w University of Washington LibrariesInterlibrary Loan and Document Delivery ServicesBox 352900 - Seattle, WA 98195-2900(206) 543-1878interlib@uw.eduWAU 1WAUWAS 1 RAPID:WAU

2/26/2019 11:05:58 AM

LendingScanRAPID

ILLiad TN: 1731091

111111111111111111111111111111111111111111111

ILL Number: -14378149

1111111111111111111111111111111111111111111111111111111

Location: Suzzallo and Allen Libraries- StacksCall #: DE2 .V42 v.82 pt.1Work Order Location: N

Borrower: RAPID:AFUOdyssey:206.107.44.64Billing Category:Needed By:Maximum Cost:

Journal Title: Babesch, bulletin antieke beschavingVolume: 82Issue: 1MonthNear: 2007Pages: 165-181Article Author: Jansen, Gemma C. M.Article Title: Toilets with a view: the luxurious toiletsof the emperor Hadrian at his villa near TivoliISSN: 0165-9367OCLC #: 3901403

NOTICE OF COPYRIGHTThe document above is being supplied to you in accordance withUnited States Copyright Law (Tille 17 US Code). It is intended onlyfor your personal research or instructional use.

The document may not be posted to the web or retransmitted inelectronic form.

Distribution or copying in any form requires the prior express writtenconsent of the copyright owner and payment of royalties.Infringement of copyright law may subject the violator to civil fineandlor criminal prosecution.

Special Instructions:Notes/AlternateDelivery:

Email:EMAIL:

Odyssey

Toiletswith a ViewThe Luxurious Toilets of the Emperor Hadrianat his Villa near Tivoli

Gemma Jansen

Abstract

An investigation of the many toilets at the villa of the Roman emperor Hadrian, near Tivoli (Italy), both multi-seat and single-seat varieties, shows that the emperor preferred to have some privacy in his toilet. The multi-seat latrines found there were for servants. For Hadrian himself his family and especially his guests, the emperorbUIlt toilets with space for a single person. These were vast rooms with high ceilings and large windows pro-viding light. They were equipped with sophisticated water systems and very luxurious decorations. A specialfeature of some of these toilets with large windows was a spectacular view. All toilets are different in shape anddecoration, and many have no equal elsewhere in the Roman Empire. It is well-known that the emperor Hadrianloved to entertain and surprise his guests. It looks like his toilets were part of the surprise too.1

From 1999 onwards, the author has studied - to-gether with a small team of colleagues and stu-dents - the many toilets at Hadrian's villa (jig. 1).This site presents an extremely valuable assemblyof thirty-five toilets. They are not only well pre-served, but they show a cross-section of toiletsthroughout the entire Roman society. Here is theopportunity to study the toilets of the lowest ser-

vant up to the toilet of the most powerful man ofthe Empire: Hadrian himself. AlI these thingsmake this site unique and allow us to study someissues impossible to investigate elsewhere: Forwhich social group was a particular toilet built?and: How much privacy was available for eachsocial group?'

0'\ 1071 ",,,

11""l\.'",,,,,E 1""

= ,- EUI Valley

.:/rue

'), ,'~

alley

\'"

/" 100

Fig. 1. Sketch-plan of Hadrian's villanear Tivoli (from MacDonald/Pinto

1995, reproduced by permission of YaleUniversity Press, copyright!,

165

Fig. 2. Public toilet with marble seats, front platesand sponge gutter near the Forum Baths at Ostia.

Fig. 3. Flush. toilet with raised tiled floor in thekitchen of the House of Apollo at Pompeii.

Furthermore, as the toilets are so well pre-served, they also provide answers to more basicquestions about water systems and decorationschemes. Especially since most of the toilets havebeen cleared of the debris on their floors, there isa lot of new information available on their watersupply and drainage systems. Besides, the drainsappeared to be full of parts of the original toiletdecoration: pieces of coloured marble, mosaicsand fragments of wall plaster. There was only onedrain, belonging to the toilet of the Smaller Baths(30), which still contained the original Romandeposit. For the many finds in this drain, repre-senting unusual luxurious decoration, see thearticle by Alessandro Blanco (in this volume pp.183-190). Because of the organic nature of thisdeposit, samples were taken and analyzed. Theresults revealed that no seeds, fruits or pollenwere preserved in the samples. Even the chrysalisof flies are absent, a very common feature in fecalsurroundings.s

166

Fig. 4. Niche toilet on upper floor of House V 1, 30at Pompeii.

This article, then, presents new information onthe water systems and the decoration of the toiletsof the villa. To put the toilets of Hadrian's villa inperspective, however, a short presentation of toi-lets elsewhere in Roman Italy is needed.

ROMAN TOILETS IN POMPEII, HERCULANEUM AND

OSTIA IIII

i

Since Scobie's inspiring publication on Romansanitation, published in 1986, modern scholarshave re-investigated Roman toilets and toiletbehavior. Several researchers have published onthe issue, though mainly focusing on Italy. Fromtheir work some general conclusions can bedrawn on toilets in this central part of the Empire'Roman toilets resemble modern western toilets

in many ways. The Romans, for example, preferredto sit down; therefore, their toilets are equippedwith seats. The Roman seats, however, have a dif-ferent design, as the Romans did not use toilet

paper; instead, they used a sponge on a stick.There was a hole in the front plate of the toiletseat in order to use these sponges (Jig. 2). Theother difference is that Romans were not alwaysalone on a toilet. Besides the many one-persontoilets in private houses, two-seaters and three-seaters also occur. And public toilets were nearlyalways group toilets, so-called multi-seaters. As thesmall private and the larger public toilets differboth in water technique and in decoration, theywill be dealt with separately.All Roman toilets needed water: both to rinse

the sponge and to flush the toilet after use. There-fore group toilets had gutters in front of the seats,filled with running water to wash the sponges. Tocreate an efficient and continuously flowing stream,this water could only be pipe water. The overflowof the sponge gutter was directed in the deep fecesgutter underneath the seats and together withsewage was led into the city sewer system. It isnot clear whether these feces gutters underneaththe seats were flushed additionally with an extrawater pipe. Recent investigations of Van Vaeren-bergh into the toilets belonging to baths showthat the flushing system was not always workingsmoothly. This is remarkable, as toilets near bathsprofit from the best circumstances: good nearbysewers and enough flushing water at hand, eitherfrom the swimming pool, the bathtubs or theoverflow from the many fountains. Private toiletsare much smaller and seldom have a sponge gut-ter. Sponges were rinsed in small basins or jars.Toilets Were flushed manually with a bucket ofwater. This led to a different design: the front plateis missing so that the toilet could be flushed.The toilets of Hadrian's villa can be compared

best with the toilets of the Roman harbor townOstia. Most of these are built in the 2nd centuryand thus provide the right timeframe with whichto contrast Hadrian's own toilets. In order to geta good chronological overview of all the possibleexamples of toilets, however, it is good to have aquick look first at the toilets found at Pompeiiand Herculaneum, which were in use more thana century earlier.

The slim evidence of only ten public group toi-lets from Pompeii and Herculaneum shows onlya glimpse of the appearance of the toilets in use inthe 1'( century Be and AD.' Based on the eviden-ce available, we can state that the public toiletswere simple in design. Small high-positioned win-dows gave the toilets a dark appearance. Koloski-Ostrow calls them 'grim'. To compensate darknessthe walls were painted white, mostly with a darkdado to hide stains. The toilets were provided with

running water for the sponge gutters (nearly allexecuted in masonry - no marble) and for flushingthe gutter underneath the seats. When connectedto the baths, they were adclitionally flushed withwater from the baths or swimming pools. All ofthese public group toilets were connected to asewer that brought the toilet waste out of the city.

