Trogira, to samo djelo-mice odgovara istini, i to kada su u pitanju spomenicipod rednim brojem 2i 3. Meutim, kada je u pitanjuspomenik br. 1, prava je istina da se njegovi ostaci u-vaju na dvije lokacije. Vei fragment (1.a) uzidan jeu Muzeju grada Trogira (!), a manji je odloen u vrtuArheolokog muzeja u Splitu (1.b). Njima treba pri-dodati i trei fragment, koji se neko nalazio u zbirciGaragnin-Fanfogna u Trogiru, ali je danas, naalost,izgubljen (1.c). injenica kako spomenuti fragmen-ti predstavljaju dijelove jednoga te istog spomenikadosad nije bila uoena, a rezultat je njegove saloni-tanske provenijencije i spleta nesretnih povijesnihokolnosti koji su te dijelove rasprili na razliite stra-ne. Kako intencija ovog rada nije objava samo toga,monumentalnou i kvalitetom izradbe zasigurnonajreprezentativnijega primjerka, ve i ostalih dva-ju, morali smo iz naslova izbaciti jednu od ustanova

i istaknuti samo onu u kojoj se uva veina grae.1.aVei komad monumentalne portretne stele kojise danas uva u Muzeju grada Trogira (sl. 1) prona-

a. Fragment in the Trogir City Museum (Fig. 1):

Find site: Solin, 1805 (?). Inventory number:3. Dimensions: height 74 cm, length 52 cm,width unknown. Literature: Babi 1984: 149,pl. XVIII, artifact and picture no. IV (drawingby I. Danilo). e same work was publishedin the Croatian language: Babi 1985: 77, fig.5, artifact and picture no. IV.

b. Fragment in the Archeological Museum inSplit (Fig. 3)

Find site: Solin, 1884. Inventory number: D116. Dimensions: height 63 cm, length 38 cm,width 28 cm. Literature: Rinaldi Tufi 1971:109, no. 34 (no picture!).

c. Lower half of a stela with almost entire in-scription field (Fig. 4)

Find site: Solin, 1805 (?). Formerly in the col-lection of I. L. Garagnin, today lost or at anunknown location. Assumed dimensionsbased on drawing by I. Danilo with addedmeasure: height 107 cm, length 86.3 cm, widthunknown. Literature: Pavlovi-Lui 1811:21f. (inscription); CIL3: 2197 (same); Babi1984: 149, pl. XVIII, artifact and picture no.XXVI (drawing by I. Danilo); Babi 1985: 77,fig. 5, artifact and picture no. XXVI (same).

2. Lower half of a funeral stela with portrait of theold woman Marcella (Fig. 8)

Find site: Trogir or its environs (?) Inventorynumber: A 22 (679). Dimensions: height 38 cm,length 28 cm, width 7.5 cm. Unpublished.

3. Fragment of a funeral stela (?) with male portrait(Fig. 9)

Find site: Trogir or its environs (?) Inventorynumber: 667. Dimensions: height 72 cm, length44 cm, width 27 cm. Unpublished.

1. Circumstances of finde title of this article cites the Trogir City Museumas the home institution of the three artifacts, which isonly partially true, as it only applies to artifacts undernumbers 2and 3. However, with reference to artifactno. 1, its remains are actually held at two locations.e larger fragment (1.a) is cemented into the wall inthe Trogir City Museum (!), while the smaller frag-ment is exhibited in the gardens of the ArchaeologicalMuseum in Split (1.b). e third fragment should beadded to these; it was formerly part of the Garagnin-

Fanfogna Collection in Trogir, but unfortunately it isnow lost (1.c). e fact that these fragments consti-tute parts of one and the same monument had not


4/36

114


en je poetkom 19. st. u Solinu, ali tone okolnostinalaza i poloaj s kojega potjee nisu poznati. Odtada se dugo vremena nalazio u zbirci trogirskogapolihistora i prvoga oficijelnog dalmatinskog kon-zervatora (glavnog nadzornika za antike i umjetni-ke spomenike) Ivana Luke Garagnina.

Godine 1805. Garagnin je vodio prva arheolokaiskopavanja na prostoru antike Salone. U travnju jeiskopavao na poloaju Grudine gdje je ve od ranijebio poznat termalni sklop sa sustavom hipokausta(Boi-Buani 1970: 155 i d.; Babi 1984: 136, 143i d., bilj. 31; Babi 1985: 69. 74, bilj. 31), a u lipnju naprostoru salonitanskog amfiteatra i u njegovoj oko-lici (Boi-Buani 1970: 156 i d.; Babi 1984: 136 id.; Babi 1985: 69). U obje ga je kampanje u svojstvudokumentarista (vjetaka-mjernika) pratio splitskislikar Ivan Danilo, koji je rezultate Garagninovih

istraivanja i konzervatorske djelatnosti dokumen-tirao dvama nacrtima izraenim 1805. godine. Namanjem je nacrtu prikazao spomenike koji su pro-naeni tijekom druge istraivake kampanje na pro-storu amfiteatra (Babi 1984: pl. XVII; Babi 1985:76, sl. 4), a na veemu (Sl. 2) spomenike koji su seu to vrijeme nalazili dijelom u Saloni, a dijelom uTrogiru, u vlasnitvu obitelji Garagnin (Babi 1984:pl. XVIII; Babi 1985: 77, sl. 5). Nije navedeno kojisu spomenici s drugoga nacrta pronaeni tijekomGaragninovih iskapanja.

Kako se ulomak portretne stele ne spominje meu

nalazima s dvaju spomenutih istraivanja, a na cr-teu se pojavljuje s drugim salonitanskim spome-nicima nepoznatog mjesta nalaza, nije jasno je liuope otkriven tijekom Garagninovih iskopavanja.U sluaju da je ipak otkriven tijekom prve kampa-nje, to je malo vjerojatno, kao mjesto nalaza valjalobi oznaiti istoni dio Salone (Grudine su, naime,naziv za prilino velik prostor unutar istonogaproirenja grada). Prvotni poloaj spomenika mo-rao bi se traiti na istonoj ili eventualno jugoisto-noj salonitanskoj nekropoli. Ako je pak pronaenna prostoru oko amfiteatra, vjerojatno bi potjecaosa zapadne salonitanske nekropole. ini se kako jeipak najvjerojatnija mogunost da je u Garagninovuzbirku doao posrednim putem otkupom.

Konzervatorska i sakupljaka djelatnost I. L. Ga-ragnina bili su jednako vani, a moda i najvani-

ji segmenti njegove aktivnosti, pa je tako jedna odpovjerenih mu zadaa bila nabavka spomenika kojisu mogli imati nacionalnu vrijednost (Boi-Buan-i 1970: 148). U tom je smislu dobivao i naputkeod zadarskoga namjesnika Tome Bradyja. Vrlo je

vjerojatno kako se nastanak drugoga Danilova na-

crta moe dovesti u vezu upravo s naputkom koji jeGaragnin dobio od Bradyja 18. lipnja 1805, u kojemga upozorava da povede vie rauna o popisivanju i

been realized until now, and its origins in Salonaand the unfortunate set of historical circumstancesresulted in the dispersal of its fragments. Since theintention of this work is not just to publish this cer-tainly most representative example in terms of itsmonumentality and quality craftsmanship, but also

the other two, one of the institutions was removedfrom the title, so that the emphasis is placed on onlythat institution that holds the majority of the artifact.

1.a. e larger fragment of themonumental portraitstela today held in the Trogir City Museum (fig. 1)was found at the beginning of the nineteenth centu-ry in Solin, but the precise circumstances surround-ing the find and the site at which it originated arenot known. Since then it was long a part of the col-lection of the Trogir polymath and first official Dal-matian conservator (chief supervisor in charge of

antique and artistic artifacts), Ivan Luka Garagnin.In 1805 Garagnin led the first archeological exca-vations in the territory of Roman Salona. In April

Slika 1.Fragment monumentalne portretne stele obitelji Attius izMuzeja grada Trogira (snimio N. Cambi).

Figure 1.Fragment of the monumental portrait stela of the Attiusfamily from the Trogir City Museum (photo by N. Cambi).


5/36

115


nabavci spomenika koji bi se mogli prenijeti u Be(Boi-Buani 1970: 151 i d., osob. 153154.; Ba-bi 1985: 78). Meu spomenicima koji su se u to

vrijeme nalazili po privatnim zbirkama i solinskimkuama mogao je biti i ulomak portretne stele 1.a,ali kako nakon nabavke nije nikad bio poslan u Be,zavrio je u Trogiru gdje je ugraen u zid uvenogaperivoja obitelji Garagnin. Tu je ostao sve do ezde-setih godina 20. stoljea. Tada je zbog svoje vrije-dnosti prenesen u Muzej grada Trogira gdje se nala-zi i danas. Uzidan je u tzv. sobi antike i zaveden podinventarnim brojem 3, kao jedna od prvih akvizicijaMuzeja.

1.b.Manji komad istoga spomenika uva se danasi u vrtu Arheolokoga muzeja u Splitu (sl. 3). ZaMuzej je otkupljen godine 1884, a za mjesto nala-za spominje se samo da je iz Salone. Za razliku odtrogirskoga fragmenta splitski je objavio S. Rinaldi

Tufi u svome poznatom radu o portretnim stela-ma u Arheolokome muzeju u Splitu (Rinaldi Tufi1971: 109, br. 34). Kratki komentar spomenika nije

he led a dig at the Grudine site, where a thermalcompound with a system of hypocausts had beenknown to exist even earlier (Boi-Buani 1970:155f.; Babi 1984: 136. 143f., note 31; Babi 1985:69. 74, note 31), while in June a dig was conductedat the Salona amphitheater and in the surround-ing area (Boi-Buani 1970: 156f.; Babi 1984:136f.; Babi 1985: 69). He was accompanied at bothprojects by painter Ivan Danilo from Split as a doc-umentarist (expert surveyor), who documented theresults of Garagnins research and conservation ac-tivities in two sketches made in 1805. e smallersketch showed the artifacts found during the secondexcavation in the amphitheater area (Babi 1984:pl. XVII; Babi 1985: 76, fig. 4.), while the largersketch (Fig. 2) shows the artifacts that were at thetime partly located in Salona and partly in Trogirand owned by the Garagnin family (Babi 1984: pl.

