the genesis of the roman public bath: recent … genesis of the roman public bath: recent approaches...

The Genesis of the Roman Public Bath: Recent Approaches and Future DirectionsAuthor(s): Garrett G. FaganSource: American Journal of Archaeology, Vol. 105, No. 3 (Jul., 2001), pp. 403-426Published by: Archaeological Institute of AmericaStable URL: http://www.jstor.org/stable/507363 .

Accessed: 20/03/2013 21:46

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .http://www.jstor.org/page/info/about/policies/terms.jsp

.JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range ofcontent in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new formsof scholarship. For more information about JSTOR, please contact [email protected].

.

Archaeological Institute of America is collaborating with JSTOR to digitize, preserve and extend access toAmerican Journal of Archaeology.

http://www.jstor.org

This content downloaded from 132.206.27.25 on Wed, 20 Mar 2013 21:46:17 PMAll use subject to JSTOR Terms and Conditions

http://www.jstor.org/action/showPublisher?publisherCode=aia

http://www.jstor.org/stable/507363?origin=JSTOR-pdf

http://www.jstor.org/page/info/about/policies/terms.jsp


The Genesis of the Roman Public Bath: Recent Approaches and Future Directions

GARRETT G. FAGAN

Abstract The problem of the origins and early development of

the Roman public bath has proven an intractable one for classical archaeologists. In the absence of hard ancient evidence, many modern propositions have been put for- ward to explain the process of development. In this paper, the six most influential of these approaches are presented and critiqued for their strengths and weaknesses. It is found that none is sufficient in itself to explain the appearance of the Roman-style bath but that most ad- vance the analysis in some measure. Consideration of the often bypassed literary and epigraphic material is also included to help identify the third and second century B.C. as the crucial period in the history of this building type's evolution. Finally, it is argued that the case for Campania remains the strongest, but not (as often claimed) as the place where Roman baths exclusively evolved. Rather, Campania provided an apt context for the final appearance of an architectural genre that had demonstrable roots extending farther afield and deeper into the past. Throughout, directions for future research are suggested.*

In the recent surge of work on Roman baths, the

problem of origins and early development ranks

high among the most contentious and insoluble issues.' In the absence of a consensus, a variety of

propositions has been advanced to explain the unusual form of the typical Roman bath, with its

system of variously heated rooms and communal

bathing pools. It is the purpose of this paper to collate, present, and analyze these propositions, and thereby help identify not only their individual

*I would like to thank the Research and Graduate Studies Office and the Institute for the Arts and Humanistic Studies at Penn State University for supporting the research for this paper with substantial funding in the summer of 1998. Parts of this paper were delivered orally at various institutions and profes- sional meetings, and I thank all the commentators for their input. Various colleagues read earlierversions of the paper (or parts of it) and offered insightful and substantial comments, notably P.B. HarveyJr., RJ.A. Wilson, and the editor and two anonymous referees of the AJA. My thanks also go to A.I. Wil- son, N. DeHaan, H. Manderscheid, and A.O. Koloski-Ostrow for assistance with bibliography, illustrations, and other mat- ters. None of the above, naturally, are responsible for any er- rors that remain.

'The main works are: Brodner 1992; DeLaine 1997; DeLaine andJohnston 1999; Ecole francaise de Rome 1991; Fagan 1999a; Heinz 1984; Manderschied 1988; Merten 1983; Niels-

strengths and weaknesses but also future directions for further research.

Before proceeding, it must be appreciated that the main problem facing an investigation into the

early history of Roman baths is the sparsity of source material, so that a handful of available archaeological sites has tended to be used as a basis for extrapolating general schemes of early development. The issue of typicality has usually not been seriously considered.2 Similarly, literary testimony either has been glossed over or addressed only in the most

general fashion.3 The epigraphic evidence, which derives from a period when bathing culture appears well entrenched in Roman life, is largely uninfor- mative as to origins but offers contemporary docu-

mentary evidence as to use and construction from the late second century B.C. onward. Overall, these difficulties inherent in the early evidence have re- sulted in a greater latitude for varied modern in-

terpretations. In any investigation, it is essential to be clear what

one is looking for. The truly vital question, therefore, is what constituted a Roman-style public bath in contrast to, for example, a Greek one. Two features define the Roman bath: first, it comprises gradations of heat in a clear sequence of rooms (usually termed in modern studies frigidarium, tepidarium, caldarium) that channels the bather purpose- fully from one room to the next; second, the Ro-

man-style bath features heated communal bathing

en 1993; Pasquinucci 1987; Weber 1996; Yegul 1992. There have also been numerous articles on the subject, especially those by DeLaine 1988, 1989,1992,1993,1999. (Most of these works feature discussions of the problem of origins and will be referred to in more detail below.) In 1993 the International Association for the Study ofAncient Baths was established, and two international conferences on Roman baths have been staged in 1992 and 1996. DeLaine (1988, 14; 1993, 354-5) has briefly surveyed the disagreement over the question of origins. 2 For a lengthier discussion of the evidence for the baths and the difficulties of interpreting it, see Fagan 1999b.

3E.g., Yegul (1992, 48-91) very briefly surveys the testimony of Plautus, Varro, Livy, and others who allude to early bathing conditions (infra, pp. 419-21). In contrast, Nielsen (1993, 1:6-36, esp. 1:28-30) addresses this evidence more fully but without close critical analysis.

403 American Journal ofArchaeology 105 (2001) 403-26



GARRETT G. FAGAN

pools (termed solia or alvei).4 It would be fair to say that a truly Roman bath cannot be identified unless both of these elements are present in conjunction. The search for origins must not therefore become sidetracked by optional secondary features that came to be associated with Roman baths as they developed, such as exercise grounds, (palaestrae), sweat baths (sudatoria or laconica), strigiling rooms (destrictaria), ball courts (sphaeristeria), or open-air pools (natationes or piscinae). All of these elements are indeed found in Roman baths, but they are not in themselves defining features of those baths. What is characteristic of Roman practice is communal

bathing in heated pools deployed in a sequential arrangement of heated rooms. For the proper func-

tioning of this sort of bathing establishment, the underfloor heating technique called the hypocaust (termed suspensura)5 proved to be particularly effective. The suspensura raised the entire floor on

pillars and, in a subsequent development, rendered the walls hollow to produce rooms all but lined with surfaces radiating heat. So many Roman baths feature a version of the suspensura that the first appearance of the hypocaust has been considered by many researchers as synonymous with the genesis of the Roman-style bath itself. This is certainly the case with those who propose Sergius Orata or the Greeks as the progenitors of the form.

THE HYPOCAUST, SERGIUS ORATA, AND THE GREEK CONNECTION

Until quite recently, the issue of the origins of Roman baths did not appear to be an issue at all. Ancient literary notices about a certain Sergius Orata seemed conclusive: Orata, living near Baiae in Campania in the early first century B.C., is re-

ported to have invented pensiles balineae (literally, "hanging baths") which he fitted into villas to great profit.6 In the minds of many the pensiles balineae were none other than the hypocaust, leading to the conclusion that Roman baths must have appeared in Campania after Sergius Orata.7 The Stabian Baths at Pompeii, with their second-century B.C. hypocaust predating Orata, generally did not feature in the discussion. The literary sources seemed clear

enough, and some proponents of Orata's role ig-

4See DeLaine 1988, 15-6; Heinz 1984, 51;Yegul 1992, 33-5. 5See Sen. Ep. 90.25; Vitr. De arch. 5.10.2. 6Fagan (1996) and Wikander (1996) collectand discuss the

pertinent references but come to divergent conclusions about them.

7For earlier, largely Italian works propounding this view, see DeLaine 1988, 14-5. See also Benedum 1967; Hilton Turner 1947-1948; Nielsen 1993, 1:20-2. Wikander (1996) considers Orata as the inventor of the system, though he admits the

0 m nn

Fig. 1. Gortys. Groundplan of the "Thermal Establishment." (After Ginouves 1959, fig. 187)

nored the question of the hypocaust in the Stabian Baths.8 In reality, Orata's contribution to the devel-

opmental process is a moot point: a full suspensura has been identified in the Stabian Baths well before Orata; the physical evidence from Greek sites demonstrates that a form of the hypocaust existed

up to 150 years before his activities; and there is

underlying uncertainty in identifying Orata's pensiles balineae with the hypocaust in the first place. Indeed, there is reason to suspect that Orata's device had nothing to do with human bathing at all.9

The greatest challenge to Orata's central role in

developing the hypocaust came from R. Ginouves, who, in two influential studies, argued that the Greeks had created the hypocaust and, along with it, the style of bathing commonly ascribed to the Romans. On this view, the Romans, as in so many other cultural respects, merely took over a form from the Greeks, elaborated it, and built on a larger scale.'? For Ginouves, Orata did no more than in- troduce this heating system to the Romans, or tam-

details are unclear. 8Di Capua (1940, 97-115) offers a good example: the hy-

pocaust is Orata's invention (97-9); the Stabian Baths are mentioned in connection with Greek gymnasia (100) andwith medicinal bathing (115), but the date of the hypocaust in the facility is not addressed.