Private toilets can be found in great abundancein Pompeii and Herculaneum. From the 200 toiletsinvestigated by the author in Pompeii and themore than sixty toilets in Herculaneum, it can beconcluded that every household must have hadat least one toilet. This is confirmed by the recentinvestigation of all the toilets of region V at Pom-peii by Hobson. He also discovered that privatetoilets are part of the houses from the very begin-ning of Pompeian urbanization. Most of the pri-vate toilets are single-seaters, though two-seatersor three-seaters occur. They tend to be dark roomswith tiny windows (if any windows at all). Theonly decorations visible are red lines on whiteplastered walls and, in a few cases, painted float-ing ornaments in between (dolphins, garlandsetc.). The toilets occur frequently in kitchens, understairs and upstairs in rented apartments. In mostcases, two brick walls or a slot can be seen to sup-port the wooden seat (now gone). There was nosponge gutter, but a basin to clean the sponges. Tofill these basins, the Pompeians seldom had run-ning water; they had to fill the basin with waterfrom the cistern in the courtyard or a street foun-tain. Besides, to flush the toilet, they had to use abucket of water, filled at the same water source.Households with good water supplies installedflush toilets, which mostly occur in kitchen areas.To facilitate flushing, a tiled chute was built in frontof the toilet seat. The chute slanted slightly towardsthe drain underneath. The fact that these toiletswere also provided with footrests shows that it wasapparently not unusual to throw water into thedrain while somebody was using the toilet (jig. 3).Households where water was more scarce, forexample the rented apartments upstairs, had toiletsbuilt in niches in the wall. Thus one could sit rightabove the drainpipe, which was led into the wall.In these cases not much water was required forgood flushing (jig. 4). Only few private toilets wereconnected to a sewer, most emptied into cesspits.To get an idea of what the toilets in Hadrianic

times looked like, we have to turn to Ostia. As thisbusy port town with a high population density, iscompletely different from the relatively quiet res-idential town Pompeii, it is not a surprise thathere more public toilets can be found: twenty-seven, of which at least twenty date from the 2nd

167

century. The most famous toilet, the one near theForum Baths !fig. 2), dates from the 4th century.Only 2nd-century toilets will be taken into accounthere.' The water supply and drainage systems ofthese toilets very much resemble those of thePompeian examples. The only difference is thathere, in addition to the sponge gutter, an extrawater basin or labrum is added in many of the toi-lets? The multi-seaters in the tenement houseshave the same somber aspect as the toilets of Pom-peii. They remain dark due to the lack of largewindows. They have only a modest wall decora-tion in the form of a simple flower motif, floorsare laid out in black and white mosaics without apattern, and sponge gutters are made of masonry,or sometimes of travertine or marble. The multi-seaters found in more public areas are somewhatdifferent, mostly with regard to decoration. Especi-ally, the toilets that are connected to baths are dec-orated with more detail. The windows in the upperparts of the walls are large and bring in a lot oflight. The wall decoration is still very plain: thesame white plaster with flower motifs or yellowand red squares. But the floors are much fancier:varying from simple black and white mosaics,decorative mosaics, small coloured marble stonesto a floor of white marble slabs. Many toilets havewhite marble sponge gutters and three 2nd-cen-tury toilets still contain marble seats.

Neudecker calls some of these toilets 'Pracht-latrinen'. From the end of the 1" century onwardsthis new phenomenon is adopted in the RomanEmpire: vast latrines in which a lot of light comesin through large windows or peristyles. They areindeed large: average seat capacity is for aboutthirty-five persons with an absolute maximum ofeighty.' These toilets are executed completely inmarble (floor, gutter, wall, and seats) and are ador-ned further with mosaics, statues and spoutingfountains. The large amount of flowing water fromthe fountains provided good flushing capacityand the vast windows guaranteed fresh air. Al-though this would have diminished the bad smell,clients were not meant to loiter in the toilets. Ac-cording to Neudecker, the decorations were in-tended to alienate one from the very fact that onewas supposed to do on a toilet: a dirty act is cam-ouflaged by shining marble and sparkling foun-tains. The three toilets of Ostia that fit into his cat-egory of 'Prachtlatrinen' are the most modest ofNeudecker's list.' 'Prachtlatrinen' do not occur sooften in Italy, anyway. The most elegant and daz-zling examples are to be found in the Near East(twenty examples) and North Africa (twenty-eight examples).

168

The more than seventy private toilets of Ostiadate mostly from the 2nd century. They have beenpreserved on ground floors only and mostly arein bad shape.t? As far as can be detected they aremost charming in their simplicity: a board with ahole fixed in a slot in the walls of a small room.t!Some had footrests. Underneath was a hole orgutter connected to a drain. These toilets hardlydiffer from the small and dark private toilets ofPompeii: many are found at the back of shops ina small space underneath staircases. No decoration(no wall paintings, no marble) has been found inthem except for one: the toilet of the Casa delFortuna Annonaria has a beautiful floor in opussectile,'?

All toilets must have been connected eventual-ly to the city sewer system, which in Ostia is laidout very systematically. How these private toiletswere flushed is still unclear as the long tile chutesthat are in front of the Pompeian toilets are miss-ing. Some toilets have an imprint of a lead pipe,but it is not clear whether this pipe was meant toflush the toilet or to fill a missing sponge basin.Four toilets have preserved their sponge basins, allwith an imprint of a supplying lead pipe. In thisrespect, the toilet of the Casa del Fortuna Annona-ria forms an exception too: the basin there wasfilled automatically with the effluence from thenymphaeum fountain in the adjacent room.In short, in the 2nd century AD at Ostia toilets

were standard commodities. Public group toiletswere spacious, well lit, Simply decorated andequipped with a rather good water supply anddrainage system, whereas the private toilets weresmall and dark. They had no decoration, seatswere made of wood and their water supply anddrainage systems were executed with less sophis-tication.

TOILETS AT HADRlAN'S VILLA

Just after becoming emperor in 117AD, Hadrianstarted to build a new and vast palace, 120 hectarelarge, in the countryside about 30 km fromRome.P The architecture Hadrian choose for thispalace was very rich, new for his times and some-times even unique in the Roman world. When-ever he was not on his travels outside Italy, hestayed either in Rome or here. He was never aloneat Tivoli, but always surrounded by 1000 peopleor more, among whom were his family, friends,the imperial administration, his other personnel,slaves ... and many visitors. The many diningrooms present at the villa suggest that Hadrianseemed to love banqueting and entertaining his

fig. 5. front plate of the round toilet at the EastTerrace (6) (photo R. Kragiing).

guests. These dinning rooms range from largeareas where a lot of guests can be received (as forexample the large Water Court, 31) to smallerareas (for example the Scenic Triclinium. 40) andvery intimate rooms for a few private guests (asat the Island Enclosure, 17). These guests couldmarvel at the beautiful architecture, pieces of artand the large surrounding gardens filled withmarvelous water displays.For all those people toilets were provided. This

article deals only with the sanitary provisions thatare built within the palace architecture. Some san-itary practices, such as relieving oneself against awall or defecating in the wild, cannot be retrievedfrom the archaeological record; other practices arehard to track down, such as the use of chamberpots and wooden toilets. The later are mentionedin ancient Latin texts, but their use in the villa hasnot been established yet.!'

Dispersed over the villa terrain the author andher team have investigated ten multi-seaters andtwenty-five single-seaters. IS Most of these toiletsare mentioned by several researchers of the villaor have been excavated recently by the Soprinten-denza di Lazio.is The toilets of this large estate,however, have not been studied systematicallyuntil now. The multi-seaters are recognizable byremnants of seats above a deep gutter in front ofwhich a small sponge gutter was laid out. Thissponge gutter is not sunken in the toilet floor, butusually placed on top of it and serves at the sametime as a convenient footrest. None of the seats arestill preserved as the wooden ones have decayedand the stone ones have been robbed. What is leftare traces of the fastenings: slots in side or backwalls, small holes for little wooden or marblebeams, rests of the marble beams themselves,larger holes for stone supports, rests of travertine

supports or supporting walls. In several toilets acombination of the different methods of fasteningwere applied. The seats were once closed with afront plate as ledges or marble rests in the wallindicate.!? We were lucky to uncover a marblefront near the round toilet on the East Terrace (6)(jig. 5). Seating capacity varied from three to twentypeople. The rooms were vast and the few toiletwalls that are standing show large windowsabove shoulder height.