XVIII; Babi 1985: 77, fig. 5). ere is no mention ofwhich artifacts from the second sketch were foundduring Garagnins excavations.

Slika 2.Nacrt salonitanskih spomenika I. Danila iz 1805. godine (prema Babi 1984; 1985).

Figure 2.Sketch of the Salona monuments by I. Danilo from 1805 (according to I. Babi 1984; 1985).


6/36

116


bio popraen i fotografskom dokumentacijom pa sezbog toga i ovaj fragment jo uvijek moe smatratineobjavljenim. Rinaldi Tufi donosi i podatak kako jeinventiran pod brojem D 416, to nije tono (Rinal-di Tufi 1971: 109). Pod tim se brojem u inventarnojknjizi D pojavljuje nadgrobni rtvenik iz Vranjica,

dok je na spomenik u stvari uveden pod brojem D116.

1.c.Za trei komad monumentalne portretne stelete drugi koji se odreeno vrijeme nalazio u Trogiruu vlasnitvu obitelji Garagnin-Fanfogna u pogledumjesta nalaza vrijedi sve ono to je reeno za fra-gment 1.a. On je takoer mogao biti pronaen u je-dnoj od dviju istraivakih kampanja iz 1805, modaprije onoj iz lipnja te godine kada su se istraivanjaobavljala na prostoru amfiteatra, ali je isto tako uGaragninovoj zbirci mogao zavriti i otkupom. Sre-

tna je okolnost to ga je 1805. nacrtao I. Danilo, ito u mjerilu izraenom u venecijanskim stopama ipalcima (sl. 4), jer danas nije poznato to se s njimdogodilo (Babi 1984: pl. XVIII, br. XXVI; Babi1985: 77, sl. 5, br. XXVI). Nema ga u perivoju obiteljiGaragnin u kojem su jo uvijek uzidani brojni salo-nitanski spomenici, nije ni u Muzeju grada Trogira,a sigurno je kako nije dospio ni u Be.1Zahvaljujuimjerilu i grafici fragmenta moe se nedvojbeno do-kazati kako je on integralni dio spomenika kojem supripadali i fragmenti 1.ai 1.b.

Ve sm pogled na Danilov crte ukazuje na to kakofragment s natpisom, na nacrtu oznaen brojemXXVI (sl. 2), veliinom nadmauje sve ostale prika-zane spomenike nadgrobnoga karaktera. Vidljivo jei to da je sastavljen od etiriju ulomaka sljubljenihpo crti loma.2Namee se zakljuak da je rije o do-njoj polovici jednog veeg nadgrobnog spomenikakojem nedostaje znatan (gornji) dio prvobitnogkorpusa, barem iste visine. Na to jasno ukazuje po-prena koncepcija natpisnoga polja koja se obino

javlja upravo kod onih nadgrobnih spomenika koji-ma su iznad natpisa likovi pokojnika.

Najvaniji detalj Danilova crtea zapravo je lomgornjeg dijela lijeve okomite i veeg dijela gornjehorizontalne profilacije titulusa i stanje prvoga redaslova koji dokazuje da natpis uklesan na spomenikunije bio integralno sauvan. Ta je injenica vrlo va-na jer se iz objave . Mommsena moe stei dru-gaiji dojam (CIL 3: 2197).3Meutim da to nije takoi da Danilov crte reproducira stvarno stanje, jasnopokazuje transkripcija natpisa koju donosi I. J. Pa-

vlovi-Lui u svome djelu Marmora Traguriensia

Since the fragment of the portrait stela is not men-tioned among the finds from these two excavations,while it appears in the drawing with other Salonaartifacts of unknown find sites, it is unclear if it waseven discovered during Garagnins excavations. If itwas in fact discovered during the first excavation,

which is highly unlikely, then the find site should bemarked in the eastern part of Salona (Grudine is thename of a rather large area within the eastern exten-sion of the city). e original position of the artifactshould be sought in the eastern or possibly south-eastern Salona necropolis. If it were, however, foundin the area around the amphitheater, it would prob-ably have come from the western Salona necropolis.However, it seems most likely that it became a part ofGaragnins collection by indirect means: purchase.

Both the conservation and collection activities of I.L. Garagnin were equally important, and perhaps themost important aspect of his activities, so that oneof the task entrusted to him was to procure artifactsthat could have national value (Boi-Buani 1970:148). He also received instructions in this sense fromthe Toma Brady, the regent of Zadar. It is very likelythat the origin of Danilos second sketch can be con-nected to the instructions that Brady issued to Ga-ragnin on June 18, 1805, in which he enjoins him toplace greater emphasis on registration and procure-ment of artifacts that could be transported to Vien-na (Boi-Buani 1970: 151f., esp. 153154.;Babi

1985: 78). e artifacts found at the time in privatecollections and in Solins households could have in-cluded the fragment of portrait stela 1.a, but sinceit was never sent to Vienna after being obtained, itended up in Trogir, where it was built into the wallof the Garagnin familys famed gardens. ere it re-mained until the 1960s. en, due to its value, it wastransferred to the Trogir City Museum where it re-mains to this day. It was built into the wall in the so-called antique room and registered under inventorynumber 3 as one of the Museums first acquisitions.

1.b.esmallerfragmentofthesameartifactistoday

held in the gardens of the Archaeological Museumin Split (fig. 3). It was purchased for the museum in1884, while only Salona is mentioned as the site of itsdiscovery. In contrast to the Trogir fragment, the onein Split was published by S. Rinaldi Tufi in his well-known work on portrait stela in the ArchaeologicalMuseum in Split (Rinaldi Tufi 1971: 109, no. 34). Abrief commentary on the artifact was not accompa-nied by photographic documentation, so that this frag-ment can therefore still be considered unpublished.Rinaldi Tufi also states that it was inventoried undernumber D 416, which is not true (Rinaldi Tufi 1971:

109). A grave altar from Vranjic can be found underthis number in inventory ledger D, while the artifactin question is actually entered under number D 116.

1 To mi je osobno posvjedoio dr. Kurt Gschwantler, voditelj zbir-ke grkih i rimskih starina (Antikensammlung) pri Kunsthisto-risches Museumu u Beu.

2 Moda je upravo to bio razlog kasnijeg raspada i unitenja.


7/36

117


(sl. 5a; Pavlovi-Lui 1811: 21 i d.). Opisat emokako Pavlovi transkribira oteeni prvi redak: Napoetku retka otisnuto je vie tokica koje sugerira-

ju da taj dio teksta nije sauvan. Od prvoga je slovapreostala samo donja polovica jedne okomite haste(zanimljivo je da nju Danilo potpuno previa). Sli-

jedi slovo A koje je, navodno, potpuno sauvano,no da to nije moglo biti tako, potvruje uuvanostokolnih slova i tokice iznad. Ovdje je oito rije otome da se to slovo moglo bez dvojbe prepoznati irestituirati (da je slovu A uistinu nedostajao vrakpokazuje uostalom Danilov crte!). Iza A slijede dvaslova od kojih su opet sauvane samo okomite haste(detalj Danilova crtea ovdje je u suglasju s Pavlo-

viem). Iza njih je sauvano slovo I, a zatim dolazidonji dio slova V, sa spojem dviju dijagonalnih ha-sta, i na koncu sauvano slovo S (Danilo prepoznajeslovo I, detalj slova V prikazuje kao jo jedno I, doktrag slova S na njegovu crteu ima oblik donjeg kra-

1.c.In terms of the find site, everything said of frag-ment 1.a also applies to the third fragment of themonumental portrait stela and the second, whichwas held for a time in Trogir under the ownershipof the Garagnin-Fanfogna family. It also could havebeen found in one of the two research projects con-ducted in 1805, perhaps more likely the one in Juneof that year, when excavations were conducted in theamphitheater area, although it may have also ended

up in Garagnins collection by purchase. A fortunatecircumstance is that it was drawn in 1805 by I. Danilo,at a scale expressed in Venetian feet and inches (Fig.4), because today its fate is unknown (Babi 1984: pl.XVIII, no. XXVI; Babi 1985: 77, fig. 5, no. XXVI). Itis not in the gardens of the Garagnin family, wherethere are still many Salona artifacts cemented intothe walls, nor is it in the Trogir City Museum, and itcertainly did not make its way to Vienna.1anks tothe scale and the graphic portrayal of the fragment, it

Slika 3. (lijevo)Fragment monumentalne portretne stele obiteljiAttius iz Arheolokog muzeja u Splitu (snimio T. Sesar).

Figure 3. (left)Fragment of the monumental portrait stela of the Atti-us family from the Archeological Museum in Split (photo by T. Sesar).

Slika 4. (gore)Donja polovica stele obitelji Attius u crteu I. Da-nila (izgubljena) (arhiv Muzeja grada Trogira).

Figure 4. (above)Lower half of the Attius family stela in thedrawing by I. Danilo ( lost) (Archive of the Trogir City Museum).

3 Mommsen sam navodi da se sluio Pavlovievom objavom pa jejasno da je prvi redak nadopunio na temelju podataka iz sau-vanog dijela natpisa.

1 is waspersonally verified to me by Dr. Kurt Gschwantler, thehead of the Greek and Roman antiquities collection (Antiken-sammlung) at the Kunsthistorisches Museum in Vienna.


8/36

118


ka slova V). U nastavku retka slijede jo tri oteenaslova: donji dio slova M, kraj vertikalne i cijela hori-ozontalna hasta slova L te okomita hasta jo jednogslova (kod Danila je prvo slovo opet prikazano kaodonji krak slova V, dok su druga dva tono prepo-znata). Na kraju retka, u gornjem desnom uglu titu-

lusa, jasno se itaju slova VSTVS (sva se slova jasnovide i na Danilovu crteu).