"Fagan 1996. 0 Ginouves 1959, 1962.

404 [AJA 105

- 10_

_j



THE GENESIS OF THE ROMAN PUBLIC BATH

pered with it in some unspecified way." On close

inspection, however, his case is far from cogent. The two cardinal buildings for Ginouves's theory

are the "thermal establishment" at Gortys in Arcadia

(fig. 1) and the "Greek hypocaust bath" at Olympia (fig. 2). At Gortys, Ginouves was convinced that the two requisite components of the Roman bath could be identified in a building dated to the middle of the third century B.C., while at Olympia a Greek an- tecedent for the developed Roman hypocaust could be found dating to ca. 100 B.C. At a stroke, not only was Sergius Orata's contribution to the history of baths

placed under a cloud, but the Greeks could now be shown, through a direct line of descent linking Gor- tys, Olympia, and Roman baths, to have fully evolved the form of the public bath that had previously been considered peculiarly Roman. While an exhaustive

description of the two buildings is unnecessary here, some discussion of their form is essential if the thrust of Ginouves's argument is to be assessed.

Gortys, Arcadia The features of the "thermal establishment" at

Gortys that most impressed Ginouves were the so-

phistication of the heating and the water supply/ drainage systems, and the complexity of the internal room arrangement (fig. I).12 The underfloor

heating system comprised a furnace (Y) with a subterranean corridor running out from it, under a water heating system (G'), immersion tubs (D), and a round room (E), and then turning south to run under the east apse of the central room (C) and the north apse of the vestibule (B). It is worth not-

ing that here, in one of the earliest examples, the

hypocaust is used to heat both water (in G' and D) and space (in E, C, and B).13 An aqueduct-fed reservoir (X) supplied the facility with water, and drains

placed to take advantage of the sloping topography dealt with the effluent. The capstone in Ginouv&s's thesis was the arrangement of the rooms, where he saw a precursor of the room sequence normally associated with Roman baths. Ginouves went so far as to reconstruct the path of the bather through the

building: from the portico (A), into the vestibule/

1 Ginouves 1962, 208-9. ' Ginouves 1959, 58-77 (heating), 78-88 (water suppy/

drainage). Heinz (1984,47) detects "die planende Hand eines geschulten Architekten" in the building's subtleties.

3 DeLaine (1988, 16) has forcefully argued that the hypoc- austwas developed first and foremost to heatwater and was only extended to heat spaces in a secondary development. The arrangement at Gortys undermines her position to some degree.

14 There are numerous examples of Greek establishments with hip-baths, usuallyfitted into rotundas: e.g., those at Olym- pia, Eretria, Oeniadae, Athens, Piraeus, and Eleusis; in Magna Graecia they are found at Megara Hyblaea, Morgantina, Syra-

changing room (B), and thence to the central room

(C). Here a series of possibilities presented themselves. One could wash at the fountains in the central room (0 and A in room C) or avail of various amenities, such as the immersion tubs (D), the sweat room (E), the waiting area (F), or the hip-baths in the adjacent tholos (G). The presence of the hip- baths placed the building firmly in the Greek tradition, since these curious little tubs are the hall- mark of Greek bathing practice.'4 For Ginouves, this ground plan constituted a recognizable series of rooms and was the final proof that the Greeks had

fully developed all the elements of Roman baths.15 The construction date was firmly fixed by numis- matic and ceramic evidence at the fourth or early third century B.C., with the hypocaust installed in the middle of the third century B.C.16

0 5 I _ J

m

Fig. 2. Olympia. Groundplan of the "Greek hypocaust bath." (After Kunze and Schleif 1994, fig. 19)

cuse, and Gela (infra, pp. 414-7; see figs. 10 and 15); inAfrica, they are found in the rock-cut baths at Cyrene. For the hip- bath as the trademark of the Greek bathing establishment, see Ginouves 1962, 101-3, 185-7; Yegul 1992, 24-9.

'5 Ginouves 1959, 46-8 and quote at 167: "Le role des Ro- mains fut de concentrer les dispositifs, en poussant plus loin l'analyse des fonctions, et de harmoniser a leurs gofits moins <sportifs-." See also Ginouves 1962, 209.

16 Ginouves 1959, 135-45. An earlier bath had been built ca. 370 B.C. butwas destroyed in the middle of fourth century and replaced with the current structure.

2001] 405



GARRETT G. FAGAN

Ginouves's propositions, however, fall into difficulties when closely scrutinized, especially his reconstructed room sequence. For instance, there is no recognizable frigidarium, since all the bathing rooms are at least partially heated, and it is equally difficult to identify specifically a caldarium or tepidarium."7 The supposed series of rooms is therefore largely illusory, especially when one considers that once in the central room (C) the bather had at least four possibilities available (five, if one counts the bench and fountain in F), with none clearly lead-

ing to any other. More importantly, there is no facility for communal immersion in either hot or cold water at Gortys, and without this we cannot think of Roman-style bathing taking place there."8 Unlike the alleged room sequence, the hypocaust is an undeniable feature of the building, but the extent to which it can be linked directly with the Roman- style suspensura is less clear-cut. The system at Gor-

tys, like other Greek hypocausts (e.g., at Gela, Syra- cuse, or Megara Hyblaea), is quite different from the familiar Roman suspensura.19 In the latter, the entire floor area of a room was raised and heated, while in the Greek "annular" models only sections of rooms were heated by means of a subterranean corridor (although at Gortys, the entire floor area of the small room E was raised). The two systems are similar only to the extent that they share the same basic principle of underfloor heating. But from this feature alone, a formal line of descent from Greek to Roman baths can be traced only very tenuously.

Finally, broader contextual considerations cast doubt on a direct link between the establishment at Gortys and the developed Roman bath. It is difficult to see how a backwater such as Arcadia could contribute a major new form to leisure architecture, especially in the context of the instability of mid- third century central Greece.20 The structure is also the centerpiece of a sanctuary of Asclepius, whose healing cult required rites of purification and bathing.21 Other sanctuaries of this god featured baths

7Heinz (1984, 48) and Ginouves himself (1959, 167-8) concede some of these weaknesses.

'8But note that roomYwas once a large cold pool, before it was converted into the furnace of the hypocaust in the mid third century B.C., see Ginouves 1959,56-7. Room D contains hot immersion pools but they are for individual use. Nothing like the Roman piscina or alveus is found here.

' See DeLaine 1989, 111-4; Nielsen 1993, 1:20-1. 20A point well made by DeLaine 1988, 15. 21 Bathing featured in the preparatory rituals before sacri-

fice and often in the cures "prescribed" by the god; see Boud- on 1994; Ginouves 1962, 349-61. A concise summary of this cult is provided byJackson 1988, 140-55 (note especially the place of bathing in the cult, ibid., 145). In fact, baths played a significant role in Greek religious life in general: over half of Ginouves's magnum opus (1962, 234-428) is concerned with

among their buildings and tended to be situated in well-watered places.22 Ginouves downplays the cultic aspects the establishment's function; he argues that it was primarily a regular thermal establishment, although under the patronage of a god.23 But this division of function strikes a hollow chord. The building's patrons would have been frequent- ing it not merely to get clean and to socialize, but to meet requirements of purification or to implement the god's remedial prescriptions.24 There are indications from the building itself that this was so. Two statue bases were found (in rooms A and I), quite possibly for images of the god or his associates (such as Hygieia). In the northeast section of the establishment, two rooms (H and I) had no discernible function in the bathing process and may well have served as a shrine or healing rooms, to be used after purification in the establishment's facilities; it may be significant that they are accessible only through the hip-bath tholos (G). An inscribed terra- cotta votive foot was found in the reservoir (X). This is a familiar element of the cult attested at other

Asclepieia, where model body parts were regularly used to call the god's attention to an afflicted limb or organ. Other inscriptions of a religious or votive nature were found reused in the Roman walls on the site, although none could be definitely con- nected with the thermal establishment.25

In short, while the establishment at Gortys shows beyond doubt that the Greeks developed certain rudimentary forms of technology later commonly employed in Roman baths, it is difficult to see the building as a direct forerunner of those baths. To the best of our knowledge, it stood in a relative backwater of Greece, lacked elements essential to Roman baths, and, for all that, operated in the religious context of a sanctuary of Asclepius. Whatever else may be claimed about the building, this last point is surely important as it suggests a function for the Gortys establishment quite distinct from the mundane servic- es offered by the civic baths of the Roman era.26

"la propret6 et la vie religieuse." 22 Such as Epidauros; see Lambrinoudakis 1994. On other

Asclepieia, see Aleshire 1989, 1991 ;Jackson 1988, 148-55. For the well-watered nature of Asclepian sanctuaries, see Burford 1969, 45-7.