The single toilets are somewhat different andconsisted of a single marble seat above a drain. Inseven toilet rooms footrests also remain. The roomswere spacious, both with respect to surface areaand height. This might be due to the round archi-tectural design, so popular at the villa, which pro-vides a lot of space for these toilets and most ofthem can be found on either side in the rest area atthe back of the apses. A lot of light came in throughlarge windows, sometimes as large as door open-ings.

As it is possible to establish the specific func-tion of the different buildings of the villa, it is alsopossible to determine which of the social groupsof the villa (the emperor and his family, his guests,his staff and slaves) used or visited a specificbuilding and the adjoining toilet." In those partsof the villa identified as the private domain of theemperor, only single toilets occur. Examples in-clude the single toilets of the Island Enclosure (17)and the toilets in the Peristyle Pool Building (27).In the large areas where guests were entertained,clusters of Single toilets were constructed. Nearthe dining halls of the Water Court (31) six singleseat toilets were built; the dining hall of the ScenicTriclinium (40)offers three separate toilets. It doesnot come as a surprise that multi-seaters werelocated in the baths and in the service areas.Relying on these data we can conclude that dif-ferent toilets were built for people of differentsocial rank. The emperor and his guests usedmainly single-seaters. Staff and slaves only hadmulti-seaters at their disposal. The lavish decora-tion of the single-seaters and their excellent watersystems (both discussed below) underline thisconclusion.

WATER SUPPLY AND DRAINAGE

Like other Roman toilets elsewhere, the villa toi-lets needed water for rinsing sponges and flush-ing, and a drain connected to a sewer to carryaway urine and feces. There was plenty of waterat the villa site, as we know that Hadrian tappedone or two aqueducts passing his villa terrain on

169

Fig. 6. Multi-seater of the North Service Building(26) (photo R. Kragting).

Fig. 7. Start of the sponge gutter in the multi-seaterof the North Service Building (26) with hole far the

supplying tead pipe.

their way to Rome." Water played an eminent rolein the villa and its gardens. Salza Prina Ricottidescribes it thus: 'COSl la villa si riempi non 501-tanto di verde, ma si illumine di fontane monu-men tali, eli graziose fontanelle, eli caseate, eli ninfei,di canali grandi e piccoli, di ampi bacilli, geome-trici laghetti piacevoli da ammirare nella caluraestiva €, rnentre il minima soffio del vento facevafremere la loro superficie ed il continuo moto del-I'acqua scorreva negli euripi suddividendo edilluminando I'area, i giardini, da belli che erano,divenivano magici.'20 Besides, water was usedinside dining rooms to enliven architecture.

Right now, several international research groupsare working on the water systems and drainagein and around the villa.s- Their research, however,is severely handicapped by the fact that nearly alllead pipes have been robbed during the last fewcenturies.22 There is only indirect evidence forlead pipes: holes in the walls for leading the lead

170

Fig. 8. End of the sponge gutter of the multi-seaterof the North Service Building (26).

Fig. 9. Drain with an a cappuccina roof connectedto the feces gutter of the multi-seater of the Central

Service Building (34) (photo R. Kragting).

ba.

" d.

~I- c. ./

Fig. 10. Schematic presentation of the water supplyand drainage system of a multi-seater: a = hole forsupply pipe, b = sponge gutter, c = feces gutterunderneath the seats, d = round corner.

pipes through, the imprint of the pipes in subfloors, and cut away ridges left by the looters. Ateam of speleologists is mapping the villa sewersystem, which is not easy as well as that the sew-ers are hardly accessible:" they are relativelysmall (50em or lower) and most of them have notyet been excavated. In addition, the channels forwater supply and the ones for drainage do notseem well systemized. There are, for example, nodistribution points built at regularly intervals noris there a main sewer with branches. To get a clearand coherent picture of how the overall watersystem and drainage was planned and how itworked, we have to wait until the results of theinvestigations of the several teams have beenpublished. On a smaller scale, at the level of thetoilets themselves, however, it is possible to recon-struct the water and sewer systems."

Water systems at multi-seaters

Themulti-seater next to the North Service Building(26) illustrates the basic principles of the watersystem in these toilets best (jig. 6). There is a stag-gered entrance with a small nvmphueum in theporch. The toilet room itself has seats along twoSides. Water was brought in once with a lead pipeof which the hole (a 5 em) is still visible (jig. 7). Itfilled the shallow sponge gutter made of traver-tine in front of the seats. The overflow was ledinto the deep feces gutter (150 em) underneaththe seat (jig. 8). The gutter, partly hewn out of theunderlying tuff bedrock and partly built in mason-ry, was made waterproof with layers of OPl/Ssigni-num. The gutter is connected with a drain a cap-puccina (conical in form, made by two tiles) to thevilla's sewer network further downwards (fig. 9).In addition to wastewater from other (unknown)places, which was led into this gutter, the spill-overs and the cleaning water from the groundwas also led into it, aided by the slightly slopingfloors. Figure 10 presents all the different ele-ments working together to establish a good watersupply and drainage system.

At five other multi-seaters a hole or imprint ofa pipe has been found.25 Only at the west toilet ofthe Canal Block (35) another washbasin has beenfound, other toilets only had sponge gutters.26

The sponge gutters of other toilets are also shal-low with an average depth 9 em and an averagewidth of 12 em. This does not suggest a largestream of water flushing the gutter underneaththe seats. The question can be asked whether thesetoilets were flushed adequately. One would ex-pect the toilets to be built above the large villa

sewers, through which all the surplus water fromthe large ponds and water displays was drainedaway. This only appears to be the case at a fewtoilets." Therefore additional flushing was neededand for this reason the toilet of the North ServiceBuilding (26) was probably connected to a smallreservoir.'" Most toilets, however, form the begin-ning of a new sewer line, and it seems that theyhad to be flushed with one jet of water only.Especially the toilets that are connected to bathshave additional flushing. Although there is noproof for it, it seems highly plausible that the toi-let of the Service Quarters (8) was flushed by thesurplus water of the above lying pool of the EastWest Terrace (14).

To get rid of waste materials round drainswould serve best. From this point of view theround toilet at the East Terraces (6) and the semi-circular one of the Larger Baths (33) are very func-tional. The other rectangular toilets show roundededges to streamline the drain. Besides, all drainshad a smooth coating of opus signinum and a slopeof approximately 30 em. Studying the feces gut-ters revealed how the multi-seaters were built. Tocreate a drain deep enough, a large rectangle wasdug out of the underlying bedrock. On top of this,the upper part of the gutter walls were built. Mostgutters were therefore partly hewn in the rockand partly built. Tiles were placed at the bottom.The gutter was then completed with a waterprooflayer of opus signinum. The gutters are about 45 emwide and about 1 m deep.

The only exception in every sense is the toiletin the Service Quarters (8). As this is situated onthe first floor, it was impossible to build a deepfeces gutter underneath the seats: it is, indeed, notmuch deeper than the floor level, starting at bothends at 5 em underneath the floor. The gutter wasmade deeper by raising the sponge gutter and thefootrests in front of the toilet seats. In the cornerthere is a large drain in the form of a round holeleading to the floor underneath. Around the holethere still is a square impression, which might in-dicate a lead sheet connected to a lead drainpipe.The lead pipe (now gone) was probably con-nected to a sewer in the underlying room.

Water systems at single-seaters

In broad terms, the water system of Single toiletsworked according to the same principle as that ofthe multi-seaters. However, their appearance wascompletely different. Instead of a sponge gutterthere was a sponge basin; instead of a feces gut-ter underneath the seats, there was a collection

171

Fig. 11. Single-seater in the west wing of the ScenicTriclinium (40) with remains of marble seat andimprint of opus sectile floor. The groove indicatesthe position of the water pipe (photo R. Kragting).

Fig. 12. Toilet room of the External Nymphaeum atthe Water Court (31) with collection box (left!, downspout for rainwater (right! and gray slates as floordecoratIOn (photo R. Kragting).