Pomnim iitavanjem Pavlovieva prijepisa i uspo-rednim promatranjem Danilova crtea (Sl. 5) ne-dvosmisleno se dolazi do zakljuka kako nedosta-

jue dijelove prvih triju slova izgubljenoga natpisasusreemo upravo na ostatku natpisa sa splitskogfragmenta (1.b, Sl. 3). Prvo slovo na splitskomulomku je M kojemu nedostaje manji trag okomitedesne haste, a upravo je on bio sauvan na natpisuiz Trogira (Pavloviev prijepis). Da je trogirski nat-

pis moraozapoinjati slovom M, govori podatak dasu dvije prvoimenovane osobe,Attius Faustusi su-pruga muAttia Secunda, bili Markovi osloboenici.Na trogirskom je natpisu slovu A nedostajao vrak(detalj Danilova crtea), a upravo je on sauvan nasplitskom fragmentu. I na koncu, tree slovo s tro-girskog natpisa nesumnjivo je T, a nedostaje mu vo-doravna hasta, koja je neobine li sluajnosti sa-uvana ba na splitskom fragmentu.

Na temelju ove usporedbe moemo zakljuiti kakonema nikakve dvojbe da natpis sauvan u Pavlovi-evu prijepisu i Danilovu crteu (1.c) te splitski fra-

gment s likovima pokojnika i gornjim lijevim uglomnatpisa (1.b) tvore jednu cjelinu. tovie, moglo bise ii toliko daleko i ustvrditi da se dva ulomka namanjem potezu natpisnog polja spajaju i direktnopo crti loma. Tome bi se u prvi mah protivila jedi-no injenica da na splitskom fragmentu slova M.ATnalazimo uklesana odmah ispod obrubne profilacije(Sl. 3), dok se s Danilova crtea stjee dojam kako

je prvi redak bio neto vie odmaknut u odnosu nagornji obrub zrcala natpisa (Sl. 4). To je, meutim,posljedica neto nemarnijeg prikaza poloaja natpi-

sa u odnosu na stvarno stanje. Danilov nemar moese dokazati cijelim nizom primjera, a za ovu e pri-liku biti dovoljno kratko se zadrati na nadgrobnojsteli mladia Pudensa, danas uzidanoj u Garagnino-

vu perivoju, koju Danilo prikazuje na crteu br. XIXvelikoga nacrta iz 1805. Dok se na Danilovu crteuposljednji redak natpisa (P.B.M = posuit bene me-renti) nalazi pri dnu titulusa, i to tako da je izmeunjega i prethodnoga retka nastala praznina u visinicijeloga jednog retka, u stvarnosti je on bio odvojenod profilacije natpisa za istu tu visinu (Sl. 6). Taj idrugi slini primjeri pokazuju kako Danilo, nakon

to je spomenike nacrtao u mjerilu pazei na visinui irinu, nije uloio previe truda na detalje natpisa idekoracije, visinu i razmak redova i sline elemente.

can be stated with certainty that it is an integral part ofthe artifact to which fragments 1.aand 1.bbelonged.

Just a perfunctory glance at Danilos drawing indi-cates that it is a fragment with an inscription, des-ignated by number XXVI (Fig. 2), with a size thatsurpasses all of the other grave monuments shown.It is also apparent that it is composed of four frag-ments placed together at the fracture lines.2 isimposes the conclusion that it is the lower half ofa larger grave monument lacking a considerable(upper) portion of the original form of at least thesame height. is is clearly shown by the diagonalarrangement of the inscription field, which usuallyappears on those grave monuments that include theimages of the deceased above the inscription.

e most important detail in Danilos drawing is ac-tually the fracture in the upper left vertical portion as

well as most of the horizontal part of the titulus, andthe condition of the first line of letters, which showsthat the inscription engraved on the monument wasnot preserved in integral form. is fact is very im-portant because the publication by . Mommsengives readers a different impression (CIL 3: 2197).3However, that this is not so and that Danilos draw-ing reproduces the genuine article are clearly shownby the transcript of the inscription made by I. J.Pavlovi-Lui in his work Marmora Traguriensia(Fig. 5a; Pavlovi-Lui 1811: 21f.). Herein follows

a description of how Pavlovi transcribed the dam-aged first line: e beginning of the line containsthe impression of numerous dots which suggestthat this part of the text was not preserved. e onlyremaining part of the first letter is the lower half of a

vertical line (it is interesting that Danilo completelyoverlooked this). is is followed by the letter Awhich, allegedly, was completely preserved, but thepreserved state of the surrounding letters and thedots above indicate that this could not have beenthe case. Obviously the letter could be unambigu-ously recognized and restored (that the letter A was

only missing its top is shown by Danilos drawing!).e A is followed by two letters, of which, again,only vertical lines were preserved (the detail fromDanilos drawing here corresponds to Pavlovi). Af-ter them the letter I was preserved, followed by thelower portion of the letter V, with the merging oftwo diagonal lines, and finally the preserved letter S(Danilo recognized the letter I, the detail from theletter V is shown as another I, while the trace of the

2 ismay in fact be the reason for itssubsequent crumbling anddestruction.

3 Mommsen states that he used Pavlovis publication, so it isclear that he supplemented the first row on the basis of datafrom the preserved portion of the inscription.


9/36

119


Prethodno razmatrane pojedinosti koje povezujufragment iz Splita i onaj izgubljeni iz Trogira ponaj-prije su oblikovno-epigrafskog karaktera. Ono to

jo eksplicitnije pokazuje kako su to dijelovi istogaspomenika s jedne je strane sadraj natpisa, koji je

u suglasju s ikonografskim sadrajem (o tome viekasnije), a s druge su strane dimenzije, odnosno for-mativni razlozi.

Fragment uzidan u Muzeju grada Trogira visok je 74cm, irok 52 cm, a debljinu mu nije mogue izmje-riti. Kada se isti komad premjeri na Danilovu na-crtu, preraunavanjem venecijanskih stopa i palacadolazi se do neto manjih vrijednosti: visina tadaiznosi blizu 71 cm, a irina 49 cm.4Visina i debljinasplitskog fragmenta nemaju za nas veliku vanost, airina mu je 38 cm. I to je, meutim, relativna vrije-dnost izmjerena na mjestu gdje je sauvan u najve-

oj irini. Integralnu irinu dvaju fragmenta mogueje izmjeriti jedino tamo gdje oni formiraju punu i-rinu izvornoga spomenika. U visini glava prikazanedjece trogirski je fragment irok 50,5 cm, splitski34,3 cm, to je ukupno 84,7 cm ili oko 85 cm, dok suu visini ruku likova roditelja te vrijednosti 47 cm i35 cm, odnosno ukupno 82 cm.

Kada se mjerilo u venecijanskim stopama i palcimaprerauna i za izgubljeni fragment s natpisom, do-lazi se do saznanja da mu je visina iznosila 107 cm,a irina 86 cm. Prema tome, razlika u irini izme-

letter S in his drawing takes the form of the lowerextension of the letter V). e rest of the line con-tains three more damaged letters: the lower portionof the letter M, the end of the vertical and entirehorizontal line of the letter L and the vertical line ofanother letter (Danilo against showed the first letter

as the lower extension of the letter V, while the oth-er two were accurately recognized). At the end ofthe line, in the upper right-hand corner of the titu-lus, the letters VSTVS can be clearly read (all of theletters are also clearly visible in Danilos drawing).

A careful reading of Pavlovis transcript and acomparison with Danilos drawing (Fig. 5) leads tothe unmistakable conclusion that the missing partsof the first three letters of the lost inscription canbe found on the remainder of the inscription fromthe Split fragment (1.b, Fig. 3). e first letter in

the Split fragment is M, on which a smaller traceof the vertical right line is missing, and this wasprecisely preserved in the inscription from Trogir(Pavlovis transcript). at the Trogir inscriptionmusthave begun with the letter M is shown by thefact that the first two persons named,Attius Faus-tus and his wife Attia Secunda, were freed slavesof Marcus. On the Trogir inscription, the top ofthe letter A is missing (detail from Danilos draw-ing), while just this is preserved in the Split frag-ment. And finally, the third letter on the Trogirinscription is undoubtedly T, while its vertical line

is missing, althoughby unusual coincidencepre-cisely this detail is preserved on the Split fragment.

Based on a comparison, we can conclude thatthere is no doubt that the inscription preserved inPavlovis transcript and Danilos drawing (1.c) andthe Split fragment with the images of the deceasedand the upper left-hand corner of the inscription(1.b) form a single piece. Moreover, one could goso far as to assert that the two fragments are con-nected directly along the fracture line in a smallersection of the inscription field. In the first instance,

this would only be contradicted by the fact that onthe Split fragment the letters M.AT are found en-graved immediately under the border profile (Fig.3), while Danilos drawing creates the impressionthat the first line was somewhat removed in rela-tion to the upper border of the mirrored inscription(Fig. 4). is is, however, the result of a somewhatmore careless portrayal of the inscription in relationto the actual situation. Danilos carelessness can beshown with an entire series of examples, but fornow it will be sufficient to briefly focus on the gravestela of the young man Pudens, today raised in Ga-

ragnins gardens, which Danilo portrays in drawingno. XIX from the large drawing of 1805. While thelast line of the inscription (P.B.M =posuit bene mer-

Slika 5.Razlike u prikazu prvog retka natpisa sa stele obiteljiAttius: a. prijepis I. J. Pavlovia-Luia (Pavlovi-Lui 1811), b.crte I. Danila (arhiv Muzeja grada Trogira).

Figure 5.Differences in portrayal of the first row of the inscriptionon the Attius family stela : a. transcript by I. J. Pavlovi-Lui (Pa-vlovi-Lui 1811), b. drawing by I. Danilo (archive of the TrogirCity Museum).

4 1 stopa=12 palaca. Duina palca iznosi 2.875 cm, dok je duinastope 34.5 cm: Madirazza 1911: 431 i d.


10/36

120


u gornjeg dijela stele sa slikovnim poljem i donjegdijela s natpisom iznosi beznaajnih 1 cm 4 cm(ova druga vrijednost zasigurno je posljedica togato prva dva ulomka nisu sljubljeni jedan s drugim),to se moe smatrati i glavnim argumentom njihovezajednike pripadnosti.