23Ginouves 1959, 47, 156. 24Ginouves (1959, 48) practically admits as much when he

concedes that the immersion baths in room D in particular, and the whole building in general, would have been visited mainly by the ill.

25 For all of this, see Ginouves 1959, 7-19 (room A), 44-6 (rooms H and I), 139 (votive foot), and 143 (inscriptions).

26 It is certainly the case that the Romans later equipped Greek sanctuaries with "regular" baths (e.g., atEpidauros, Olym- pia, Delphi, and Eleusis, among others), which were identical in form to those found in civic contexts, but the role such sanc-

406 [AJA 105




Olympia The other key facility in Ginouves's case is phase

four of the baths at Olympia, usually termed the "Greek hypocaust bath." The establishment as a whole had a long history stretching back at least to the fifth century B.C. The first three structures were all Greek-style facilities, equipped with hip-baths.27 With phase four a new departure was evident. A room was built featuring a full Roman-style suspensura on 90 pillars supporting the entire floor area of the room (fig. 2). In addition, there was a heated communal pool (A) at one end, and, at the other, an apse (B) that yielded vestiges of an emplacement for a labrum (a raised basin for dousing). There was even a place in the rear wall of the communal

pool for setting the device (C) called a "pool tor- toise" (testudo alvei), which was used to keep all the water in the pool at an even temperature.28 This bath was dated by the excavators, rather shakily, to ca. 100 B.C.29 Ginouves believed that it clinched the case for the mainland Greeks' development of a Roman-style caldarium, in all its details.30 More- over, the building was a decade or more older than

Sergius Orata and had been constructed a century and a half after the baths at Gortys. It appeared to be a missing link.

As with Gortys, however, there are serious questions. The excavator himself expressed the opin- ion that phase four marked a sharp change in bath-

ing practices at the site, here expressed architec-

turally.3' What could have caused such a change? There is no evidence for a local, gradual evolution from the use of hip-baths to communal bathing.32 Rather, the change at Olympia is abrupt. A reasonable explanation is that phase four, rather than

evolving organically from what came before, marked the grafting of new, already evolved forms onto earlier structures. The source for those forms is not far to seek: Rome and Italy. By the time the Olympia caldarium was built in the first century B.C., the Roman public bath was well established in Italy.

tuary facilities played in the religious life of their clients and how their use differed, if at all, from "regular" Roman baths remains in need of closer investigation. Among the Greeks, however, where "civic" baths were far more rare, the appearance of baths in religious contexts probably indicates an association of these facilities primarily with the purifying aspects of worship (this is especially true of the Asclepian cult). For a recent assessment of the role of baths in Greek medical sanctuaries, see Ginouves 1994.

27Kunze and Schleif 1944, 51-6; Mallwitz 1972, 270-3. 28The terms are given in Vitr. De arch. 5.10.1. 29 Kunze and Schleif 1944, 79-80 (on the basis of pottery). 30Vitr. (De arch. 5.10.) provides the classic description of the

caldarium. See also Ginouves 1959, 168; supra, n. 11. '3 Kunze and Schleif 1944, 51. 32This excludes the occasional Greek communal cold pool,

..-...._r~~~~~ _. J _vii. 114

vii.1.9 vii.l.9a

0 5 10 m

Fig. 3. Pompeii. Groundplan of Stabian Baths. (After Yegul 1996, fig. 59)

Furthermore, Romans had been directly involved in Greek affairs for over a century, and some had

emigrated to live there. In particular, there is evidence for Roman architects at work in the Greek east in the early second century B.C., and Romans had been given permission to participate in pan- hellenic athletic games at least since the late 190s B.C. Romans would undoubtedly have been spec- tators at the great Greek athletic festivals throughout the second century B.C. Roman suppression of the Achaean revolt and the destruction of Corinth in 146 B.C. can reasonably be seen as heralding a more lasting Roman presence in the area.33 Given all this, it is more likely that phase four of the baths at Olympia is the result of Roman influence on Greek

bathing practice, so that the supposed "Greek hypocaust bath" was really a "Roman hypocaust bath at Olympia." Recent work at the site redates it to ca. 40 B.C. and argues forcefully for the structure be-

ing an import from Italy into Greece.34 The absence of the other requisite rooms of a Roman-style bath

found usually in association with sanctuaries and therefore possibly of religious function (e.g., at Olympia, Delphi, Delos, Nemea).

33 The architect Cossutius, a Roman citizen, worked for Antiochus IVEpiphanes (175-163 B.C.), see Vitr. Dearch. 7.15- 17. For Romans living in Greece, see Wilson 1966, 94-8. M'. Acilius Glabrio, among other honors, had been granted the right to participate in athletic contests in 191/90 B.C., see SIG3.608 (Delphi). On the Achaean revolt and its aftermath, see Gruen 1984, 514-27.

34 See Ladstatter 1995. Ladstatter's work corroborates the earlier date already proposed by Nielsen (1985, 101-4; 1993, 1:22). DeLaine (1988,15) brands ita "Roman import." Recently, Farrington (1999, 62 n. 27) rather elusively considers phase four at Olympia a "transitional stage" and maintains its date of ca. 100 B.C.

2001] 407



GARRETT G. FAGAN

Bathing wing 0 10

T rli li n nt m

Fig. 4. Pompeii. Eschebach's phase one for the Stabian Baths. (After Eschebach 1979, fig. 43a)

may suggest that phase four was experimental in some way, a local and partial adaptation of the more elaborate forms of a Roman bath. It certainly has no

parallels among Greek baths and, like Gortys, it is located in a religious sanctuary.

While Ginouves's work established beyond doubt that the Greeks had evolved a culture of public bath-

ing and had developed technology to serve it well before the Romans, his Greek connection is not sufficient in itself to explain the origin of the Ro-

man-style bath as defined above. When one inspects the evidence, there are serious problems with the two sites he adduces as the chief support for his view. Nevertheless, his opinions appeared to find

strong confirmation when the Stabian Baths at

Pompeii were subjected to a close investigation in the early 1970s.

THE STABIAN BATHS AT POMPEII: FROM

GREEK TO ROMAN?

Eschebach's Scheme The Stabian Baths at Pompeii are the oldest near-

intact set of Roman public baths to survive from

antiquity (fig. 3). H. Sulze was carrying out analytical work on the structure when his investigations were interrupted by the Second World War. His notes were subsequently destroyed in the firebomb-

ing of Dresden. His student, architect H. Esche-

35 For the original dating of the building's construction, see Mau and Kelsey 1907,189-201. For Eschebach's investigations, see Eschebach 1970, 41-5; 1979.

36 For a summary of the building's history, see Eschebach 1979, 51-3 (on the Greek bath), 64-73. For acceptance of Eschebach's phases, see, e.g., Gros 1996,391; Heinz 1984,52- 7; Nielsen 1993, 1:25-36; Weber 1996, 34-9, Yegil 1992, 61, 373-4. Zanker (1998, 50) implicitly accepts a pre-second century existence of the Stabian Baths, presumably in deference to Eschebach's reconstruction of the building's history. Niels-

vii. 1.9a

Fig. 5. Pompeii. Eschebach's phase two for the Stabian Baths. (After Eschebach 1979, fig. 43b)

bach, returned in the 1970s and carried out a close

analysis both of the building's surviving fabric and, through sondages at various points around the

building, of the underlying strata. His results were

startling. Although the middle of the second cen-

tury B.C had been established as the construction date for the Stabian Baths, Eschebach deduced a traceable history in seven distinct phases dating back into the fifth century B.C. What is more, for the first three phases-covering the fifth to the second century B.C.-he reconstructed the building as largely Greek in form.35 Here, at a single site, the evolution of a set of baths from Greek-style to Ro- man could be charted (figs. 4-7). Ginouves's Greek

origin for Roman baths seemed fully vindicated, and Eschebach's reconstruction of the building's history has met with widespread acceptance, so that it is now frequently cited as established fact.36 In

reality, it is largely conjecture. Eschebach's principal means for deducing his

building phases is construction technique, a weak basis for determining detailed chronological rela-

tionships within or between structures. On a broad

perspective, it ought to be a matter of concern that this analytical method works so well for the early periods of the building's alleged evolution, for which the evidence is scarcest, but less effectively

en (loc. cit.) goes so far as to use Eschebach's scheme for the Stabian Baths as a blueprint for the development of the Ro- man-style public bath in general. One prominentvoice of pro- test is Richardson (1988, 100-5), but since his overall synthe- sis has failed to garnerwidespread support, his perceptive comments about the Stabian Baths appear to have gone largely unnoticed, although see the comments ofYegul 1992, 434-5 n. 19. Koloski-Ostrow (forthcoming) takes a moderate and sensible line on these issues.