172

box with a connection to the sewer. The systemcan be illustrated best with the three toilets at theScenicTriclinium (40).In the dining area arou.ndthe Canopus where Hadrian must have en ter-tained his guests, two toilets were in the sid ebuildings and one behind the cupola. The toiletsin the side buildings are rather similar (jig. 11): avast room in which - at the back - there is a mar-ble toilet seat, more a throne than a toilet. On thewalls are the remains of marble decoration andthe imprint of an elegant pattern of marble piecescan be seen on the floor. Here is also a deep im-pression of a lead pipe preserved going in the di-rection of the toilet seat, stopping at about 30 ernin front. This implies that once a lead pipe fedsomething in front of the toilet seat: this must havebeen a small charming, bubbling fountain inwhichsponges could be cleaned. The toilet at the eastside has, though placed in a symmetrical position,something extra: footrests under which the over-flow of the little sponge basin is led. Both toiletsare connected to a collection box, which in turnwas connected to a drain. The toilets appeared tobe the start of a small sewer lines with a roof acappuccino, connecting with the large villa sewera little further away. The third toilet of this areacan be found near the cupola, it differs from theother two: here the toilet seat is built into a niche,the collection box is very shallow and connectedto a very small drain channel. A drain from a roofabove also ends into this channel. The water sup-ply system with a pipe leading to the basin infront of the seats is the same, however.The toilet in the External Nymphaeum at the

Water Court (31) gives more specific information

Fig. 13. Detail of the toilet floor in the ExternalNymphaeum at the Water Court (31) with leadwater pipe ending in collection box (top left) andsmall pipe next to metre-stick (photo R. Kragiing).

about the water system !Jig. 12). The toilet is builtin a niche of this rather large room. The collectionbox is connected to a drain wi th an a cappuccinoroof. In one of the corners a drain from abovecomes down. The floor decoration in gray schistand the skirting in yellow marble are partly pre-served. The lead pipe that flushed the toilet is stillin situ, together with a small rolled lead pipe, adiversion, which had to fill a sponge basin (jig. 13).The center of the basin was 60 em in front of theseat. The pipe had a diameter of 4.5 cm and thediversion 2 em.We discovered parts of a marble sponge basin

at only two toilets." At eleven others an imprintof a pipe or a real pipe was heading in the direc-tion of a sponge basin.w The basins are not farfrom the seats, mostly within a range of 30 to 40cm. The water pipes entered the room mostlythrough the door openings. In three toilets a di-version of pipes can be seen. This means that withthis stream of water several things at one timewere supplied. How the water pipes of the toiletsfit in the larger scheme can be seen very well onthe drawing of the floor of the Water Court (31)by Rakob, showing how from two central pointseach toilet gets water with a separate water line."This suggests that the taps that operated all thewaterlines were operated from these two centralplaces.Most single-seaters seemed to be flushed with

only one water jet. Maybe this was sufficient, as thecollection boxes are still very smooth today andhave a chute downwards (with a decline between3 and 15 ern) to the sewer !Jigs. 14, 15). The depthsof the boxes depended on the location of thesewer to which they had to be connected. Mostare 76 em deep, though some are rather shallow(10-25 em), The later ones are mostly found onupper floors, where one has to drain the toiletbetween the ceiling of the ground floor and thefloor of the first floor. Some special maneuveringwas needed also at the Island Enclosure (17),where the drains had to dive underneath thecanal. Not all toilets were connected to drains. Itis remarkable that some toilets seem to disposetheir material into gardens or down the cliffs ofthe valleys.P Maybe the pipes that once carriedthe waste of these toilets have since disappeared.

The drain from above in the toilet at the Exter-nal Nymphaeum at the Water Court (31) and inother toilets, and the find of a 'toilet' in the theatre,seduced Hidalgo and Leon to speculate about aflushing system with higher placed reservoirs.PThis is a rather interesting theory which would bevery sophisticated if true. A systematic check of

Fig. 14. Single-seater at the West Belvedere (37)with modern fence in the middle (photo A. Vela).

Fig. 15. Detail of the collection box at the toilet atthe West Belvedere (37) (photo A. Vela).

all these pipes and their provenance revealed thatrainwater from adjoining roofs was led into thesedown pipes and in the already existing drain.This water was not meant for flushing, as in thiscase the toilets could be flushed only during arain shower.

The principles of the water supply anddrainage systems are summarized in figure 16.These toilets were equipped with sophisticatedwater systems feeding charming marble basins,in which fresh water bubbled. The water systemdefinitely added to the comfort of these luxurioustoilets.

173

Fig. 16. Schematicpresentation of the watersupply and drainage systemsof a single-seater:a = imprint of lead supplypipe, b ~ imprint of thesponge basin, c = drain forthe overflow of the spongebasin underneath the footrests, d ~ collection boxunderneath the seat, e = rainwater pipe discharging incollection box, f = connectedsewer line.

d.

a.

Fig. 17. Smallfinds from thedrain of the westtoilet of the IslandEnclosure (17).

r-

Irrr...J

Fig. 18.Reconstructionof the opus sectilefloorin the two toilets of thewings of the Scenic'Iriclinium (40) byGuidobaldi (1994).

Fig. 19. Reconstructionof the opus sectilefloorin the toilet of the East

Belvedere (10) byGuidobaldi (1994).

FORM AND DECORATION

The robbing of art and decorations of the villamust have started shortly after the abandonmentof the site in the 5th century.That is the reason whyalmost all decorations are gone. Only tiny piecesshow us a glimpse of how beautiful and magnif-icent the villa must have been once. The walls

174

must have been decorated with frescos, marbleslabs or intarsia, the latter made of delicatelyshaped small pieces of coloured marbles set innatural slate fields to form floral or architecturaldesigns. Ceilings once were plastered with stuccodecorations. Floors were covered with colouredmarbles from all over the world, alternating withblack and white mosaic floors with floral or geo-metric patrons.Although toilet rooms did not escape the rob-

beries, there are enough clues left to reconstruct thedecoration of these rooms. For example the subfloors, and especially those that have been pro-tected by a roof, show the negative impressionsof opus sectile floor patterns. Plaster on the wallsshows the negative of the revetment with marbleslabs, while clamp holes in the walls indicate thepresence of wall decoration. Besides, during ourcleaning operation of toilet floors and drains wefound parts of decoration in many toilets: smallparts of wall plaster with colourful painting (red,blue, green, yellow), small mosaics cubes in allcolours, parts of many types of coloured marbleand other stones, and intarsia !jig. 17). Unfortun-ately, most of these finds do not reveal anythingof the decoration of the specific toilet room. Thefinds were dumped in the toilet drains - togetherwith modern material - probably during the sev-eral restorations." Alternatively, the finds seem tobe transported - after the collapse of the villa -from elsewhere (far or near) along the drains.wSome fills tell something about the history of thevilla after abandonment, for example the fill inthe toilet of the West Belvedere (37) with glazedshards from post-antiquity inhabitants (1500-1800).Only in two toilets we can state that the finds aredefinitely from the toilet rooms themselves.On the basis of all the gathered information it

can be concluded that the Single-seaters were ex-tremely rich in decoration. Toilets seats and frontplates that have survived are made of marble;remaining footrests are covered with marble slabs.Of two toilets a part of the floor survives: the Sta-dium Garden (28) toilet had a floor with smallmarble squares in red and white, the toilet of theExternal Nymphaeum at the Water Court (31)hadrectangular gray slates on the floor with a skirtingof yellow marble ljig. 12).30Of eight other toiletswe have imprint of various patterns in opus sec-tile in the sub floor: the imprints are either rectan-gles or diamonds ljigs. 11,18,19).Thresholds andskirting boards are preserved better. Seven toiletsstill have one or more marble (white, pink or yel-low) or slate slabs left. Toiletwalls were plasteredbut had no paintings, they show signs of marble

Fig. 20. View over the east valley from the single-seater at the West Belvedere (37) (photo H. Heuts).

revetment, though at no toilet walI are the actualmarble slabs still attached.