Na kraju je potrebno konstatirati kako Garagnin niuz najbolju volju nije mogao zamijetiti injenicu dasu dva komada u njegovoj zbirci, Danilovi fragmentibr. IV i XXVI, dijelovi istoga spomenika. Dva ulom-ka jednostavno se nisu spajala po crti loma; donji diofragmenta 1.anije mogao nalijegati na najvii dio na-tpisa1.c(gornji desni ugao) jer je u tom dijelu bio osa-kaen za bazu i donji dio stabla tordiranoga stupia.

2. Tone okolnosti, mjesto i godina nalaza donjegadijela male portretne stele iz Muzeja grada Trogiratakoer nisu poznati (sl. 8). Gotovo je sigurno kako

potjee iz samoga Trogira ili njegove najblie oko-lice, ali to nije mogue potkrijepiti nekim vrimargumentima. U Dravnoj upravi za zatitu kultur-ne i prirodne batine, u Konzervatorskome odjeluu Splitu, poznat je od godine 1973, kada je doku-mentirana cijela tadanja graa trogirskoga muzeja.Prema tome, moe se jedino rei da je naen prijespomenute godine.

3. O treem spomeniku iz Muzeja grada Trogira(sl. 9) raspolaemo gotovo jednakim informacija-ma kao i za prethodni. I on je po svoj prilici pro-naen u Trogiru ili njegovoj neposrednoj okolici,ali nita drugo osim toga nije poznato. Nije isklju-eno da se pod fragmentom zapravo krije ostataknekog od ranije poznatoga nadgrobnog spomeni-ka. Tu mogunost u ovome trenutku nije moguerasvijetliti.

2. Oblik

1.a. Trogirski fragment monumentalne portretnestele pripada slikovnome polju izraenom u obliku

duboke nie uokvirene polustupiima tordiranogtijela. Sauvala se gornja polovica desnoga polustu-pia prislonjenog uz istak zamiljenoga zida nie(antu). Stupi zavrava pseudokorintskim kapite-lom sa stiliziranim, zailjenim akantusovim listii-ma. Tijelo stupia od kapitela odvaja dvostruka pr-stenasta profilacija (anuli). Na abaku kapitela poi-

va impost iji je vrh viestruko profiliran: u donjemdijelu S profilacijom, a u gornjem jednostavnomravnom trakom. Nad njim je sauvan mali, ali zna-ajan trag kose stranice zabata (katete). On svjedoida je stela zavravala slobodnostojeim trokutastim

zabatom. Zabat je sa strana imao akroterije, a jedanje mogao stajati i na njegovu vrhu. Od njih nemanikakvih tragova.

enti) is found near the bottom o the titulus in Dani-los drawing, such that between it and the precedingline there is a void with the height of an entire line,in reality is was actually separated from the inscrip-tion profile by that same height (Fig. 6). is andother similar examples show that Danilo, after draw-

ing the monuments to scale and seeing to the heightand length, did not pay excessive attention to thedetails of inscriptions and decorations, the heightand spaces between lines and similar elements.

e similarities considered above which link thefragment from Split and the lost fragment fromTrogir have a primarily figural-epigraphic charac-ter. More explicit evidence that they are parts of thesame monument is the content of the inscriptionon the one hand, which corresponds to the icono-graphic content (more on this later), and the dimen-

sions, i.e. formative reasons, on the other.e fragment built into the wall of the Trogir CityMuseum is 74 cm high and 52 cm long, while its widthcannot be measured. When the same piece is meas-ured by the scale of Danilos drawing, a recomputa-tion of Venetian feet and inches produces somewhatsmaller values: the height is then near 71 cm, and thelength 49 cm.4e height and width of the Split frag-ment bear little importance to this discussion, whilethe length is 38 cm. is is also, however, a relativevalue measured at the point where it was preserved at

the greatest length. e integral lengthof thetwo frag-ments can be measured only where they form the fulllength of the original monument. At the height of theheads of the portrayed children, the length of the Tro-gir fragment is 50.5 cm, while that of the Split fragmentis 34.3 cm, which is a total of 84.7 or approximately 85cm, while at the height of the hand of the parents im-ages that value is 47 cm and 35 cm, for a total of 82 cm.

When the scale in Venetian feet and inches is rec-omputed for the lost fragment with the inscription,then one arrives at the data that the height was 107cm, while the length was 86 cm. erefore, the dif-

ference in width between the upper portion of thestela with its image field and the lower portion withthe inscription is an insignificant 1 to 4 cm (the sec-ond value is certainly the result of the fact that thefirst two fragments are not placed one next to theother), which can be deemed the principal argu-ment in favor of their common origin.

In the end it is necessary to assert that Garagnin,even under the best circumstances, could not havenoticed the fact that the two fragments in his collec-tion, Danilos fragments no. IV and XXVI, were parts

4 1 foot = 12 inches. e length of an inch is 2.875 cm, while thelength of a foot is 34.5 cm: Madirazza 1911: 431f.


11/36

121


Skromni ostatak kose stranice zabata dostatan je iza analizu izgleda njegovih stranica. One su iznu-tra bile obrubljene tzv. S profilacijom, a izvana je-

dnom ili dvjema ravnima trakama. Zabat nije imaogredu ni druge horizontalne elemente klasinetrabeacije jer je to onemoguio prikaz pokojnika,odnosno oblik nie izdubljene unutar stupia. Nia

je u donjem dijelu imala pravokutan oblik, ali je svo-jim vrhom probila zamiljenu horizontalnu stranicuzabata i srasla s njim. Na taj su nain slikovno polje sportretima i zabat izraeni kao jedna cjelina.

Unutar nie vidljivi su ili se naziru likovi etvero

pokojnika. Dvoje pokojnika mlae ivotne dobi

prikazano je u prednjem planu, a njihovi roditelji

stoje iza njih. Sva etiri lika izraena su u visini po-lufigura. Desni par je vrlo dobro uuvan, ako zane-

marimo portret djeaka, dok se lijevi par do otkria

of the same monument. e two fragments simplydid not fit together at the fracture line; the lowerportion of fragment 1.a could not connect to the

highest portion of inscription 1.c(upper right-handcorner) because here it is bereft of the base and low-er portion of the trunk of the small twisted column.

2. e precise circumstances, site and year of dis-covery of the lower portion of the small portrait ste-la from the Trogir City Museum are not known (Fig.8). It is almost certain that it comes from Trogir itselfor its immediate vicinity, but this cannot be backedby any solid arguments. e National Cultural andNational Heritage Protection Agencys Conserva-tion Department in Split has had knowledge of its

since 1973, when all of the materials from the TrogirMuseum at the time were documented. erefore,one can only state that it was found prior to that year.

Slika 6.Usporedni prikaz Danilova crtea stele djeaka Pudensa (arhiv Muzeja grada Trogira) i spomenika uzidanog u perivoju Gara-gnin-Fanfogna (snimio D. Mari).

Figure 6.Comparison of Danilos drawing of the stela of the boy P udens (archive of the Trogir City Museum) and the monument raised inthe Garagnin-Fanfogna gardens (photo by D. Mari).


12/36

122


splitskog dijela spomenika mogao smatrati gotovo

izgubljenim. O ikonografskim, modnim i stilskim

znaajkama portreta pokojnika bit e rijei neto

kasnije.

U gornjem dijelu trogirskog fragmenta, iza glavamukarca i njegove supruge, a u funkciji njihovepozadine, izraena je koljka sa irokim, plastinonaznaenim rebrima koja se izvijaju iz donjeg dijelakoljke prema vanjskoj strani. koljka nema samodekorativnu, ve i isto praktinu svrhu jer je nje-zinom izradom ublaen sraz visoke nie s portreti-ma i nepostojanoga zabata. Pri tome je vanu uloguodigrala praksa takve njezine uporabe u stvarnojarhitekturi.

1.b. Splitski fragment istoga spomenika sauvan jeu neto manjoj visini, i to tako da je gornji dio niesa koljkom u potpunosti izgubljen, ali je uz poprsja

pokojnika preostao manji gornji lijevi ugao natpi-snoga polja. Od dimenzija je svakako najzanimlji-

vija debljina, koja nedostaje kod trogirskog dijela, aiznosi 28 cm.

Splitski komad monumentalne portretne stele upogledu ikonografije i rasporeda likova pokojnikane nudi nita novo. Na njemu prikazani lik starijegpokojnika dominira pozadinom nie, a u prednjem

je planu lik djeteta. Via je figura odbijena od vra-ta navie, a nioj je sauvan samo oval desne stranekalote, dok je lice takoer odbijeno. Najzanimljiviji

detalj splitskog ulomka zapravo je izgled tordiranogpolustupia. On je izraen isto kao i onaj na desnojstrani nie (trogirski fragment). Pritom su tordiranekanelire izraene u suprotnom smjeru, ali je zani-mljivo da stablo stupia od baze odvaja jednostavnai niska prstenasta profilacija. Druga je zanimljivostizrada baze koja nije sauvana kod desnoga stupia.Ona je od rubne trake natpisnoga polja i stajne trakeportreta odvojena jedino visinom, ali s prednje stra-ne nema nikakve stube ili profila.

1.c. Izgubljeni trogirski fragment pokazuje kaoto se uostalom moglo i oekivati da je donji diomonumentalne portretne stele zauzimalo natpisnopolje izraeno poput pravokutnoga, po horizontaliizduenog titulusa. On je bio uokviren neukrae-nim ravnim trakama, a sam je natpis bio obrubljenklasinom S profilacijom. Unutar tako priree-nog okvira bio je uklesan natpis rasporeen u ak

jedanaest redaka. U donjem dijelu titulusa izraenje uobiajeni sadraj za ovakav tip nadgrobnih spo-menika: ravno i zacijelo neto grublje obraeno po-dnoje. U neko je vrijeme na njega uklesan jo jedannatpis, zbog ega je epigrafski sadraj ispunio cijelu

donju polovicu izvornoga spomenika. Tako je po-dnoje postalo sastavni dio titulusa ne samo stru-kturalno, nego i sadrajno.