408 [AJA 105




Fig. 6. Pompeii. Immersion tub in north wing of Stabian Baths (measuring tape set at 0.5 m) (G. Fagan)

for the later periods, when more of the structure is available for examination. It seems strange that his first three (Greek) phases can be so precisely dis-

tinguished and dated while the last two phases, the sixth and seventh, embrace over a century of the

building's history during which numerous alter-

ations, repairs, and extensions were carried out that Eschebach has great difficulty identifying as dis- crete operations. More troublingly, Eschebach's

chronology rests on unstable stratigraphic founda-

tions, a fact commented upon by reviewers when his major publication of the site first appeared but

largely overlooked since.37 The dating of his early phases remains unclear. Indeed, only the fifth

phase is truly datable, on the basis of an inscription recording the activities of the duoviri Uulius and Aninius in ca. 80 B.C.38 A pottery sherd discovered under the floor of a room in the north wing (the present-day latrine) establishes only a terminus post quem for the laying of that floor, that is, sometime after the fourth century B.C. No more precise a date for Eschebach's phase three can be offered. Aside from these dating difficulties, the form and nature of Eschebach's first three phases raise more questions than they answer. A closer look is in order.

The supposed fifth-century bath (phase one) is most unusual, without parallel in any Mediterra- nean context (fig. 4). It comprised a hip-bath com-

37 E.g., Ling 1981;Yegfil 1981. For the unreliability of draw- ing chronological conclusions primarily from construction technique at Pompeii, see Fulford and Wallace-Hadrill 1998.

plex, in which the tubs were deployed in five individual "bathing cells" (Badezellen), as opposed to

being arranged around the inside walls of an open room, which was the norm for such early balaneia

(compare figs. 1, 11, 13). The hip-baths themselves were of unusual form, not least their orientation within their cells: the bather would have been re-

0 10 m

vii. 1.9a

Fig. 7. Pompeii. Eschebach's phase three for the Stabian Baths. (After Eschebach 1979, fig. 36b)

38 CIL 1.1635 = CIL 10.829 = ILS 5706 = ILLRP 648 = Fa-

gan 1999, 250 (no. 61).

2001] 409



c

D E

Fig. 8. Pompeii. Some entrances to the Stabian Baths. (G. Fagan). A, entrance vii.1.48 on Vicolo del Lupanare; B, entrance vii.1.9a on Via dell'Abbondanza; C, entrance vii.1.17 on Via Stabiana; D, entrance vii.1.51 on Vicolo del Lupanare; E, entrance vii.1.14 on Via Stabiana.




quired to sit in the tub facing the far wall with his or her back to the door, which was the main light source for these gloomy little spaces. This disposition con- travenes all known Greek examples, where the hip- baths face outward from back walls. Eschebach's curious hip-bath complex was then associated bodily with a palaestra to form what he believed was an "ath- lete's bath" (Athletenbad). No such association of hot- water bathing and palaestra came into existence on the Greek mainland until the third century B.C.39 It strains credibility to accept that here, at the very edge of the Greek world, such a farsighted development had taken place two centuries earlier. Furthermore, the shape of Eschebach's palaestra in this early bath

presupposes the later road system at Pompeii; his

attempt to explain the shape by arguing that the

original town wall ran along the western side is un-

convincing and lacks supporting evidence.40 The same can be said for his belief that this early bath was buried in an eruption of Vesuvius at the end of the fifth century B.C. If this were the case, it is curious that the building was reconstructed along identical lines in Eschebach's phase two-dated on no secure

grounds to the fourth or third centuries B.C.-but now stripped of its most innovative feature, the hip- baths in their bathing cells (fig. 5). Instead, the re- built cells were fitted with individual immersion tubs (still to be seen today, fig. 6), although they remained associated with a palaestra. He contends that some ancillary "gymnasial" elements of unclear function were added in this phase along the northern and southern flanks of the exercise court (A in

fig. 5). As with phase one, this whole arrangement is

entirely without parallel in the contemporary Medi- terranean basin. It is also worth noting that the bath-

ing cells with their supposed immersion tubs show no sign of plumbing and may well have served needs unrelated to bathing proper.41

With Eschebach's phase three, the rudiments of a Roman-style bath make their first appearance in

39SeeYegiil 1992, 21-4 with ample reference to earlierstud- ies. Later Greek-style baths in the West (e.g., at Gela, Syra- cuse, Megara Hyblaea, all dating to the fourth and third centuries B.C.) were not associated with a palaestra, so one cannot argue that this was a regional innovation.

40 See Richardson (1988, 100-5) for a counterview. On the obscurity of the city's early history, see Cerchiai 1995, 129-39; Fulford and Wallace-Hadrill 1998; Zanker 1998, 31,44-53.

41 On the absence of water supply or drains in the "bathing cells," see Eschebach 1979,38-9. Compare the uncertainty over the function of a similar row of cells in the later Sarno Baths at Pompeii; Ioppolo 1992, 72; Koloski-Ostrow 1990,94-5.

42 Note here the "rudimentary" private bath in a house at Roccagloriosain Lucania; see Fracchia 1999. Public bathing does not feature as an attested part of Samnite life; see Salmon 1967, 50-64; Tagliamonte 1996, 156-78, 236-42. Most of the remains of "Oscan Pompeii" are usually dated to the second or

the scheme (fig. 7). There was now a sequence of rooms (I-III) arranged diagonally, although the function of each remains uncertain and there was no clear evidence of heating or plumbing in any of them; the use of braziers should not be ruled out. What is more, the bathing cells continued in use. Eschebach gives no indication how this building was to be used. The inspiration for the introduction of a room sequence is also unclear, since the

period to which Eschebach assigned this phase (the fourth century or earlier) was that of the Oscan oc-

cupation of Pompeii, and the Oscans were known neither for their devotion to public bathing culture nor their architectural innovation.42

Eschebach's remaining phases (four through seven) rest on more solid ground, since they cover the period of the building's generally accepted existence, from the mid second century B.C. onward. Throughout these periods the baths were fully Roman in type, with communal pools and clear se-

quences of rooms heated with hypocausts. It is therefore unnecessary to review them in detail. The difficulties lie primarily with Eschebach's first three

phases, which are critical for establishing the Greek- to-Roman development of the Stabian Baths. Rich- ardson has offered an argument that, in my opin- ion, settles this debate. He noticed that five of the

public entrances to the building, on the three sides that front onto streets, were constructed of the same material (tufa) and fashioned in the same style (e.g., fig. 8a-c; the location of all entrances is indicated in fig. 3 with the tufa examples underlined). Three entrances are of a noticeably different de-

sign and construction (e.g., fig. 8d-e). The conclusion seems inevitable that the whole building, as

suggested by the five tufa entrances, was conceived as an entirety and erected in a single operation, probably ca. 140-120 B.C., as originally proposed by Mau, or slightly later.43 The alterations and repairs of the Sullan, Augustan, and early Imperial

early first century B.C., see Zanker 1998, 32-60. 43 Mau based his dating on overall construction technique

(Mau and Kelsey 1907,35-44, esp. 40; see also Richardson 1988, 103). This method has come under serious assault recently (Fulford and Wallace-Hadrill 1998), but the presence of a sun- dial bearing an Oscan inscription in the east portico of the palaestra (Eschebach 1979, 17 and abb. 5) probably suggests an early date for the building, before 100 B.C. at the latest. Stratigraphic work on the site could clarify the matter. What- ever the case, Eschebach's scheme requires that each of the tufa entrances was constructed at different times over his early phases (compare figs. 5 and 7 with fig. 3), necessitating the builders to adhere closely to a particular door design up to his fifth phase, when they changed over to a new model. This is an example of the illogic ofEschebach's entire reconstruction criticized by Richardson (1988).

411



GARRETT G. FAGAN

Fig. 9. Pompeii. Niches in the women's apodyterium of the Stabian Baths. (G. Fagan)

periods-corresponding to Eschebach's fourth

through seventh phases-then account for the three,

evidently later entrances. Finally, no features of the construction techniques employed in the north wing suggest that it existed independently of and earlier than the rest of the facility. If these arguments are

accepted, there was no Greek bath on the site.44

Nielsen's Scheme The notion of its existence, however, persists.

Nielsen, who accepts Eschebach's overall scheme in a slightly modified form, has offered an alternative version of the Greek-to-Roman evolution of this

building. She reconstructs Eschebach's phase four

(of the second century B.C.) as a Greek balaneion, with dozens of hip-baths occupying the bathing rooms in the east wing (fig. 3). In contrast to Esche-

bach, she believes the building lacked hypocausts at this time: they were added later. Unfortunately, no trace of hip-baths remains in the Stabian Baths,

44 It remains unclear what exactly Eschebach found in the Badezellen, since he provides no photographs of the hip-baths in his book, only drawings (1979, 51 abb. 16 and taf. 33c). Indeed, there are neither photographs nor full-scale drawings of the crucial north wing in the book, although the other wings enjoy ample illustration. In an attempt to clarify this issue, the author plans to conduct investigations in the cells in the future.

45 See Nielsen 1985; 1993, 1:27-30. Nielsen's reconstruction is accepted by, for example, DeLaine 1989, 117-20; Gros 1996, 392.