It is remarkable that most toilet seats are builtin niches in the roomY These niches offer a spe-cial place for decoration: the apse. One of the toi-lets in the Peristyle Pool Building (27) is knownto have a mosaic in this apse in the form of aoelum.ss Ceilings are mostly made in the form ofcross vaults or semi-domes covered with plainwhite stucco. Windows in these vaults light twotoilets. Besides all these luxurious decorations thesingle-seaters had yet another special item: a view.Eight toilets have large windows, sometimes aslarge as a doorway opening to a wide variety ofviews: to the artificial canal around the island, thevilla gardens, and the dramatic valIeys on eachside of the villa terrain. The views can always beenjoyed best from the toilet seat (fig. 20).39

In general, the multi-seaters are more soberlydecorated than the Single-seaters, although thereis some variety within this framework. Their floors

can range from opus signinum or opus spicatum, tocrude white mosaics or more elegant black andwhite mosaics to marble floors. The sponge gut-ter goes from masonry built and coated with opussigninum or plaster to travertine and marble ex-amples. In most cases the seats seem made ofwood, sometimes of heavier material like traver-tine or marble. Walls were covered with opus sign-inurn or plastered white without any paintings,except for the toilets of the Smaller Baths (30),where waII decoration with eniarsia can be recon-structed.w Most walIs are preserved only at sucha low height that it is not possible to concludeanything about windows and the amount of lightin these rooms. Only three toilets have (remainsof) windows that were placed high in the wall.They were placed too high to enjoy the surround-ings." From this, however, it can be concludedthat these toilets were well-lit.

The simplest toilets occur in the slave areas: thetoilet of the Service Quarters (8) and of the servicearea of the Heliocaminus Baths (20)42 The threemost beautiful toilets, alI with marble decorationsare the ones in the Smaller (30) and Larger Baths(34) and the round one at the East Terraces (6).Dueto their position they do not seem to have beendestined for slaves or servants, but for the guestsof the Emperor. Their lavish decorations alsopoint into this direction.

From this survey of the decorations in the toi-lets, we can conclude that especially the single-seaters were decorated extremely colourfully withshiny materials. This indicates that the toilet roomswere not isolated entities in the vilIa, but that theywere part of the overall villa conception, wherelavish and variegated decorations are the stan-dard. Some toilets had exactly the same decora-tions as the surrounding rooms, as for examplethe six toilets at the Water Court (31).

We can observe tha t no two toilets at the villagrounds are the same. Though some are placed ina symmetrical position and their ground plansmay be the same, furnishings and decorationsshow many smalI variations. One of the two toi-lets in the wings of the Scenic Triclinium (40), forinstance, has a footrest, whereas the other one doesnot. Besides, the toilets at the vill a are not onlyvariations of the common types: two are so excep-tional in form that they have no paralIels in theRoman Empire. They can be regarded as noveltiesof Hadrian or his architect. Their form is so dif-ferent that at first we had serious doubts whetherthese rooms were toilets at alI.43 Only after clean-ill!, their floors and drains did they prove to betoilets.

175

Fig. 21. Gilded mosaictesserae from the apse ofthe toilet in the FountainCourt East (12)(photo R. Kragting).

Fig. 22. Round group toilet with seven separate seatsat the East Terraces (6) (photo R. Kragiing).

One of them is the vast two-person toilet in theFountain Court East (12). The room has an areaof 12 square m and is 4 m high. A large windowis overlooking the adjoining garden with foun-tain. The toilet seats themselves were placed intwo niches, one in front of the another. Beneath theseats were the normal collection boxes, smoothen-ed with several layers of fine opus signinum. Infront of the seat some parts of marble spongebasins remain. Though floors, walls and ceilingwere decorated just like the other villa toilets, theniches of the seats, however, had a special adorn-ment. In the apse of the east toilet (the apse of thewest one is a modern reconstruction) many im-prints of mosaics can still be detected. The mosaicsthemselves have been discovered in the drain andon the toilet floor: eighty-eight turquoise tesseraeand eight gilded ones (fig. 21). This toilet, whichis matchless in the Roman world due to its twoopposing seats, is also unique because of its shin-ing decoration in the apses.The other exceptional toilet is the round group

toilet at the East Terraces (6) (fig. 22). This veryelegant toilet, decorated completely with marbleslabs, had a small fountain in the middle of theroom. Although there are a few other round grouptoilets in the Roman Empire.w this is the only onewith separate seats for each user: seven users saton seven individual seats. The round shape ofthis toilet echoes the continuous repetition of

176

round forms found everywhere in the archi tec-ture of the villa. More than in the other toilets, thevisitors must have felt surprised. In this sense itis a pity that we can not establish the user grou pof these two toilets: we do not know exactly whatfunction the Fountain Court East (12) had, norwhat other buildings surrounded the round toi-let. From their exclusiveness, however, we canconclude they must have been for the most priv-ileged visitors.

TOILETS IN USE

It would be naive to assume that all toilets dis-cussed here were built in one stroke and were inuse at the same time. At least seven toilets showalterations that provide clues for dating or thattell us something about their period of use. Mosttoilets seem part of the general building process ofHadrian.s? One room seems to have been pi armedas a toilet at first, but somewhere during the build-ing process its design was changed.'. Two moretoilets were built later than their surroundingbuildings and thus form no part of the originalplan.v It is remarkable that both are multi-seatersfor personnel.After Hadrian's death the villa was in imperial

hands for at least some eighty more years. This canbe deduced from the imperial portraits of allHadrian's successors that have been found on theterrain, as late as the reign of Septimius Severnsand his son Caracalla. We can conclude that thevilla had not yet been abandoned then from a fewbrick stamps dating from the beginning of the 4thcentury. This implies that the villa was in use formore than 300 years'S It seems only natural thatsome toilets have been altered as, for example,one of the doors of the eastern toilet of the Resi-dence (22)has been dosed.s? Two other toilets wereput out of use while the villa was still inhabited.50The finds in the drain of the Smaller Baths (30),

parts of terra cotta cooking pots, so-called casse-roles, from the early 3,d century AD and a strongcorroded coin that can be dated to the years 296-305AD, probably do not tell us anything about thelate use of this toilet, but instead give us Some-thing about the date of dismantling this toilet andthis part of the villa.v'

/

CONCLUSIONS

The people that built and designed the toilets ofPompeii and Ostia were severely restricted: spaceand money determined the appearance of theirtoilets. Hadrian, who was not limited in any way,

Tab.1. Multi-seaters of Pompeii, Ostia and Hadrian's villa.

Pompeii/Here. Ostia (2nd century) Hadrian's villa

rene- ment houses baths simple luxurious

10 examples 12 examples 8 examples 7 examples 3 examples seat

(woodwood, marble

wood marble travertine, marble wall

with flower motif with flower motif white plaster marblewhite plaster white plaster white plaster opus s;gnilllll1l, intarsin floor

colored stones spicatum, crude marbles, finewhite mosaics mosaics

simple simple marble, mosaics, opus signinum, white or colored windowssmall; high position small; high position smal1; high position large; high position large with a view

dark/light dark dark light light light

water pipe water pipe water pipe water pipe water pipe water

sponge gutter not preserved not preserved marble masonry or marbletravertine

extra wash basin - labrumdischarge sewer sewer sewer sewer sewer

I

iTab. 2. Single-seaters of Pompeii, Herculaneum, Ostia and Hadrian's villa.