3. Almost the same amount of information exists onthe third monument from the Trogir City Museum(Fig. 9). It was also found in Trogir or its immediate

vicinity, but nothing more is known about it. epossibility that the remains of some earlier knowngrave monument are concealed under the fragment

cannot be excluded. But no light can be shed on thisprospect at the moment.

2. Form

1.a.e Trogir fragment of a monumental portraitstela has a pictorial field made in the form of a deepniche framed by twisted demi-columns. e upperhalf of the right-hand demi-column leaning on a pro-trusion of an imaginary wall of the niche (ante) is stillvisible. e column ends with a pseudo-Corinthiancapital and stylized, sharpened Acanthus leaves. etrunk of the small column is separated from the capi-tal by a double ring-like profile (anuli). ere is animpost on the abacus of the capital and its top hasmultiple profiles: an S profile in the lower portion,and a simple straight border on the upper portion. Asmall but significant remainder of the slope of a gablehas been preserved above it. It testifies to the fact thatthe stela ended with a free-standing triangular gable.e gable had acroteria on the sides, and one mayhave stood on its peak. ere are no traces of these.

emodestremainsofthegablesslantingsideissuf-ficient for an analysis of the appearance of its sides.ese were bordered with so-called S profiles fromthe inside, with one or two straight borders on theoutside. e gable did not have a column nor anyother horizontal elements of classical trabeation,because this prevented portrayal of the deceasedand the form of the niche impressed within the col-umn. e niche had a rectangular form in the lowerportion, but its top passed through the imaginedhorizontal side of the gable and extends with it. Inthis manner the pictorial field with portraits and the

gable were crafted as a single whole.eimagesoffourdeceasedpersonsarevisiblewith-in the niche. Two youths are shown in the forefront,while their parents stand behind them. All four imag-es are crafted at the height of semi-figures. e cou-ple to the right is very well preserved, if the portraitof the young boy is overlooked, while the couple tothe left can be deemed almost completely lost. eiconographic, fashionable and stylistic features ofthe portraits of the deceased will be dealt with later.

e upper portion of the Trogir fragment, behind

the heads of the man and his wife as part of theirbackground, contains a shell with broad, plasti-cally-designated ribs that curve from the lower part


13/36

123


Pitanja na koja u ovom trenutku nije mogue sa si-gurnou odgovoriti jesu: spajaju li se sauvani tro-girski i splitski komad direktno po crti loma i kako sespomenik u svojoj punoj visini uglavljivao u postolje?

Na prvo bi pitanje bilo mogue odgovoriti samo akose pokua direktno spojiti dva sauvana fragmenta,

dakle ako se kratkotrajno posudi splitski fragmentMuzeju grada Trogira ili obratno. U ovom trenutkuto nije mogue, no svakako se treba napraviti u bu-

of the shell toward the outside. e shell does notsimply have a decorative purpose; it has an equallypractical purpose, as it alleviates the clash betweenthe high niche with the portraits and the gable with-out a base. Here an important role is played by thepractice of such use in actual architecture.

1.b.e Split fragment of the same monument hasbeen preserved at a somewhat lower height, such thatthe upper portion of the niche with shell has beencompletely lost, although the smaller upper left-handcorner of the inscription field has been preservedalong with the bust of the deceased. Among the di-mensions, the most interesting is certainly the width,which is lacking for the Trogir portion, and it is 28 cm.

e Split fragment of the monumental portrait stelaoffers nothing new in terms of iconography and thearrangement of the images of the deceased. On it,

the image of the older persons dominates the back-ground of the niche, while the children are in theforeground. e taller figure is broken off from theneck upward, while only the right side of the crani-um on the shorter figure has been preserved, whilethe face is also broken off. e most interestingdetail of the Split fragment is actually the appear-ance of the twisted demi-column. It is made in thesame manner as that on the right-hand side of theniche (Trogir fragment). Herein the twisted flutingis crafted in the opposite direction, but it is interest-ing that the trunk of the columns is separated from

the base by a simple and low ring-like profile. Alsointeresting is the crafting of the base, which was notpreserved on the right-hand column. It is dividedfrom the border line of the inscription field and thebase border of the portrait solely by height, althoughthere are no steps or profiles on the front.

1.c.e lost Trogir fragment showsas could havebeen expectedthat the lower portion of the monu-mental portrait stela was an inscription field madeas a rectangular horizontally extended titulus. It wasframed with non-ornamented straight borders, andthe inscription itself was bordered by a classic S pro-file. Within this frame, the inscription was engravedin as many as eleven lines. In the lower portion of thetitulus, the customary content for this type of gravemonument: a straight and overall somewhat coarselymade base. At some point another inscription wasengraved on it, which is why the epigraphic contentfilled in the entire lower half of the original monu-ment. us the base became a component of the titu-lus not just structurally, but also in terms of content.

Questions that cannot be answered with certaintyat this moment are: can the preserved Trogir and

Split fragments be connected at the fracture lineand how was the monument set onto the base at itsfull height?

Slika 7.Rekonstrukcija stele obitelji Attius(D. Mari & J. Zaninovi).

Figure 7.Reconstruction of the Attius family stela(D. Mari & J. Zaninovi).


14/36


15/36

125


polufiguru mukarca smjetenu u niu ovalnogapresjeka. Prema lomu desnoga ruba spomenika idubini nie na tom mjestu lako se moe zakljuitida se lijevo od mukarca nalazio jo jedan lik, vjero-

jatno lik njegove supruge.

3. Tipologija

Iz prethodnog se poglavlja moe zakljuiti da suprve dvije stele razliite i sa strukturalnog i s forma-tivnog aspekta, tj. da je rije o razliitim izvedbenimuzorcima. To e nas u konanici odvesti do zaklju-ka o njihovoj razliitoj tipolokoj pripadnosti.

1.a-c.Prva je stela izraena po uzoru na arhitektu-ru mauzoleja, to znai da je pomno arhitektonskiosmiljena. To pokazuje analiza vanjske forme spo-

menika i njegova unutarnjeg ustrojstva te meuso-bni odnos tih dviju sastavnica. Gornji dio spome-nika strukturalno je koncipiran u obliku edikule.Elementi su edikule stupovi i drugi karakteristinidijelovi klasine trabeacije, a od posljednjih je pose-bice znaajna prisutnost slobodnostojeeg (plasti-nog) trokutastog zabata. To je najei nain obli-kovanja gornjega dijela nadgrobnih stela zbog egase takvi primjerci definiraju kao arhitektonske stelezabatne forme, tj. stele u obliku edikule ili naiskosstele (Behrens 1951: 39 i d.; Mansuelli 1956: 368 id.; Mansuelli 1967: 115 i d.; Gabelmann 1972: 75;

Pflug 1989: 39 i d.).Najea inaica stela u obliku edikule jesu steles pravokutnim slikovnim poljem u kojem je prika-zan lik jednog ili vie pokojnika. Mogao bi se navestiuistinu velik broj stela toga tipa s prostora rimskeprovincije Dalmacije: stela Gaja Licinija iz Hardo-milja kod Humca (Pakvalin 1985: 120, br. 1, sl. 1),stela Gaja Utija iz Salone (Rinaldi Tufi 1971: 9495,br. 6, tab. IV, sl. 1; Cambi 1989: 34, bilj. 1, tab. I), ste-la Sergija Enija Fuska (Rinaldi Tufi 1971: 99, br. 13,tab. V, 3; Cambi 2000: 50, br. 63, tab. 82), stela Gaja

Publicija Romana iz Narone (Prijatelj 1952: 149, tab.XI; Cambi 1980: 135, sl. 8) itd. Neto je rjea inaicas istim poljem u kojem je izdubljena luno zasvoe-na ili jednostavno natkriljena nia. Najljepi primjerstele toga tipa je ona Lucija Cezija Basa, danas izgu-bljena (Cassas 1802: 138, tab. 31 [=1997: 233, tab.139]; Hofmann 1905: 5859, sl. 38). U oba sluajagornju stranicu slikovnoga polja zatvara arhitrav te

je rije o temeljnoj tipolokoj varijanti arhitekton-skih stela stelama u obliku edikule s arhitravom(Gabelmann 1972: 73 i d.; Pflug 1989: 39 i d.). Tre-a je inaica (inae druga po rairenosti, odmah iza

prvospomenute) ona kod koje slikovno polje nijenatkriljeno gredom ili nekim drugim elementomu funkciji meupolja prema zabatu. Ono je svojim

fied based on this alone, but the inscription clearlystates that it is an elderly woman. In this sense, it isnot without merit to note that the clothing on thedeceased is also important. is will be covered ingreater detail in the section on portraits.

3. e form of the third Trogir fragment can be ex-amined within the framework imposed by its state ofpreservation. e monument is broken on all sides(it would appear that only the back is undamaged),while the only recognizable aspects of its compositionelements are the semi-figure of a male set in a nichewith an oval cross-section. e fracture of the rightedge of the monument and the depth of the niche atthis point leads to the conclusion that another fig-ure stood to the left of the man, probably his wife.

3. Typology

e preceding sections indicate that the first twostelae are different from both the structural andformative aspect, i.e. that these are differently ex-ecuted samples. is will ultimately lead to conclu-sions on their different typologies.

1.a-c.e first stela was made in compliance withmausoleum architectural models, which meansthat it was carefully designed architecturally. isis shown by an analysis of the external form of themonument and its internal structure, and the inter-

action between these two components. e upperportion of the monument was structurally con-ceived in the form of an edicule. e elements of theedicule are columns and other characteristic com-ponents of a classical trabeation, and among the lat-ter the presence of a free-standing (plastic) triangu-lar gable is particularly significant. is is the mostfrequent method for forming the upper portion ofthe funeral stela, which is why such examples aredefined a gable-form architectural stelae, i.e. edi-cule-form stelae or naiskos stelae (Behrens 1951:39f.; Mansuelli 1956: 368f.; Mansuelli 1967: 115f.;Gabelmann 1972: 75; Pflug 1989: 39f.).

e most frequent variant of edicule-form stelaeare stelae with rectangular pictorial fields in whichthe image of one or more deceased persons is shown.A truly immense number of this type of stelae in theterritory of the Roman province of Dalmatia can becited: the stela of Gaius Licinius from Hardomiljenear Humac (Pakvalin 1985: 20, no. 1, fig. 1), thestela of Gaius Uttius from Salona (Rinaldi Tufi 1971:9495, no. 6, tab. IV, fig. 1; Cambi 1989: 34, note 1,tab. I), the stela of Sergius Enius Fuscus (Rinaldi Tufi

1971: 99, no. 13, tab. V, 3; Cambi 2000: 50, no. 63, tab.82), the stela of Gaius Publicius Romanus from Nar-ona (Prijatelj 1952: 149, tab. XI; Cambi 1980: 135, fig.