46DeLaine (1989,118) suggests that as many as 35 hip-baths stood in the women's apodyterium alone. As Nielsen (1993,

so Nielsen deduces their existence from the presence of niches in some of the surviving structure's rooms. They are most clearly visible in the men's and women's changing rooms (apodyteria), although she presents evidence that they once featured also in the other bathing rooms (fig. 9). Nielsen reasons that, since analogous niches were found over

hip-baths in Greek-style establishments to accom- modate the bathers' belongings as they soaked in the tubs, so hip-baths must once have stood beneath these niches in the Stabian Baths. Therefore, the Stabian Baths of the second century B.C. may be identified as a fully-fledged Greek balaneion.45

Once more, however, closer inspection renders this position questionable. In the first place, that no trace of a single hip-bath has survived in the Stabian Baths is at least suspicious, since they would have numbered 100 or more, and the niches supposedly associated with them can still be traced with such facility.46 Second, the presence of niches is

1:27) reconstructs four or five hip-bath chambers, a total of 150 hip-baths is not inconceivable for the complex. Nielsen (1985, 86) avers that the hip-baths were made of terra-cotta and may have been movable. But even portable hip-baths often leave some trace of their presence, as at the Greek Har- bour Baths at Eretria, or the Greek bathhouse at Morgantina; see Ginouves 1962, 183-98 (Harbour Baths) and Allen 1974, 379,381 (Morgantina). Also, Greekbaths thatwere later transformed into Roman-style facilities can leave indications of their original Greek character; see the examples cited by Nielsen (1993, 1:101) from Athens, Egypt, and Pergamon. This is not so for the Stabian Baths.

412 [AJA 105




Fig. 10. Pompeii. Niches in the men's tepidarium of the Forum Baths. (G. Fagan)

not, in itself, justification for postulating hip-baths beneath them. The Forum Baths at Pompeii, for

instance, feature niches in the men's and women's

tepidaria (fig. 10). There is no suggestion that hip- baths ever stood under these niches, since the Fo- rum Baths were built in ca. 80 B.C. and indisput- ably served Roman-style bathing habits. The same can also be said for other examples of Roman public baths, where the niches appear to have served as both decorative elements and temporary storage shelves for bather's gear as they proceeded through their routines.47 Third, according to Nielsen the

hip-bath rooms were later transformed into hypocausted and tubulated Roman-style bathing rooms. This renovation would require that the cavea of the

hypocaust was installed by digging down below the

hip-bath floor-level, a dubious procedure that risked destabilizing the foundations; thus the hy-

47 For instance, the Central Baths at Cales (built ca. 90-70 B.C. according to Nielsen's catalogue entry, C.35), featured niches which Nielsen (1993, 1:32 n. 59) explains as "presumably merely decoration adopted from the earlier establishments at Cumae and Pompeii." Other examples of niches in hypocaust-heated bathrooms are found in Spain at, e.g., Baetulo (C.100 in Nielsen's catalogue; first century B.C.) and Los Banales (C.113; mid first century A.D.). On the functional

pocaust and the niches are more likely contempo- raneous.48 Finally, if all the rooms in the east wing were fitted with hip-baths, there seems little reason for their linear sequence and their deployment on either side of a centrally located furnace. Rather, the room sequence alone suggests that a Roman-

style bathing habit, not a Greek one, was being ca- tered to in the building's design.

After all this discussion, it must be concluded that the Stabian Baths cannot offer any secure evidence concerning the supposed transition from Greek to Roman baths. The evidence of the site is

ambiguous and debatable, and short of total exca- vation of the lower strata the picture is unlikely to be clarified. It seems more plausible that this build-

ing was constructed as a Roman-style facility in ca. 140-120 B.C., with its sequence of rooms, hypocausts, and heated communal pools all present at

purpose of such shelves and niches in heated rooms, see Hei- nz 1984, 52-3.

48As noted, DeLaine also accepts the presence of hip-baths in the Stabian Baths (supra nn. 45, 46) and therefore seems to think the hypocaust and hip-baths were utilized contempo- raneously; see also DeLaine 1989, 119-20. This seems unlikely, since hip-bath chambers and hypocausts are routinely sep- arate installations in survivng Greek baths.

2001] 413



GARRETT G. FAGAN

0

A m

r l l

I!

Fig. 11. Syracuse, Sicily. Groundplan of the Greek Bath. (After DeLaine 1989, fig. 5b)

the outset. There are indications in the confident hand of the architect that these features had as yet unidentified predecessors of uncertain location and date, perhaps reaching back into the early years of the second century or beforehand. A detailed examination at some future date of other early but

ill-investigated establishments-such as the Cen- tral Baths at Cumae or Cales-may answer some of

the questions that remain open about the early appearance of the Stabian Baths, but for now only spec- ulation prevails. In any case, even if a Greek-to-Ro- man evolution could be securely identified at the Stabian Baths, the issue of typicality surely advises caution before extrapolating wholesale from this one building an entire development scheme for the Roman public bath as an architectural genre.

THE ROLE OF MAGNA GRAECIA

While the case for the Romans adopting their

bathing habits wholesale from the mainland Greeks no longer appears convincing, that is not to deny the influence of Greek culture on Roman bathing habits entirely. The Greeks had a fully developed culture of public bathing, not just in athletic facilities such as gymnasia and palaestrae but also in bal-

aneia, genuine public baths with heated water. In-

deed, the initial stimulus for the Italic adoption of the public bathing practice in an urban context most

likely came from the Greeks, since other Italic cultures (such as Oscan) show no overt penchant for a

bathing in public.49 Ancient Italians would have en- countered many models of public baths in the Greek

poleis of Magna Graecia, and several balaneia have indeed been found in the Greek cities of the West. It therefore seems reasonable that some scholars have credited the Western Greeks with taking the critical steps toward developing Roman-style bath-

ing habits, which were then picked up and adapted

Fig. 12. Syracuse, Sicily. Rooms A (top right) and B (foreground) in the Greek Bath. (G. Fagan)

49 Supra n. 42. The intense hellenization of Etruscan civilization makes its contribution to Italian public bathing moot.

[AJA 105 414




by neighboring Italians. Pointing especially to the

fourth-century Greek baths at such sites as Syracuse and Gela in Sicily,J. DeLaine has detected the presence of communal immersion pools heated by hypocausts, used in conjunction with a tholos fitted with hip-baths (figs. 11-14). It was but a short step from this arrangement to extending the hypocaust under an entire room, a process DeLaine sees pre- saged in the parallel heating channels of the Greek baths at Gela (figs. 13-14).50 With these communal immersion pools, we are clearly on the road to the

Roman-style bath. Thus the Roman bath owes its

origins to the Greeks, not of the mainland, but of

Magna Graecia.

Despite the cleverness and attractive elegance of this theory, a closer look at some of DeLaine's model Greek establishments generates doubts. None of them, for instance, has a clear room se-

quence; all retain the randomized internal ar-

rangement that typifies the traditional Greek bath-

ing establishment. The revolutionary feature of the Greek baths at Syracuse (which functioned in the fourth and third centuries B.C.) is the room B with an associated annex A that DeLaine inter-

prets as a communal bathing pool heated by a hypocaust (see figs. 11-12). But DeLaine's communal pool A, while superficially resembling the Roman solium or alveus, was 0.35 m deep, which is considerably shallower than the habitual depth of Roman pools and rather meager for a sufficient

bodily immersion.51 Yet pool A was certainly meant to contain water, since a drainage channel flowed from it into room B. The question really is how the

pool was used. A genuine possibility is that it served as a tank out of which individual bathers collected their hot water for use in the associated larger room. That room B at Syracuse, and the analogous room in the Greek bath at Megara Hyblaea, bore indications of permanent fixtures around the inside surfaces of its walls (indicated in fig. 11) could be a significant factor: the bathers may have collected water from pool A and used it individually in room B at wall emplacements, whether they were benches or basins. Such a practice would be consistent with what we know of Greek bathing practices that stressed the separateness of each bath-

50DeLaine 1989. Broise (1994) has argued a parallel position, apparently unaware of DeLaine's important article. Note that both authors consider the hypocaust primarily a water- heating device.

51 See the original excavator's report (Cultrera 1938, esp. 268-70) on this room. For parallel examples from Megara Hyblaea and Morgantina, see Broise 1994. The depth of the Syracuse bath's "pool" (0.35 m) was measured by the author at the site in June 1998. By way of comparison, solia in typical

er's experience and, coincidentally, was precisely the procedure followed by the author on a visit to a Tunisian hammam in 1995, where patrons collected water from a large communal pool and washed with it while seated on built-in benches. If

so, the function of such heated pools at Syracuse and elsewhere in western Greek sites was quite different from that of the later Roman solium/ alveus, and the bathing habits the pools served were more Greek than Roman.