Pom peii / Herculaneum Ostia (2nd century) Hadrian's villa

200 and 60 examples 70 examples 25 examplesseat wood wood white marblewall white plaster not preserved white plaster or marble revetmentfloor tiles not preserved opus sectile in many colorswindows small; high position not preserved large with viewsdark/light dark - lightwater bucket of water bucket, sometimes pipe water pipe water

I Sponge basin or gutter basin or bucket basin or bucket marble basindischarge cesspit sewer sewer

could order the toilets to be built just the way heliked. The interesting question is: 'Do these toiletsdiffer from the toilets of Pompeii and Ostia, dis-cussed in the first part of this article?' When wecompare the toilets of Hadrian's villa with othertoilets in the Roman world (tab. 1), it appears thatthe multi-seaters for slaves are just as somberlydecorated as in other places of the Empire, where-as the water systems very much resemble the sys-tems of toilets elsewhere. Compared to the multi-seaters of Pompeii they receive a lot of light - justas do the ones in Ostia. None of these servant toi-lets, however, fits into Neudecker's category of'PrachtIatrinen'. In contrast, Hadrian's multi-seat-ers for guests and his Single-seaters are the mostluxurious toilets to be found in the entire RomanEmpire (tab. 2). They are the absolute pinnacle,not only with regard to space and decoration, butalso in terms of their hydro-technology. Large,

lowly placed windows not only provided anabundance of light, but also spectacular views.The toilets cannot be regarded as separate enti-

ties. They form an integrated part of the villacomplex. They are not only imbedded in the infra-structure of water and sewer systems, but theyare also part of the architectural program anddecoration schemes. Indeed they fit in very wellwith the many other playful surprises in the villa.Just as MacDonald and Pinto (1995, 170) discov-ered that no two rooms in the villa look the same,no two toilets are equal. Even within one build-ing toilets differ in form and decoration. Apartfrom that, the two toilets with remarkable groundplans, the round one of the Terrace (6) and that ofthe Fountain Court East (12),have no parallels inthe Roman world. Hadrian's guests must havemarveled and felt surprised at the sight of thesetoilets, just as they did at the view of the uncon-

177

ventional architecture of the buildings, the daz-zling gardens and unexpected water displays.

It is much discussed whether Hadrian was thearchitect of his own villa. It is a fact,however, thatwhoever made the drawings and sketches had toplease him. The buildings of the villa must havebeen approved by Hadrian and must have beenexecuted according to his taste. In this way thebuildings and toilets still visible do reveal some-thing of Hadrian's preferences and gives us aglimpse of his character. He seems to have beensomeone who wanted to impress others, andsomeone who did this in a creative, artistic andeven frivolous way. To find out whether thesemany luxurious toilets are typical for Roman em-perors or for Hadrian alone, more comparativeresearch must be done in other imperial palacesand villas. The fact that his direct successors long-ed for more soberness can be read in the Meditati-ons of Marcus Aurelius, who recommends hisreaders to live a simple Stoic life. One piece ofadvice he offers is: '[ ] Do all things as a discipleof Antoninus (Pius) [ ] how he would remain thewhole day at the same occupation, owing to hisspare diet not even requiring to relieve natureexcept at the customary time.''' This quote givesus at least a hint that not all emperors wanted(and needed) a luxury toilet around every corner.

NOTES

1 r want to thank Soprintendente A.M. Reggiani of theSoprintendenza ArCheologica per il Lazio for her kindpermission to research the site. My thanks go also toBenedetta Adembri and the excellent staff at the exca-vation site, Mr Di Croce and Mr Galli, who werealways willing to help. Pina Cinque owes my gratitudefor sharing her information on recently discovered toi-lets. Furthermore, I would like to express my kind grat-itude to Hubertus Manderscheid for all his muchappreciated help and advice regarding this research.Special thanks are for Ann Koloski-Ostrow, who notonly discussed the subject of latrines with me manytimes, but who also kindly corrected and revised theEnglish of this paper. Above all, I would like to thankmy fel.low toilet researchers at the site for their creativeideas and enthusiasm: Carmen Plegiello, Beatrice deFraiture, Nathalie de Haan, Lynn Hagendoorn, RobertKragting, Frederike Schipper, [ercen Van Vaerenbergh,Anneleen Van de Water, Mandy van der Velden andEveline Vermeulen. Many of their ideas and thoughtshave found a place in this article. lowe them for that.This investigation is financially supported by the Nether-lands Organization for Scientific Research (NWO).Photographs and drawings in this article are by theau.thor, ~ess stated ?therwise. The photographs arepnnted WIth the permission of the Ministerio per i Beni~ le Attivita Culturali, Soprintendenza Archeofogica per11Lazio.

2 For a discussion of these problems, see Jansen 2000 and2005.

3 The samples were analyzed voluntarily by Lieselotte

178

Meersschaert and Bart Klinck of the archaeobotanicallaboratory of the Archaeology Department Waasland atSint-Niklaas (Belgium). For more detailed informationabout these investigations see Van Vaerenbergh, forth-coming.

4 This paragraph is based on Neudecker 1995, Koloski-Ostrow 1996 and 2000, Hobson 2004 and forthcoming,and Van Vaerenbergh 2006. Interesting are remarks onlatrines by Carbrecht and Manderscheid (1994, 23-24and 66-67). The results of the author's own investiga-tions at Pompeii, Herculaneum and Ostia (jansen 1993,1997 and 2002) are included too.

5 These toilets are not only few in number, but do not allprovide crucial data: one is overbuilt by the modernrestaurant (toilet in the Forum Baths), and three otherswere not yet finished when Vesuvius destroyed the city(toilet at the Forum, the Great Palaestra and the CentralBaths).

6 In comparison to Pompeii, it is much more difficult todistinguish between public toilets and private ones atOstia. In this article public toilets are the independentones, the ones found in/near baths, in guild houses, inother public buildings as the fire department, and onthe ground floor of the tenement houses. Most of thesetoilets are directly accessible from the streets. The pub-lic ones accommodating one or two persons only (sixexamples) are not taken into account "here.The function of this extra basin is not clear, it eouId befor washing hands and faces, or for storing the sponges.

8 Neudecker 1995, Abb. 72.9 These are the two large toilets of the Baths of Neptune

and the toilet in the south part of the palaestra of theBaths of the Forum.

10 Boersma (1996) has raised a lot of confusion, arguingthat down pipes on the outer site of a house are for dis-charging rainwater of the roofs and that down pipesinside are for toilet waste of upper floors. As no toilethas been found yet on an upper floor in Ostia, his state-ment is regarded as an assumption. In the Casa delleVolte Dipinte is a 'possible' toilet on the upper floor con-nect.ed .with a down pipe and this is a down pipe onthe inside. Boersma does not refer to this toilet in hisargument. '

11 In several toilets dispersed over the city these boardshave been reconstructed in concrete.

12 The fact that this toilet is the only single-seater of whichthe marble seat has been preserved is at least suspi-cious. Especially since the marble is not weathered andis completely intact, it could be one of the many recon-stru~tions. by the excavator Calza. Besides, .he (Giornalidegll SCflVl25, 1938-1939,58) does not mention a seat inhis excavation report, he only mentions the beautifulfloor: '[ ... ] neI sottoscala eben conservato un gabinettopersonale con una pavimentazione in mattonelle esago-nab di J?~umo giallc ~tico [... ['. I thank Jane Shephardand LUIgIMastromarmo, both of the Soprintendenza diOstia, for providing this useful information.

13 For this paragraph see MacDonald/Pinto 1995 andSalza Prina Ricotti 2001. The reference system to thebuildings (names and numbers) used in this article isthat of MacDonald/Pinto 1995, see also figure 1.

14 See for a list of these texts, Salza Prina Ricotti 1995a108-110. Besides, she (1995a, 77-110; 2004, 39-40) argu"';that the women in the palace did not use the normaltoilets, but (now disappeared) chamber pots and woodentoilets.

15 This is the number known in November 2005. In addi-tion to the catalogue published in Jansen 2003 sevennew toilets have been detected. The toilet at FountainCourt East (12), though offering seats for two persons,is regarded here as a single-seater as it has all other

II/I!