16/36

126


gornjim dijelom sraslo sa zabatom, zapravo probilozamiljenu liniju arhitrava, pa govorimo o stelamau obliku edikule bez arhitrava ili stelama s tzv. pre-kinutim zabatom (Gabelmann 1972: 81 i d. s listomprimjera i tablicom tipova na sl. 42; Pflug 1989: 39i d.). Takav nain spajanja slikovnoga polja sa zaba-

tom kod stela u dalmatinskome priobalju rijeen jena dva naina: kod jednih je greda odnosno vijenackoji bi fiktivno trebao nositi teinusvoda sa zabatomipak vidljiv iza lea pokojnika, dok je kod drugih izalea pokojnika, zapravo iza njihovih glava, izrae-na koljka. Prva je varijanta rijetka i zastupljena sasvega dva primjerka, oba iz Kouta kod Garduna(Tilurium): stelom Melvadija, konjanika Nove kla-udijevske ale (Hofmann 1905: 47; Rinaldi Tufi 1971:97, br. 10, tab. IV, sl. 3) i stelom s posve izlizanimnatpisom, po svoj prilici podignutoj vojniku iste je-dinice (Rinaldi Tufi 1971: 9798, br. 11, tab. V, sl. 1).

Evidentno je kako naa stela, rekonstruirana na te-melju dvaju trogirskih i jednoga splitskog fragmen-ta (1.a-c), pripada upravo posljednjoj tipolokoj

varijanti. Ona je u gornjem dijelu zavravala troku-tastim zabatom koji nije imao horizontalnu profila-ciju ili gredu, ve je taj prostor ispunjavala koljkaizraena kao pozadina portretima pokojnika.

Zbog irokoga simbolikog spektra koljka je imalaznaajnu ulogu u rimskoj sepulkralnoj umjetnosti,posebice na sarkofazima gdje se javlja kao sredinjimotiv (clipeus) s likovima pokojnika (Brandenburg

1967: 195 i d.; Matz 1971: 104 i d., sl. 11; Koch &Sichtermann 1982: 196). Na nadgrobnim se stela-ma, pa tako i onima u dalmatinskome priobalju,takoer javlja u funkciji pozadine polja s likovimapokojnika. Ulogu pozadine slikovnoga polja kojenije arhitektonski uokvireno koljka ima samo nadvjema stelama. Prva je stela, zapravo njezin ma-nji fragment, pronaena na istonoj salonitanskojnekropoli (Cambi 1993: 351 i d., sl. na str. 352).koljka je izraena unutar polukrunog okvira (lu-nog zabata), a njegov vanjski rub izveden u obliku

jednostavne trake dodatno je dekoriran urezanomvalovnicom. Druga je stela izvorno bila kudikamomanjih dimenzija, a potjee iz Narone (Rinaldi Tufi1971: 107, br. 28, tab. IX, sl. 1; Cambi 1980: 135, sl.9). koljka ulijepljena u prostor trokutnoga poljakod nje ima oblik pseudozabata, i to tako da se rebrakoljke penju prema njegovim stranicama. Za po-sljednju je stelu sigurno, a za prvu gotovo sigurno,da nije imala arhitektonsku dekoraciju.

Spomenutim bi se primjerima eventualno mogaopribrojiti i fragment salonitanske stele s glavommlaega mukarca koji se uva u Arheolokome

muzeju u Splitu (Rinaldi Tufi 1971:94, br. 5, tab. III,sl. 2). I tu je koljka umetnuta u polje polukrunogzavretka s rubom izvedenim ravnom trakom. U

8) etc. A somewhat rarer variant features the samefield with a concave, arched or simply covered niche.e most attractive example of this type of stela isthat of Lucius Caesius Bassus, today lost (Cassas1802: 138, tab. 31 [=1997: 233, tab. 139]; Hofmann1905: 5859, fig. 38). In both cases the upper side

of the pictorial field is closed by an architrave, andthis is a fundamental typological variant of the ar-chitectural stela: stelae in the form of an edicule witharchitrave (Gabelmann 1972: 73f.; Pflug 1989: 39f.).e third variant (otherwise the second most wide-spread, immediately after the first-mentioned vari-ant) is that with a pictorial field that is not covered bya beam or some other element in the function of aninterval space toward the gable. e upper portionmerges with the gable, actually breaking throughthe imaginary line of the architrave, so we speak of

edicule-form stelae without architraves or stelaewith so-called disjointed gables (Gabelmann 1972:81f. with a list of examples and a table of types in fig.42; Pflug 1989: 39f.). is manner of connecting thepictorial field with the gable in stelae in the Dalma-tian coastal belt was resolved in two ways: for somethe beam or wreath that should fictitiously hold theweight of the arch with the gable is nonetheless vis-ible behind the backs of the deceased, while for oth-ers there is a shell made is behind the backs of thedeceased, actually above their heads. e first vari-ant is rare and present in only two examples, both

from Koute, near Gardun (Tilurium): the stela ofMelvadius, the cavalryman of the New Claudian ala(Hofmann 1905: 47; Rinaldi Tufi 1971: 97, no. 10,tab. IV, fig. 3) and a stela with a completely worn in-scription, probably raised to a soldier from the sameunit (Rinaldi Tufi 1971: 9798, no. 11, tab. V, fig. 1).

It is evident that our stela, reconstructed on thebasis of two Trogir and one Split fragment (1.a-c),belongs to the latter typological variant. It ended ina triangular gable in its upper portion which did nothave a horizontal profile or beam, rather this spacewas filled by a shell made as the background to theportrait of the deceased.

Due to its broad symbolic spectrum, the shell played a

major role in Roman sepulchral art, especially on sar-

cophagi, where it appears as a central motif (clipeus)

bearing images of the deceased (Brandenburg 1967:

195f.; Matz 1971: 104f., fig. 11; Koch & Sichtermann

1982: 196). It also appears as a background to the field

containing portraits of the deceased on grave stelae,

including those on the Dalmatian coast. The role of the

background to the pictorial field which is not an archi-

tecturally framed, shell is only present in two stelae.

The first stelae, actually a smaller fragment, was foundon the eastern Salona necropolis (Cambi 1993: 351f.,

fig. on p. 352). The shell is crafted within a semi-circu-


17/36

127


koljci je stajao samo lik mlaeg mukarca prikazanvjerojatno u visini polufigure. To zorno dokumenti-raju rebra koljke koja se spajaju u srcu tono iznadglave pokojnika. Problem ovog fragmenta vezan jeuz mogunost postojanja arhitektonskog okviraispod lunog zavretka (dakle, kao njegova po-

tpornja) ili eventualno u donjem dijelu spomenika.

Kod svih ostalih primjera koljka stoji iskljuivo ufunkciji pozadine likova pokojnika i obavezno je ar-hitektonski uokvirena tako da je flankiraju stupii ilipilastri, a svojim gornjim dijelom probija horizon-talnu stranicu zabata. Prema poloaju koljke i ta senajbrojnija skupina spomenika moe podijeliti nasljedee dvije grupe.

Kod prve je grupe koljka poloena tako da je nje-zin iri kraj okrenut prema gore, a ui (srce) premadolje, pa se tako rebra radijalno ire odozdo prema

van. Pritom srednja dva rebra obino prolaze tonoizmeu glava prikazanih pokojnika. Takav nain po-stavljanja koljke unutar slikovnoga polja prisutan

je na unitenoj salonitanskoj steli otkrivenoj 1826.na zapadnoj nekropoli (Lanza 1856: 33, tab. VII), nasteli M. Antonija Celera sauvanoj na crteu Filiber-ta Pingonea (CIL 3: 3162a; Hofmann 1905: 49 i d.),steli Marka Pythe, vojnika 2. kohorte Kiresta iz Gor-njih Rakiana kod Garduna (CIL 3: 14934; Prijatelj1952: 143; Rinaldi Tufi 1971: 98, br. 12, tab. V, sl. 2;Cambi 2000: 49, br. 61, tab. 80) te na fragmentu steleiz Salone s glavom starijeg mukarca (Rinaldi Tufi

1971: 95, br. 7, tab. III, sl. 3; Cambi 1987: 146, br.46 sa sl.; Cambi 2000: 34, br. 23, tab. 19). Kod svihnavedenih primjera, osim kod posljednjeg, jasno jemjesto koljke u odnosu na strukturalnu kompozi-ciju spomenika i broj prikazanih likova pokojnika.Iako je odlomljen sa svih strana, nema nikakve dvoj-be kako i posljednji primjer pripada steli u oblikuedikule bez arhitrava. U slikovnom polju stajali sulikovi dvoje pokojnika (zacijelo brani par), ali jesauvana samo glava starijeg mukarca. To jasnopokazuje poloaj dvaju sredinjih rebara koljke slijeve strane glave mukarca. Ona se pruaju iz srca,

smjetenog otprilike iza glava pokojnika, premavanjskoj strani. Ostat e nepoznato koliko je likovaukupno bilo prikazano jer su oni mogli biti raspore-eni u jednom polju, u dva superponirana polja (nastelama u obliku visoke edikule s dva registra kolj-ka je preferirana kao pozadina gornjemu registru!) ionako kako je to uinjeno na steli iz ovoga rada.

Druga je grupa zastupljena stelama kod kojih jekoljka postavljena obrnuto u odnosu na prvu gru-pu, tj. srce je okrenuto gore, a iri dio dolje. U tugrupu ide samo jedan arhitektonski koncipirani pri-

mjer fragment gornjeg dijela stele s figurama bra-nog para iz Ekvuma (Cambi 1989: 42 i d., sl. 5).