DeLaine argues that the double-channel hypocaust in the Greek bathhouse at Gela, dating to the late fourth and early third century B.C., offers a precursor to the hypocaust in the so-called Republi- can Baths in Pompeii, which was raised not on pillars but on walls radiating from the furnace (figs. 13-15). On the basis of this unusual hypocaust,

Fig. 13. Gela, Sicily. Groundplan of the Greek Bath. (After Orlandini 1960, fig. 1)

Roman baths tend to be deeper: the two pools in the caldarium of the Baths of Neptune at Ostia, 1.09 n and 1.19 m; the three caldarium pools in the Forum Baths at Ostia: 1.21 m, 1.06 m, and 0.62 m; the men's caldarium in the Forum Baths at Pompeii, 0.6 m; the women's caldarium pool in the Stabian Baths at Pompeii: 0.61 m; the caldarium pool, Suburban Baths at Pompeii, 0.66 m. Cultrera (1938, 269) was struck also by the Syracuse pool's shallowness and designated it "un ampio ma non profondo canale."

2001] 415



GARRETT G. FAGAN

Fig. 14. Gela, Sicily. Hypocaust channels in the Greek Bath, southward view (above) and northward view (below). (N. De Haan)

DeLaine proposes a date for the Republican Baths in the mid second century B.C.52 DeLaine considers the system at Gela a departure from the arrangement at Gortys, since at Gela she identifies a "series of heating channels in parallel," which is an ante- cedent to the multiple heating-channel system found in the Republican Baths (figs. 14-15). But the Gela "series" of channels is, in reality, a minor variation on the familiar Greek annular system where a single heating channel has been replicat- ed to form a double channel (figs. 13-14). Further-

52 DeLaine 1989, 120. On the Gela bath's hypocaust, see Orlandini 1960,189-95. The fullest description of the Repub-

more, this duplex annular hypocaust is located di-

rectly opposite the entrances to the two bathing chambers fitted with hip-baths (rooms A and A' in

fig. 13), and one of the heating channels from this

admittedly curious double arrangement continues further south, entering a rectangular subterranean area (B in fig. 13) from which two channels emerge as vents. The excavator could not decide whether this whole hypocausted arrangement was a contin- uous paved sweat bath or a partially paved system intended to heat water for use in the hip-bath cham-

lican Baths at Pompeii remains Maiuri 1950.

416 [AJA 105




bers; no walls survive to help determine its function. To my mind, it looks like both, a minor variation on the near-contemporary system at Gortys, where an annular hypocaust heated both space and a water tank located outside a hip-bath room (see the box Y in fig. 13; cp. the Gortys hypocaust, de- scribed above, p. 405). In terms of form and function, the Gela bath clearly belongs to the bathing tradition represented at Gortys rather than that served by the Republican Baths at Pompeii. In addition, the relatively crude nature of the hypocaust system in the Republican Baths need indicate noth-

ing as to the earliness of its date but could have been generated by any number of intangible factors, such as cost-cutting or individual preference.53

What DeLaine has shown unquestionably, however, is that the technological potential for the de-

velopment of the Roman-style suspensura was

present in the cities of the Greek West. Moreover, these baths seem to offer varieties of bathing op- tions to their customers, a feature later common in Roman facilities. The physical remains, however, are insufficient to prove that the Western Greeks in the Hellenistic Age had moved toward the familiar Roman practices of communal immersion and

sequentially heated spaces. Instead, they appear to have developed and expanded techniques to facilitate their traditional bathing procedures.

THE INFLUENCE OF PRIVATE BATHS ON PUBLIC ONES

The archaeological record for private baths in

Italy has also been brought into the debate, on the reasonable assumption that private leisure habits

precede public ones. Seneca, in a famous passage (Ep. 86.4-12), describes the primitive simplicity of

Scipio Africanus's small bathing suite in his rustic villa and then rails against the bathing excesses of his own time, and Varro (Ling. 9.68) mentions the

simple "washing room" (lavatrina) as a predeces- sor to the fully-fledged private bath, or balneum.

53Fagan 1999a, 59-60. Note a fragmentary bath constructed in opus reticulatumat Tibur that features an apsidal room with a central wall dividing it into two halves, lengthwise. This arrangement of substructures seems to be another variant of the wall-supported hypocaust. No firm date can be assigned to the facility, but the "eta repubblicana" was proposed by Carducci (1940, 61-3). Similarly, Roman suspensurae in Britian (by definition of imperial date) can be of the channel variety, sometimes found side by side with the more familiar pillared form, as at the Chedworth villa (Goodburn 1979, 13-24), or even in combination with it in single rooms (Yegil 1992, 357- 61, who also cites other examples from Asia Minor and Gaul). These hypocausts attest variations in design that do not re- flect chronological or evolutionary relationships.

54 The major study remains Fabricotti 1976. The article has

_A Onm _^ m

Fig. 15. Pompeii. Groundplan of the Republican Baths. (After Nielsen 1993, 2:fig. 77)

Numerous country villas and townhouses in central Italy have therefore been the focus of scholarly attention.54 But there are several problems with us-

ing this type of evidence. Chief among them is the issue of variation in bathing habits and the deter- minative effect of personal taste on the form of private bathing suites. Initially, when Roman-style bath-

ing was not yet fully formed and alternatives were on hand from Greek sources, these factors make

discerning a scheme of development among private baths from different regions of Italy extremely difficult. Is the presence of a simple bathing suite a

sign of early date, or of the owner's personal pro- clivities or limited means, or all of the above? The

imprecise dating of several crucial examples of villa baths is also a genuine concern, since many date to the first century B.C. or A.D., by which time public bathing habits were well entrenched and the valence of influence is hard to determine.55 The

many flaws and is gradually being replaced by the work of N. De Haan (e.g., 1996, 1997, 2001). Note also the comments of Di Capua 1940, 124-38; Lafon 1991;Yegfil 1992, 50-5.

55 See, for instance, Fabricotti's discussion of simple private suites in villas such as those near S. Basilio in Nomentana (first century B.C.), at Centopiedi near Pompeii (late second/early first century B.C.), or at the Villa of Mysteries at Pompeii (late second century B.C.); see Fabricotti 1976, 32-3, 35, 37-8 respectively. In each case, the private bath suite dates to periods when Roman public bathing was already established (i.e., the second century B.C. and later), and this is true also of many of her other examples; seeYegil 1992,377. Similarly, simple two- roomed bathing suites may seem, on formal grounds, to predate more complex suites with three or more rooms. In fact, many bicameral suites are coeval with or later than more com-

2001] 417



GARRETT G. FAGAN

Fig. 16. Sperlonga. Villa Prato bathroom. (EFR, neg. SP1254)

evidence from the houses at Cosa reflects this vari-

ety in the archaeological record. Here, several houses of quite late date (early first century B.C.) still contained very simple bathing suites, comprised of a single room attached to the kitchen, when both

private and public bathing suites elsewhere in Italy were at this time displaying greater complexity and

pretension.56 Some recently investigated facilities shed partic-

ularly interesting light on the issue. At Vulci a villa

(the so-called House of the Cryptoporticus) was

equipped with a four-roomed bathing suite. Recent

arguments have redated the hypocaust in the sweat room (sudatorium) back from ca. 50 B.C. to 100

plex sets of bathing rooms; see Fabricotti 1976; Bruno and Scott 1993, 161-91 (on the House of the Birds at Cosa; Augustan). In the imperial period, lavish public baths appear to have influ- enced the form, scale, and decoration of private ones; see De- Laine 1999, 73. Netzer (1999) notes that, among the bath- houses built by Herod the Great in his palaces ca. 35-15 B.C., some were of the olderJudeao-Hellenstic style (with a minimal decoration and a characteristic stepped ritual bath) while others were of the fully Roman type. This constitutes a striking example of personal taste determining the form of private baths.

56 For instance, the House of the Skeleton and its neighboring house, with their kitchen/washroom arrangements, were dated by coin finds to the period 89-70 B.C.; see Bruno and Scott 1993,123-7 (House of the Skeleton), 153-8 (neigh- bor's property). The House of the Treasure acquired a simple

B.C., making it one of the earliest examples of the

system in a private bath anywhere in Italy, and the earliest in Etruria.57 Assuming the earlier date is

correct, this site demonstrates conclusively that private bathing rooms fully fitted with hypocausts could be found by 100 B.C. Less clear is the direct influence of the private habits of the rich (who owned the villas) on the form of the public baths used by all. While such examples as the Vulci villa may suggest a pioneering role for private baths, other ex-

amples confuse the issue. A villa unearthed at Ci-

ampino near Rome, and perhaps dating to the mid- second century B.C., contained a bathing suite with a communal pool but no hypocaust.58 The Villa Pra-

bathing room only in ca. 90 BC; see Bruno and Scott 1993,21, 81-97. Such basic facilities are more evocative of supposedly early Roman washrooms (Sen. Ep. 86.4-5) than the elaborate bathing suites found at other Italian sites, which are often contemporarywith and sometimes predate these Cosan examples (e.g., supra, n. 55).