/rIII

characteristics of a single-seater. Multi-seaters: eigh-teen-person toilet at Service Quarters (8), eleven-persontoilet at Hall of the Cubicles (15), seven-person toilet atthe East Terrace (19), three- or four-person toilet at theHeliocaminus Baths (20), twelve-person toilet at NorthService Building (26), two- or three- person toilet at theSmaller Baths (30), twenty-person toilet at the Lars;erBaths (33), six- or seven-person toilet at Central ServiceBuilding (34), a ten- person and a sixteen-person toiletat the Canal Block (35). Single-seaters: one toilet at theEast Terraces (6) (new), one toilet at East Belvedere (10),two-person toilet at Fountain Court East (l2), three toi-lets at the Island Enclosure (17) (one new), two toiletsat the Residence (22), one toilet at the ArcadedTriclinium (23) (new), two toilets at ResidenceFountains (24) (both new), two toilets at the PeristylePool Building (27), one toilet at the Stadium Garden(28) (new), six toilets at the Water Court (31), one toiletat the External Nymphaeum at the Water Court (31),OI~e toilet at the West Belvedere (37) (new), and threetoilets at Scenic TricIinium (40). The author does notregard the 'toilet' discovered recently at the NorthTheater (2) by Hidalgo and Le6n (2004) as a toilet, butrather as a lIymphaeum room: there is no sign of a seat,the collection box underneath the seat does not resem-ble the others in the villa, nor is it in the middle of theback ~all as with the others. There are still some poten-tial toilet rooms which have not been cleaned by ourteam: In the Northern Ruins (3), one at East Belvedere(10), one at Hall of the Cubicles (IS), two toilets on theterrace of the Central Service Building (34), two toiletsnext to the nymphneul1l behind the Scenic Triclinium(40), and in tlie Circular Hall (45).

16 See, Adembri 2000, 77; MacDonald/Pinto 1995, 172;Neudecker 1995, 68-69; Salza Prina Ricotti 1995a, 106 n.44; 200~, pas-?im. In several monographs of buildingsof the vlll~ toilets are mentioned. More recently the fcl-lowing toilets have been excavated and are studied atthe moment: the toilet in the Arcaded Triclinium (23)by Letizia Rustico, the toilet of the Stadium Garden (28)by Benedetta Adembri and the toilet of the West Bel-vedere (37) by Alessandro Vella and Marella Maccarone.

17 Ledges indicating front plates have been found at thetoilets of the Smaller Baths (30), at three of the toi.lets ofthe Water Court (31) and at one of the toilets of theScenic Triclinium (38) a small part of the front plate isstill in situ.

18 See for an extended discussion of this theme, Jansen 2003." MacDonald/Pinto 1995, 170-182.20 Salza Prina Ricotti 1995b, 367. See also Ehrlich (1989)

for an evocative description of the 'structural role ofwater architecture in Hadrian's villa' in which she por-trays Hadrian as 'a virtuoso amateur architect, whothrived upon the creation of magnificent water shows'.

21 For many years Hubertus Manderscheid (2000, 2002and 2004) has been studying the waterworks of thevilla, including those of the baths and the Scenic Tricli-niurn (40). Since 2001 a Scandinavian group has carriedout a multi-disciplined survey of the water supply ofVilla Adriana and its possibfe relations to the urbanRoman aqueducts with participants from the Univer-sities of Copenhagen, Lund and Oslo under the direc-tion of Iergen Hansen, Copenhagen, and sponsored bythe Carlsberg Foundation. The results are expected toappear as a supplementary volume 'Water for the Em-peror - Survey of the External Water-Supply of VillaAdriana' in the series of the Accademia di Danimarcain Rome in 2007. More recently, a team of German engi-neers led by Henning Fahlbusch (2003) started study-ing the overall waterworks in the project 'Wasserkulturin dec Villa Adriana in Tivoli'.

22 Among the few that have survived is the pipe in themouth of the crocodile fountain statue found at the bot-tom of the Canopus basin (the original is now in thesite museum) and one found by our team in one of thetoilets in the External Nymphaeum of the Water Court(31). See pp. 168-169 and figs. 12, 13.

2J The speleoiogists of the Associazione Roma Sotterraneaare presently mapping the underlying sewer system.

24 Dunng our research Hubertus Manderscheid helped uswith read tests to establish whether the holes andimprints reaUy indicate the presence of lead pipes. Thename of the test is Rrnerckoquant R 10 077 Pb 2+ andits use is described by Eggert 1988. [0. most cases theresults of this test were positive. Furthermore, we triedto trace lead pipes still hidden underneath floors witha metal detector. This method proved to be successfulat Pompeii earlier (see Jansen 2001). However, in thevilla it was used in vain. We want to thank dottoressaBenedetta Adembri for lending us a metal detector.

25 These can be found in a toilet at the East Terraces (6),Heliocaminus Baths (20), Smaller Baths (30), CentralService Building (34) and the east toilet of the CanalBlock (35).

26 In two multi-seaters, the one at the HeJiocaminus Baths(20) and the one of the Smaller Baths (30), no sign of agutter can be detected. As these are the smallest of themulti-seaters, serving only few people (respectively forthree to four and two to three persons), there mighthave been a sponge basin. Research of the author atOstia has revealed that sponge basins were used at toi-lets servicing up to three or four people, while spongegutters were available at toilets for more than three tofour people.

27 The round toilet at the East Terraces (6), the multi-seaterof the Central Service Building (34) and both single-seaters of the Residence (22).

28 Although wastewater from elsewhere was dischargedin some gutters, this was not done to flush the toilet butrather to find an easy way to dispose of that water inan already existing drain.

29 In the toilet of the Fountain Court East (12) and theStadium Garden (28).

30 Rakob (1967, 12-15) describes this feature for the six toi-lets of the Water Court (31). He calls the way the waterwas provided on these toilets 'extravagante sonst nichtubliche Weise'.

31 Published in Rakob 1967 Abb.1 and Guidobaldi 1994,Tav. F. The team of Fahlbusch has investigated the watersystem of the Scenic Triclinium (40) and made a recon-struction of the water system including the toilets. For afirst impression see Manderscheid 2004, 118 and Fig. 3.

32 Gardens: toilets at the Residence (19) and the SmallerBaths (30). Cliffs: toilet at East Terraces (6) and WestBelvedere (37).

33 See Hida.lgo/Le6n 2004. The other toilets with drainsfrom upstairs are three of the toilets at the Water Court(31) and the one described above, the toilet behind thecupola i.n the Scenic Triclinium (40).

34 Modem bottles and plastic bags in the toilet drain of theround toilet on the East Terraces (6), a cement sack andparts of a grenade in the toilet drain of the ResidenceFountains (24), some lead sheets, a small drawn leadwater pipe from the first half of the nineteenth century,glass and drinking vessels in the toilet drain of ExternalNymphaeum at the Water Court (31). The drain of theeast toilet ~f the Canal ~lock East (36) appeared a dumpfor every kind of material: dead animals, parts of mosaicfloors from elsewhere and all kinds of plastic objects.

35 For example, the finds in the drains of toe toilets at theIsI.an.d .Enclosure (17), the Residence (22), the ArcadedTriclinium (23) and the Peristyle Pool Building (27).

179

36 In the drain we have found many more slabs of the yel-low marble skirting.

37 Only three single-seaters have no niche: the east toiletof the Island Enclosure (17), and the two larger toiletsof the Scenic Tricliniurn (40).

3S Sear (1997, 109) describes his find of yellow and red glasstesserae; these are not visible anymore. Another toilet withmosaics in the apse is the east toilet of the FountainCourt East (12). In the drain of the west toilet of theResidence Fountains (24) we found a part of stucco ofthe apse with two white mosaics still in place and manyimp:rints.

39 In the toilets at the rim of the East Terraces (6) and atthe West Belvedere (37) the walls are preserved at tooIowa height to conclude whether here was a windowor not. Due to their location next to the valley panora-mas, it seems most probable that they had windowsoffering a spectacular view.

4() See Blanco forthcoming and pp. 183-190 in this volume.41 Except for the window in the toilet of the Smaller Baths

(30), which must have been lower in the wall andopened into the garden.

42 The toilet for the visitors of these baths has not beenfound yet.

43 Others, too, had doubts, for example, Guidobaldi (1994,91 and 76) who names the room of the Fountain CourtEast (12) 'latrina?' and the round room at the East Ter-races (6) 'fontana prolibata'.

44 Other round toilets can be found in the Maxentius' Bathson the Palatine in Rome and at Piazza Armerina (Sicily).

45 See for building periods of the villa MacDonald IPinto1995,33, Adembri 2000,35 and Salza Prina Ricotti 2001.

46 This is the eastern room of the External Nymphaeumat the Water Court (31).

-17 The multi-seater at the Hall of Cubicles (15) has beenbuilt in a previously existing: room and the one at theNorth Service Building (26) IS built against the alreadystanding building. It is hard to tell how much later. Sal-za Prina Ricotti (2001, 153-156, 157~158) supposes thatthey were ready at Hadrian's return to Italy in AD 125.