Iako zastupljene sa svega dva-tri primjera, stele sa

lar frame (arched gable), and its external edge was ren-

dered in the form of a simple border additionally deco-

rated with an engraved wave-motif. The other stela orig-

inally had somewhat smaller dimensions, and it comes

from Narona (Rinaldi Tufi 1971: 107, no. 28, tab. IX,

fig. 1; Cambi 1980: 135, fig. 9). The shell is affixed to a

triangular field with the form of a pseudo-gable, so thatthe ribs of the shell rise toward its sides. It is certain that

the latter stela had no architectural decoration, while it

is almost certain that this is the case for the former.

e fragment of a Salona stela with the head of ayounger man that is held in the Archaeological Mu-seum in Split (Rinaldi Tufi 1971: 94, no. 5, tab. III,fig. 2) could possibly be counted among the afore-mentioned stelae. Here the shell is also inserted intoa field with a semi-circular end and an edge ren-dered in a straight border. e shell only contained

the figure of a younger man probably shown inhalf-size. is is clearly evidenced by the ribs of theshell which merge in the heart precisely above thehead of the deceased. e problem of this fragmentis associated with the possibility of an architecturalframe under the arch ending (thus, as its support)or possibly in the lower portion of the monument.

In all other examples, the shell is exclusively used asa background to the deceased and it always architec-turally framed so that it is flanked by small columnsor pilasters, while its upper portion breaks throughthe horizontal side of the gable. Based on the posi-

tion of the shell, this most numerous group of monu-ments can be divided into the following two groups.

In the first group the shell is arranged so that itsbroader end is turned upward, while the narrowerend (heart) is turned downward, so that the ribs ex-tend radially outward from the bottom. Here the twomiddle ribs normally pass exactly between the headsof the deceased. Such an arrangement for the shellwithin the pictorial field is present in the destroyedSalona stela discovered at the western necropolis in1826 (Lanza 1856: 33, tab. VII), on the stela of M.

Antonius Celerus preserved in the drawing by Fi-liberto Pingone (CIL 3: 3162a; Hofmann 1905: 49f.),the stela of Marcus Pytha, a soldier in the secondcohort of Cyrestus from Gornji Rakiani at Gardun(CIL 3: 14934; Prijatelj 1952: 143; Rinaldi Tufi 1971:98, no. 12, tab. V, fig. 2; Cambi 2000: 49, no. 61, tab.80) and on a fragment of the stela from Salona bear-ing the head of an elderly man (Rinaldi Tufi 1971:95, no. 7, tab. III, fig. 3; Cambi 1987: 146, no. 46 withfig.; Cambi 2000: 34, no. 23, tab. 19). In all of theseexamples, except the last, the position of the shellis clear with regard to the structural composition of

the monument and the number of deceased personsportrayed. Although broken on all sides, there isno doubt that this fragment belongs to an edicule-


18/36

128


koljkom ije je srce okrenuto prema gore iznimnosu vane za pitanje podrijetla toga motiva. Inspiraci-

ju za takvo ralanjivanje slikovnoga polja zasigurnotreba traiti u arhitekturi lararija i nimfeja, a najvea

je vjerojatnost da se ona na nadgrobnim spomeni-cima poela upotrebljavati zbog imitiranja naina

postavljanja statua u niama slubenih i privatnihgraevina (Gabelmann 1972: 99, s lit. u bilj. 97 i 98).Na to upuuje i veza koljke s polukrunim poljem(u funkciji zabata) posvjedoena na fragmentu steles jugoistone salonitanske nekropole, a vjerojatno ina fragmentu s glavom mlaega mukarca iz Arhe-olokog muzeja u Splitu koja je mnogo puta potvr-ena u rimskoj arhitekturi (Bratschkova 1938: 30 id.). Uz koljku kao kalotu svoda nie-slikovnog polja(to se iskazuje prisutnou vijenca na donjem kra-

ju koljke) takvu vezu izravno potvruje i fragment

stele iz Ekvuma.Zbog naina izrade i postavljanja koljke te zbogpruanja rebara iz srca iza lea pokojnika naa sestela nedvosmisleno moe povezati s prvom gru-pom stela kod kojih je koljka arhitektonski uokvi-rena i iskljuivo u funkciji pozadine. Takve su ste-le najbrojnije na podruju Salone. Uz onu koja jetema ovoga razmatranja, jo su ak etiri iz Salone,a samo jedna iz Tilurija. Ako se u statistiku ana-lizu uzmu svi ostali primjerci, vidjet emo da ih jeak est iz Salone, a po jedan iz Tilurija, Ekvuma iNarone. Jasno je kako takva slika mora imati veze s

radionicom u kojoj je nastao opisani uzorak. S obzi-rom na to da je stela iz Tilurija izolirani primjerak,a one iz Ekvuma i Narone ne samo pojedinani vei vrlo kasni primjerci, oito je kako samo salonitan-ske stele predstavljaju homogenu, kontinuiranu imanje-vie tipiziranu (nepromjenjivu) seriju spo-menika. Zato ne treba dvojiti da je rije o modelukoji je potekao iz klesarskih radionica Salone, alitako da je i u ovom sluaju preuzeta shema koja jeotprilike u isto vrijeme kolala prostorima sjeverneItalije, ponajprije istonim Venetom i Altinom (usp.

Pflug 1989: 120 i tamo navedene primjere, posebnotab. 20, sl. 2). U salonitanskim je radionicama tashema dominantan sepulkralni izraz neposrednoprije i iza sredine 1. stoljea, kada i nastaju najre-prezentativniji primjerci. Moe se ustvrditi da salo-nitanske stele raene po tom modelu predstavljaju inajljepe primjerke takve vrste nadgrobne plastike.U flavijevsko doba model arhitektonske stele bez ar-hitrava s portretnom koljkom preuzet e, ini se, itilurijske radionice, ali nee nikad zaivjeti u unu-tranjosti provincije.6Zanimljivo je i znakovito dana podruju Tilurija nije pronaen ni jedan slino

izraen spomenik koji bi pripadao vojniku 7. legi-je, a isti je sluaj i s prostorom sjeverne Dalmacije(Liburnije). Prema tome, u dalmatinskom priobalju

form stela with no architrave. e images of two de-ceased (surely a married couple) stood in the picto-rial field, but only the head of an older man has beenpreserved. is clearly shows the position of thetwo central ribs of the shell to the left of the manshead. ey extend outward from the heart, located

approximately behind the head of the deceased.How many figures were shown in all will remainunknown, because they may have been arranged ina single field, in two superimposed fields (in highedicule-form stelae with two registers, the shell ispreferred as the background in the upper register!)and as rendered in the stela considered in this work.

e second group encompasses stelae in whichthe shell is placed in a manner opposite to the firstgroup, i.e. the heart turned upward, while it expandsdownward. is group includes only one archi-

tecturally conceived example: the fragment of theupper stela with figures of a married couple fromEquum (Cambi 1989: 42f., fig. 5).

Although present in only two to three examples,the stelae with shells wherein the heart is turnedupward are extremely important to the question ofthe origin of this motif. e inspiration for such apictorial field so arranged should be sought in thearchitecture of lararia and nymphea, while it ismost likely that it began to be used in grave monu-ments to imitate the manner of placing statues inthe niches of official and private buildings (Gabel-

mann 1972: 99, with bibliography in notes 97 and98). is is indicated by the link between the shelland the semi-circular field (to function as a gable)proven to exist on the fragment of a stela from thesoutheastern Salona necropolis, and probably onthe fragment bearing the head of a younger manfrom the Archaeological Museum in Split which hasbeen confirmed many times in Roman architecture(Bratschkova 1938: 30f.). Together with the shell asa section of the arch of a niche/pictorial field (shownby the presence of a wreath on the lower portionof the shell) such a link is directly confirmed by thestela fragment from Equum.

Based on its crafting and the placement of the shell,and on the extension of the ribs from the heart be-hind the backs of the deceased, our stela can unmis-takably be associated with the first group of stelae,wherein the shell is architecturally framed and ex-clusively a function of the background. Such stelaeare the most numerous in the territory of Salona. Inaddition to the stela under consideration here, thereare four more from Salona, and only one from Tilu-rium. A statistical analysis of all other stela shows

that there are six from Salona, and one each fromTilurium, Equum and Narona. It is clear that thissituation must have something to do with the work-


19/36

129


opisani sepulkralni izraz u prvom je redu karakteri-stian za salonitanske klesarske radionice i njihovudjelatnost iz sredine 1. st.

2. Manja portretna stela iz Muzeja grada Trogiraima slian raspored polja na prednjoj plohi ploekao i ona salonitanske provenijencije; u donjem

je dijelu natpis, sredinu je zauzelo slikovno polje sportretom pokojnice, gornji dio spomenika nije sa-uvan, ali se o njegovu karakteru moe raspravljati.Iza tako slinog rasporeda kriju se, meutim, obli-kovno bitno razliita rjeenja (sl. 8).

Natpisno polje druge stele izraeno je na najjedno-stavniji mogui nain. Na ravnoj plohi uklesan je

shop in which this sample emerged. Since the stelafrom Tilurium is an isolated example, while thosefrom Equum and Narona are not only individualbut also very late examples, it is apparent that onlythe Salona stela are a homogenous, continual andmore or less type-specific (unalterable) series of

monuments. is is why there should be no doubtthat this is a model that emerged in the stone-ma-son workshops of Salona, although in this case alayout was taken which was present in northernItaly at about the same time, predominantly in east-ern Veneto and Altino (cf. Pflug 1989: 120 and theexamples cited there, especially tab. 20, fig. 2). InSalona workshops, this scheme was the dominantsepulchral expression prior to and after the mid-first century AD, when the most typical examplesappeared. e Salona stela made in compliance

with this model can be said to represent the finestexamples of this type of grave sculpture. During theFlavian Era the architectural stela model withoutarchitraves and featuring a portrait shell would beassumed, it appears, by the Tilurium workshops,but it never took root in the provinces interior.6Itis interesting and significant that no other, similarlycrafted monument was found in the territory of Til-urium that could belong to a soldier of the seventhlegion, and the same is true of northern Dalmatia(Liburnia). us, in the Dalmatian coastal belt thissepulchral expression was primarily characteristic

of the Salona stone-mason workshops and their ac-tivities in the mid-first century.