57Broise andJolivet 1991, esp. 85-8. For the redating, see De Haan 1996. (My particular thanks go to N. De Haan for sending me the text of her paper and for several other relevant references.)

58De Rossi 1979, 64-5 (no. 89); Lafon 1991, 113-4. Broise (1994, 28) implicitly dates the Ciampino bath to the ca. 150- 100 B.C., while De Rossi (1979,65) postulates the first century B.C. This is a good example of the difficulty that prevails in dating many examples of private baths.

418 [AJA 105




to at Sperlonga, dated to the third quarter of the third century B.C., contains a most unusual bicameral bathing suite. One room, with a pool and traces of a labrum, appears to have been a cold-bath and

changing room. The other, apparently for hot bath-

ing, contained a single-person immersion tub along- side an unusual variant of the hip-bath called a "boot- bath," which on the basis of its shape and dimen- sions cannot have been used to wash more than the feet and lower legs. Both the immersion tub and the boot-bath formed a single architectural element, a sort of raised bathing platform into which the tubs were sunk and which was accessible by a single step from the floor level (fig. 16).59 The nearly contem-

porary date of this odd arrangement with the private hypocaust bath at Vulci or the non-hypocausted private suite at Ciampino both highlights the vari-

ety apparent in private suites and raises questions about how direct the influence of private baths was on public ones, since nothing comparable with the Villa Prato arrangement is discernible in the public record.60 That influences flowed between public and private facilities is beyond question, but their direction and extent need to be more fully investigated before reliable conclusions can be drawn.

One scholar has detected the influence of private baths in one detail of the Roman public bath's form: the linear sequence of rooms. In all the early private baths, the bathing rooms are located near the kitchen, often in a restricted space. As a result, bathing rooms with two or more elements tend to assume a simple row arrangement with respect to each other and the heat source in the kitchen. In our earliest public baths in the Vesuvian cities, the

bathing rooms are also in a simple linear alignment, conceivably derived from this domestic arrange-

59 Broise 1994, 26-7; Lafon 1991, 102-11. A terra-cotta tub similar to the Sperlonga "boot-bath" was found broken and reused in the House of the Birds at Cosa (and is now in the museum at the site). The tub belonged to an undetermined phase predating the extensive remodeling of the Augustan era; see Bruno and Scott 1993, 38-42. Given the evidence for other private baths at Cosa (supra n. 56), the boot-bath is unlikely to be any older than the early first century B.C. and thus appears to offer yet another example of contemporary variations in private bathing practices at a single site. Another relevant site is the private bath in the peristyle house at letas (Monte Iato) in western Sicily, which featured two rooms and a furnace. A single-person immersion tub and an emplacement for a labrum stood in one of the main rooms (cf. the arrangement at Sperlonga). Interestingly, in this facility the underfloor heating system for the immersion tub was discontinued in the third and final phase-the precise opposite ofwhat one would expect, and yet another indication of the impact of personal tastes or circumstances on the form of private baths. The date of the bath has been suggested as falling between the fourth and second century B.C.; see Dalcher 1994, 37-9.

ment.61 The suggestion is a plausible one but is vitiated by the very simplicity of so many early private baths: most have two elements and one cannot

really talk of a "row" of two. More importantly, the sequence of rooms in Roman baths does not seem to be the product of circumstance, as this suggestion holds, but rather a purposeful arrangement that met a need for varying grades of heated space and water. The origins of this need are far murkier and will be addressed below.

THE LITERARY AND EPIGRAPHIC EVIDENCE

In studies of the origins of Roman baths, the lit-

erary sources have tended to be treated only per- functorily while archaeological data have dominat- ed the discussion. Aside from the difficult issue of

Sergius Orata, the literary evidence is quite infor- mative, although there are certainly challenges in

interpreting it. Livy (23.7.3) provides one of our

very earliest references to public baths when he describes Hannibal suffocating the councilors at

Capua in a bathhouse in 216 B.C. Given Livy's research methods and the transferal of the tale to Nuceria by Dio (15.57.30) and Valerius Maximus (9.6 ext. 2), it seems possible that we are dealing with a later tale inserted into the tradition about Hannibal. Nevertheless, it is noteworthy that the anecdote locates early baths in Campania, a region later attested as a focal point of bathing culture (and excess) in the Italian peninsula. Plutarch puts an accusation of ill-discipline into the mouth of Bibu- lus, tribune in 209 B.C., who accuses Marcellus of dereliction of his command against Hannibal: Mar- cellus, rails Bibulus, has devoted himself to Cam- panian pleasures, hot baths prominently among them.62 Somewhat later, a speech attributed to C.

Given the form of the bath, a date not much earlier than ca. 100 B.C. is more likely; see Wilson 1990, 24-5.

60 Broise (1994, 26-8) sees things a little differently and argues that the Villa Prato immersion tub was inspired by those he identifies in Greek public baths in Sicily. Even by his own argument, however, the public examples were for collective immersion (which is crucial) and the Villa Prato tub for individual use. In all such cases, the interpretation of the physical record remains difficult and highly contentious, especially in detecting and tracing the causal connections between specific facilities across regions and over time.

61 For this observation, see Yegfil 1992, 63. For domestic examples, see Fabricotti 1976, 31-41.

62Plut. Marc. 26.4-27.2. For analogous sentiments about the ruinous effects of hot baths on military discipline (often set in Campania) see Livy 23.18.12; Plut. Mor. 785F; Dio 27.94.2, 62.6.4; SHA Comm. 11.5, Avid. Cass. 5.5, Pesc. Nig. 3.10, Alex. Sev. 53.2. Campania was also home to the notorious pleasure resort at Baiae, see Yegul 1996. Other thermal sites in the region also show clear evidence of development, see Crova 1953; Houston 1992.

2001] 419



GARRETT G. FAGAN

Gracchus locates baths in three Campanian towns, Gales, Ferentinum, and Teanum Sidicinum (Gell. NA 10.3.1-3). Thus the early Roman tradition ex-

plicitly associated public bathing with Campania and

adjacent municipia (in Latium Adiectum, for instance). It may well be that this tradition reflects the broader and later association of Campania with lux- urious living and excess, but that observation does not detract from the historical significance of the notices: later Romans nevertheless associated very early baths with these regions.

With Plautus we are on more solid ground. In several plays written at the end of the third and the beginning of the second century B.C., the play- wright refers to public baths and bathing as a regular part of the urban scene.63 While Plautus's de-

pendency on his Greek models might raise legit- imate concerns about these passages' applicabili- ty to Roman conditions, there are good grounds to believe that Plautus was relying on his audi- ence's familiarity with public baths to render his observations effective.64 A reference in Seneca

supports the conclusion that Plautus's public baths were real enough and not shadows cast by his Greek sources. Seneca (Ep. 86.10) comments that among the duties of aediles of old was the task of checking public baths for cleanliness and

suitably heated water. Seneca specifically mentions the Cornelii, Fabius Maximus, and Cato the Elder in this connection-all contemporaries of Plautus. These references reveal unequivocally that public baths were a regular feature of Rome's

cityscape by ca. 200 B.C. What type of bathhouse was Plautus alluding

to? Nielsen has suggested that one particularly il- lustrative passage in Plautus (Rud. 382-85) indicates they were Greek-style balaneia.65 The passage reads:

etiam qui it lavatum in balineas, cum ibi sedulo sua vestimenta servat, tamen surripiuntur, quippe qui quem illorum observet falsust; fur facile qui observat videt: custos qui fur sit nescit.

When a man goes to the public baths to wash, though he watches his clothes like a hawk, they're stolen all the same, as he doesn't know which of his fellow bathers to watch. The thief easily spots the watcher; the guard doesn't know who the thief is.

13 Plaut. Asin. 356-57, Persa 90-91, Poen. 703, Rud. 382-85, Trin. 405-8, Truc. 322-5.

4 Fagan 1999, 45-6; see also Fraenkel 1960; Harvey 1986;

Moore 1998, 50-66. 65Nielsen 1993, 1:27 n. 1 3. 66See Ter. Haut. 655, Eun. 592,596,600; Caec. fr. 98R; Non.

108M (155L), s.v. "ephippium." For discussion of these references, see Fagan 1999, 46-7.

67It is noteworthy that the comment in Trogus (a writer of

Nielsen argues that, since the bather takes his clothes with him into the bathroom, Plautus is de-

scribing a Greek-style balaneion where the bather

deposited his or her belongings in niches over the

hip-baths. But early Roman baths can also have niches in the bathing rooms and presumably, if one could not afford to hire a clothes guard, such niches would have been used to house the bathers' street clothes as well as their bathing gear. If the bather in Plautus's bath were attending to his ablu- tions in a hip-bath while his belongings sat in an overhead niche, would not the thief have to lean over the bather to reach the clothes? How, then, could the bather be unclear as to who the thief was? But if, instead, the bather is envisaged sitting in a communal pool or perspiring on a bench in a crowd- ed Roman-style tepidarium or caldarium equipped with storage niches, then Plautus's scenario makes perfect sense: the bather has to watch his clothes from a distance, uncertain who in the surrounding crowd harbors larcenous intent. Plautus's baths are Roman balnea, not Greek balaneia.