48 MacDonald/Pinto 1995, 198-199.49 In the toilet of the East Belvedere (10) footrests were

added later.50 The drain of the western toilet of the Peristyle Pool

Building (27) was deliberately blocked by building aterrace room over viewing the Stadium Garden (28).After blocking the drain this room was altered at leastthree times, as is indicated by the remains of three dif-ferent floors. The other toilet that has been put out oforder is the round toilet at the East Terraces (6) of whichone of the front plates was used outside the toilet. Thatthis plate was used as a sewer drain tells us that onewas afraid of falling in a still functioning sewer andthus that the toilet was put out of use while the villawas still inhabited.

51 See for more details Van Vaerenbergh, forthcoming.52 Marcus Aurelius, Meditations VI.30 quoted from Loeb

translation.

BmUOGRAPHY

Adembri, B. 2000, Hadrians Villa, Milan.Blanco, A. forthcoming, La latrina delle Piccole Terme, in

AUi del Convegno "Rileggere l'antico", Roma 13-15 dicem-bre 2004, Rome.

Boersma, J. 1996, Private Latrines in Ostia: a Case Study,BABesch 71, 151-160.

Egg~rt, G. 1?88, Eine einiache Feldmethode zum qualita-tiyen Blemachweis, AKorrBI18, 417·420.

EhrlIch, T.L. 1989, The Waterworks of Hadrian's VillaIounwl of Garden History 9.4, 161-176. '

180

Fahlbusch, H. 2003, Villa Adriana: attivita e progetti,Orizzonti, Rasegna di archeologia 4, 119-123.

Carbrecht, G./H. Manderscheid 1994, Die wesserbewtrt-schaftung romischer Thermen, archaologische und l-ry-drotechnische Untersuchungen, Mitteilungen des Leich t-uieiss-lnetitur jUr Wasserbau der technischen UniversitiitBraunschweig, Heft 118 A-C.

Guidobaldi, F. 1994, Sectilia Pavimenta di Villa Adriana(Studi Monografici Mosaici antichi in Italia), Rome.

Gusman, P. 1904, La villa imperiaie de Tibur (Villa Adriana),Paris.

Hidalgo, R./P. Leon 2004, Excavaci6n Arqueol6gica en el"teatro greco" de villa adriana. Resultados de Ia prime-ra campafta (2003), Romula 3, 173-218.

Hobson, B. 2004, Toilets of Pompeii; Human waste disposalpractices over four centuries, in regio VI insula l, unpub-lished dissertation Bradford University.

Hobson, B. forthcoming, Times and Motions. Domestic toi-lets in Pompeii and the Roman world.

Jansen, G. 1993, Paintings in Roman Toilets, in E.M.Moormann (ed.), Functional and Spatial Analysis of WallPain.ting, Procee1in$s of the Fifth International Congress ofAncient Wall Painting (BABesch suppl. 3), Leiden, 29-31.

Jansen, G. 1997, Private toilets at Pompeii: appearance andoperation, in S.E. Bon/R. Jones (eds), Sequence and Spacein Pompeii, Oxford, 121-134.

Jansen, G. 2001, Water Pipe Systems in the Houses ofPompeii: Distribution and Use, inA.O. Koloski-Ostrow(ed.), V'(ater Use and Hydraulics in the Roman City, AlACoiloouia and Conference Papers New Series 3, Boston 27-40. '

Jansen, G. 2002, Water in de Romeinse stad. Pompeji -Herculaneum - Ostia, Leuven.

Jansen, G. 2003, Social distinctions and issues of privacyin the toilets of Hadrian's Villa, IRA 16, 137-152.

Jansen, G. 2005, Stiller Ort mit Aussicht. Die kaiserlichenToilettenanlagen in der Hadriansvilla bei Tivoli warenein Besuch wert, AW 3,1-6.

Koloski-Ostrow, A. 1996, Finding social Meaning in thepublic Latrines of Pompeii, in N. de Haan/G. Jansen(eds), Cum Aquarum in Campania, Proceedings of the 9thInternational Congress on the History of Water Mnnagementand Hydraulic Engineering in the Mediterranean Region1994 (BABesch suppl. 4), Leiden, 79-86.

Koloski-Ostrow, A. 2000, Cacator cave malum: the Subjectand Object of Roman public Latrines in Italy during thefirst Centuries BC and AD, in G. Jansen (ed.) CuraAouarum in Sicilia, Proceedings of the 10th Interl1~tionalCongress all the History of Water Mmzagement and HydraulicEngmeerillg in the Me,{jtermneall Region 1998 (BABeschsuppl. 6), Leiden, 289-295.

MacDonald W./J.A. Pinto 1995, Hadrian's Villa and itsLegacy, New Haven. ..

Manderscheid, H. 2000, Uberlegungen zur Wasserarchi-tectur und ihrer Funktion in der Villa Adriana, RM 107109-140. '

Manderscheid, H. 2~02, L'ar0itettura dell' acqua a VillaAdn~na. con par~cu1are. nguardo al. Cenopo, in A.M.Reggiaru (ed.], Yiila Adriana, paesagglO antico e ambieniemcderno: elemenii di novita e ricerche in corso, Milan, 84-89.

Maf.lderscheid, H. 2004, Speisen von Wasser umgeben.Uberlegungen zum sag. Serapeum, Orizzonti, Rasegnadi archeologw 5, 117-121.

Neudecker, R. 1995, Die Pracht der Latrine, ZUlli Wandeloffent.ficher Bediirftlisallstaltel/ hI der kaiserzeit1ichen Stadt,Muruch.

Rakob, F.L. 1967, Die Piazza d'Oro in der Villa Hadriana beiTivoli, Munich.

Salza Prina Ricotti, E. 1995a, Lavacra pro sexibus separ-avit, RendPontAc 66, 77-110.

Il

i

Salza Prina Ricotti, E. 1995b, Adriano: Architettura delverde edell' acqua, in M. Cima I E. la Rocca (eds), HortiRomani, Atti del ccuzegno il1feYl1nzionale ROllin 4-6 /1Iagg1o

1995, Rome, 363-399.Salza Prina Ricotti, E. 2001, Villa Adriana, 1150gl10 di 1m

imperatcre, Rome.Salza Prina Ricotti, E. 2004, Sabina, in A.M. Reggiani (ed.),Adriano.Le memcne al [ennninite, Milan, 31-45.

Scobie, A. 1986, Slums, Sanitation and Mortality in theRoman World, Klio 68, 399- 433.

Sear, F. 1977,Roman Wall and Val/It Mosaics (RM Sonderheft23), Heidelberg.

VanVaerenbergh, J. 2006, The Latrines in and near RomanBaths of Italy A nice Compromise with a bad Smell. inG. Wiplinger (ed.), Cum Aqllamm.,ill Epneecs 2004(BABesch suppl. 12/ Sonderschrift DJh 42), Leuven/Vienna, 463-469.

Van Vaerenbergh, J. forthcoming, The Stunning La trine ofthe Smaller Baths. Smelly or not?, in Afti del Conuegno"Rileggere l'anticc", ROllla 13 - 15 dicentbre 2004, Rome.

PROFESSOR MOSERSTRAAT 40NL-6224 BL MAASTRICHTgemma.robert@wanadoo.nl

181

BABESCHBULLETIN ANTI EKE BESCHAVING

Annual Papers on Mediterranean Archaeology

No. 82 - 2007

Volume 1

STICHTING BULLETIN ANTIEKE BESCHAVING

© 2007 Stichting Bulletin Antieke Beschaving, The NetherlandsPrinted in Belgium by Orientaliste, LeuvenISSN 0165-9367eISSn 1783-1369

Fig. on cover and title page:Greek Geometric horseAmsterdam, Allard Pierson Museum inv, 1344, formerly Scheurleer Collection