2. e small portrait stela from the Trogir City Mu-seum has a field arrangement in the foregroundsurface similar to those of Salona provenance; thereis an inscription in the lower field, the mid-sectioncontains a pictorial field, while the upper portion ofthe monument has not been preserved, even thoughits character can be analyzed. However, this similararrangement conceals an essentially different solu-tion in terms of form (Fig. 8).

e inscription field of the second stela was madein the simplest possible manner. e inscription isengraved on a smooth, flat surface with no borderprofiles, while its rendering greatly differs from theinscription on the first stela which was made like aclassical titulus.

e pictorial field is inserted between the inscrip-tion and the lost upper portion (battlement), whichis customary for monumental and architecturallyconceived stela. However, it does not have an archi-

Slika 8.Donja polovica stele starice Marcelle iz Muzeja gradaTrogira (snimio D. Mari).

Figure 8.Lower half of the stela of the old woman Marcella fromthe Trogir City Museum (photo by D. Mari).

6

Na tom prostoru, koliko je meni poznato, nije pronaena nije-dna stela s portretnom koljkom. Najblii primjer predstavljastela iz Bosanskog Novog kod koje su ulogu koljke preuzelarairena krila orla: Sergejevski 1939: 10 i d., br. 1, sl. 2.

6

To this authors knowledge, not one stela with a portrait shellwas found in this area. e closest example is a stela fromBosanski Novi, where the role of the shell is assumed by theoutspread wings of an eagle: Sergejevski 1939: 10f., no. 1, fig. 2.


20/36

130


natpis koji nema obrubnu profilaciju te se izvedbombitno razlikuje od natpisa prve stele koji je izraen uformi klasinog titulusa.

Slikovno je polje umetnuto izmeu natpisa i izgu-bljenoga gornjeg dijela (krunita), to je uobiajenokod monumentalnih i arhitektonski koncipiranihstela. Meutim, ono nema arhitektonski karakter,nego je izdubljeno u obliku plitke nie u koju jesmjeten lik pokojnice. injenica da je gornji dionie odlomljen zajedno s vrhom oteavajua je, aline i presudna okolnost za moguu rekonstrukciju.Naime, sauvani su poeci rubnih stranica polja, sastrana pokojniina poprsja, koji jasno ukazuju daono nije moglo imati trokutasti oblik (za primjeretakvog tipa nie usp. Rinaldi Tufi 1971: 106107, br.27, tab. VIII, sl. 4, 108, br. 28, tab. IX, sl. 1), ili, re-cimo oblik pseudoklipeusa (Cambi 1990: 111 i d.).

Meu cijelim korpusom rimskih portretnih stela sobalnog prostora provincije Dalmacije nema niti je-dnog primjera gdje je slikovno polje, od natpisnogpolja odvojeno jedino svojom visinom ili eventual-no urezanom linijom, imalo neki drugi oblik osimpolukrunog ili priblino potkovastog. To pokazu-

ju ne samo one portretne stele koje su izraene upriblino isto vrijeme kad i trogirska, kao npr. stelaJulije Valerije (Prijatelj 1952: 148, tab. XIII; Rinal-di Tufi 1971: 102103, br. 19, tab. VI, sl. 3; Cambi2000: 84, br. 138, tab. 179), Aurelija Valerina (Ren-di-Mioevi 1957: 156 i d., sl. 1; Rinaldi Tufi 1971:

102, br. 18, tab. V, sl. 5; Cambi 2000: 83, br. 136, tab.177) i stenografa Asterisa (Rinaldi Tufi 1971: 112 id., br. 41, tab. XI, sl. 1), nego i one koje su u to dobaprepravljene od starijih spomenika, meu kojimatreba istaknuti stelu iz vrta Garagnin-Fanfogna uTrogiru (Abrami 1937: 18, sl. 12) i onu uzidanu u

juni zid Crkve sv. Duha u Splitu (Cambi 1987: 7 id., sa slikom).

Razlog zbog kojeg je takav oblik slikovnoga polja bioprimjeren jednostavno ralanjenim stelama vrlo

je prozaian. Nia koja je zasvoena ili natkriljenalukom najbolje prati liniju tijela pokojnika, osobitoglave. Osnovni je pak razlog esto koritenje takvihnia u samoj arhitekturi, takoer kao prostora pre-dodreenog za smjetaj skulptura. Iz arhitekture je

vjerojatno i doao impuls za njezino koritenje unadgrobnoj umjetnosti. Mogli bismo prema tomekazati da je i druga stela nastala po uzoru na jedno-stavne polukrune nie iz stvarne arhitekture, alitakve nie koje su bez arhitektonske dekoracije, stu-pova, pilastara i slinih elemenata.

Prema oblikovnoj koncepciji sauvanoga dijela kor-pusa oito je kako manja trogirska stela pripada gru-

pi anarhitektonski koncipiranih stela. Najbrojniji tiptakvih stela u svakom kutku Carstva jesu stele kojenjemaki autori pojmovno odreuju kao Profilge-

tectural character, rather it is concave in the form ofa shallow niche in which the figure of the deceased isplaced. e fact that the upper portion of the nicheis broken off together with the top is a complicating,but not crucial factor influencing a possible recon-struction. Namely, the beginnings of the fields bor-

der side have been preserved from the side of thebust of the deceased, which clearly indicates that itcould not have had a triangular form (for examplesof this type of niche, cf. Rinaldi Tufi 1971: 106107,no. 27, tab. VIII, fig. 4, 108, no. 28, tab. IX, fig. 1),or, say, the form of a pseudo-clipeus (Cambi 1990:111f.). Among the entire body of Roman portraitstelae from the coastal belt of the Dalmatian prov-ince, there is not one example in which the pictorialfieldwhether divided from the inscription fieldeither by its own height or possibly by an engraved

linehad a form other than semi-circular or ap-proximately horseshoe-like. is is shown not onlyby the portrait stela made at nearly the same time asthe Trogir examples, such as the stela of Julia Vale-ria (Prijatelj 1952: 148, tab. XIII; Rinaldi Tufi 1971:102103, no. 19, tab. VI, fig. 3; Cambi 2000: 84, no.138, tab. 179), Aurelius Valerinus (Rendi-Mioevi1957: 156f., fig. 1; Rinaldi Tufi 1971: 102, no. 18, tab.V, fig. 5; Cambi 2000: 83, no. 136, tab. 177) and thestenograph Asteris (Rinaldi Tufi 1971: 112f., no. 41,tab. XI, fig. 1), but also by those which were madefrom remodeled older monuments, among them the

stela from the Garagnin-Fanfogna gardens in Trogir(Abrami 1937: 18, fig. 12) and the stela mortaredinto the southern wall of the Church of the HolySpirit in Split (Cambi 1987: 7f., with picture).

e reason why this pictorial field form was suit-able for simply arranged stelae is very prosaic. Aniche covered by an arch best follows the line of thebody of the deceased, especially the head. e ba-sic reason is the frequent use of such niches in ac-tual architecture, also as a space designated to placesculptures. e idea to use it in grave art probablyderived from architecture. We could therefore saythat the second stela emerged based on the modelof simple semi-circular niches from actual architec-ture, although niches without architectural decora-tion, columns, pilasters and similar elements.

According to the formative concept of the preservedportion of the corpus, it is obvious that the smallTrogir stela belongs to the group of non-architec-turally conceived stelae. e most numerous type ofsuch stelae in every corner of the Empire are thosestelae which German authors have terminologicallydesignated as Profilgerahmtenstelen, i.e. profiled

stelae (e.g. Gabelmann 1972: 67f.), while Italian au-thors call them stele semplicemente riquadrate(Mansuelli 1956: 368f.) or stele corniciate (Rebec-


21/36

131


rahmtenstelen, tj. profilirane stele (npr. Gabelmann1972: 67 i d.), a talijanski kao stele semplicementeriquadrate (Mansuelli 1956: 368 i d.) ili stele cor-niciate (Rebecchi 1972: 181 i d., osobito 189 i d.).One najee imaju neke elemente arhitektonskogapodrijetla kao to su zabat i akroteriji, ali je njihov

najvaniji strukturalni element zapravo profilacijakoja uokviruje cijelo natpisno polje ili barem jedannjegov dio.

Najvei broj profiliranih stela je zabatne forme. Umanjem broju sluajeva to je slobodnostojei (pla-stini) trokutasti zabat, sa strana flankiran akroteri-

jima, a u neto veem broju zabat upisan u pravoku-tni zavretak spomenika s profilacijom naznaenimpseudoakroterijima ili jednostavnim trokutastimpoljima ostavljenima u funkciji pseudoakroterija.I kod jednih i kod drugih slikovno polje moe biti

uklopljeno izmeu zabata i natpisa, unutar zabata ivrlo rijetko unutar natpisnoga polja.

Mala trogirska stela nema najvaniji element profili-ranih stela profilaciju. Nema je ni uokolo natpisa,to bi se moglo oekivati, ni uokolo nie, to bi biomanje logian izbor, a nije ju realno traiti ni u izgu-bljenom dijelu spomenika. Slino oblikovan prednjidio ploe imaju gotovo identine stele Julije Valerijei Aurelija Valerina. Prvu odlikuje slojevita unutarnjaralamba svojstvena arhitektonskim stelama. Donjidio s natpisom ima svojevrsnu funkciju potpornjapolukruno zasvoenom slikovnom polju koje je s

gornje strane obrubljeno uskom ravnom trakom tose produuje na zabat. Prostori koji su ostali izme-u polja i horizontalne stranice zabata ispunjenisu stiliziranim vegetabilnim motivima. Kod drugestele luk slikovnoga polja nema rubne trake. Arhite-ktonizacija tih dviju stela manifestira se kroz pojavuslobodnih trokutastih zabata i polukrunih akrote-rija (Rinaldi Tufi 1971: 125 i d.). Asterisova je stelakoncipirana neto druki

Documents

5 Drazen Marsic