For the remainder of the second century we have merely bits and pieces of testimony. There are some references in Terence, a fragment of Caecilius Sta- tius, and a questionable passage cited by Nonius.66 Two references in later writers are of particular in- terest. Varro (Ling. 9.68) implies that the first baths in the city were introduced to Rome from elsewhere, and Pompeius Trogus, as epitomized byjustin (Epit. 44.2.6), refers to the Roman introduction of hot- water bathing to the Spanish following the Second Punic War. Assuming both of these statements are accurate and do not anachronistically represent later conditions, they show, first, that the Roman bath evolved elsewhere than at Rome and, second, that the habit was sufficiently well ingrained by 204 B.C. to allow the Romans to export it to Spain.67 The latter conclusion, particularly, is consistent with the regular appearance of public baths in the works of Plautus. The epigraphic record is of little use, since most Republican inscriptions pertaining to baths date to the first century B.C., when public baths were already well established in Italian com- munities.68 A notable exception is the late second- century B.C. inscription from Aletrium in the hills bordering southern Latium, which records the gen-

the Augustan Age) is evocative of Tacitus's acerbic observation in the Agricola (21.3) that the once hardy Britons were weakened by the pleasurable amenities of Roman civilization, including "porticoes, baths, and elegant dinner parties "(porticus et balinea et conviviorum elegantiam). The notion of intro- ducing enervating hot baths to hitherto stout native populations might therefore have emerged as a literary topos by Tro- gus's day.

8 See Fagan 1999, 53-4.

420 [AJA 105




erosity of a local benefactor in gracing the town with numerous functional structures. Among them is a lacus balnearius.69 Even if the precise meaning of this term remains unclear, the adjective balnearius sure-

ly suggests a familiarity among the people of Aletri- um with baths and bathing; and since the benefactor's other actions pertain to public structures, the lacus balnearius also must have served the local

populace. The text, then, offers epigraphic support for the presence of public baths bordering Rome by this date, but we can say nothing as to the

physical form of those baths, their construction

date, or when they first had been introduced to the

region. The literary evidence, sparse though it is, plays

an essential role in establishing the third century B.C. as a pivotal period in the early history of Ro- man public baths. By the end of that century, the Romans had already taken up the bathing habit in the capital; by the end of the following century it can be archaeologically attested in different regions of south and central Italy. The archaeological material would further suggest that the distinctively Roman form of bathhouse was already in existence

by the middle of the second century B.C., quite possibly earlier. The question then begs, how and where did this form evolve?

THE CASE FOR CAMPANIA

It was long ago suggested that Roman baths originated in Campania, a position recently supported forcefully by I. Nielsen.70 The chief basis for the

centrality of Campania is twofold: first, the tradition about Sergius Orata and his elusive pensiles balineae and, second, the presence there of volca-

nically heated pools and vents, especially in the

Campi Flegrei near Cumae. While the role of Ora- ta remains too uncertain to offer a concrete basis for argument, the case to be made from the Campi Flegrei is more cogent. Italian scholars of the early 20th century argued that local populations in Cam-

pania had used the natural pools for communal immersion bathing and the vents for steam-bath-

ing, and that a desire to recreate artificially such conditions generated the Roman-style bath.7" This

69 CIL 12.1529 = CIL 10.5807 = ILS5348 = ILLRP528 = Fagan 1999, 285-6 (no. 157). For the location of Aletrium, see Tal- bert 2000, 44 D2.

70Nielsen 1985; her position is recapped in ead. 1993, 1:20- 2. Nielsen's work builds on the observations of earlier scholars; see infra, n. 71.

71 Crova 1953; DiCapua 1940; Sgobbo 1929. It should be noted that the political climate surrounding Sgobbo and Di- Capua's work made arguing for an entirely native Italian origin for Roman baths advisable, although this observation does not in itself vitiate their arguments.

is essentially an evolutionary proposition: artificial-

ly-heated Roman baths grew out of the natural conditions occurring in Campania.

In recent years, however, there has been an ap- peal to move the focus of investigation away from

Campania. It has been argued, for instance, that

naturally heated pools and vents occur elsewhere than in Campania or that, as we have seen, the con- tributions of private baths in central Italy or public baths in Magna Graecia were crucial in determin-

ing the form of Roman baths.72 In a parallel devel-

opment, other leisure buildings traditionally ar-

gued to have originated in Campania have had their

origins reassigned to other locations.73 Many scholars of Roman baths today, therefore, downplay the

importance of Campania and turn to other sources as critical to the genesis of the Roman public bath.

In my view, however, the case for Campania remains the strongest. Three factors in particular make it so. First, most of our earliest evidence for Roman baths points to Campania. Here are found our earliest physical remains (at Cumae, ca. 180

B.C., and the facilities in the towns buried by Ve-

suvius), and the region provides the setting from

many early literary references to public baths (419- 21). Even if the plays of Plautus suggest that there were public baths in Rome (or at least in an un-

specified Roman urban setting) by 200 B.C., it will be remembered that Varro (Ling. 9.68) expressly states that public baths were introduced to the city from without. At the very least, the focus of so much of the earliest material on Campania needs to be

respected. Second, wider contextual considerations make

Campania an excellent candidate for the appearance of the Roman public bath. Campania in the third and second centuries B.C. was a rich and prosperous place. Many of Pompeii's first grand houses, such as the House of the Faun or House of the

Figured Capitals, were constructed during this

period and several major public buildings were added to the city's fabric, such as those clustering around the theater at the southern end of the city. But Pompeii was not alone in prosperity: other sites, too, reveal the region's wealth.74 The region was also

72 De Haan 1996; DeLaine 1989; Yegul 1992, 48-91, 377-80. 73 See the ingenious arguments of Welch (1994; forthcom-

ing) for the non-Campanian sources of the amphitheater or the basilica respectively. See also the perceptive observations of Gros 1996, 235-44 (on basilicas) and 317-23 (on amphi- theaters).

74Richardson 1988, 67-127; Zanker 1998, 32-60. On the history of Campania in this era, see Cerchiai 1995, 195-224; Frederiksen 1984, 221-84; Pugliese Carratelli 1991, esp. 151- 91 (P. Somella on urbanization) and 193-233 (F. De Martino on economy and society).

421 2001]



GARRETT G. FAGAN

culturally diverse and vibrant in the years before and after 200 B.C. Greeks, Oscans, and Romans all enjoyed a permanent presence in the area, and

exposure to more far-flung cultures came daily through Campania's many ports. An apt illustration of this cultural interaction comes from the House of the Faun at Pompeii. Here, the affluent Oscan owner of the property adorned one room of his palatial house with a magnificent mosaic. The subject he chose was thoroughly Greek, al-

though set in the Near East: a victory of Alexander's over Darius III, which was copied in tesserae from a painting of ca. 300 B.C. by Philoxenos of Ere- tria.75 The introduction into Campania of foreign cultural influences and the subsequent folding of those influences into local tastes could not be clearer. Although not specifically a Campanian phenomenon, concrete had also been introduced as a building material by ca. 200 B.C.76 The importance of this substance for the construction of Ro-

man-style baths needs little elucidation. Without it, the erection of a series of barrel-vaulted chambers of the sort that characterizes surviving exam-

ples of early Campanian baths would have been far more difficult. All of the preceding factors found public expression in new architectural forms that are first found in Campania, such as the basilica and the amphitheater. In my view, the Ro- man public bathhouse can be added to the list. Recent objections notwithstanding, Campania still offers the first clear examples of each type, even if further levels of inspiration can be detected earlier and elsewhere.77 In this sense, Campania pro-

75 Zanker 1998, 33-43; see also Cohen 1997. 76 See Adams 1994, 79-81; Anderson 1997, 145-51; Boeth-

ius 1978, 126-9; Lamprecht 1996. The observations of Blake (1947, 324-30) remain instructive. It is not possible to locate the discovery of concrete in any particular region of Italy. The substance has long been thought to have arrived in Rome by 174 B.C., as represented in the remodeled Porticus Aemilia along the bank of the Tiber (Gatti 1934); however, a recent reevaluation of this structure (Tuck 2000) proposes a later date for it, in the Sullan era. The Temple of Magna Mater on the Palatine (constructed 204-191 B.C.) may well represent the earliest appearance of concrete in the city; see LTUR 3.206-8 (P. Pensabene).

77Even if, as Welch argues (supra, n. 73), the amphitheater had its ultimate roots in Rome or the basilica in Hellenistic palaces, there is no reason to suppose thatwhat might be termed the definitive form of each genre-that commonly designated "Roman"- did not first appear in Campania. The evidence would still suggest that it did.

78Futrell (1997, 33-44) comes to an analogous conclusion about the origins of the stone amphitheater: it appeared in prosperous and innovatory Campania but under the influence of Roman colonists. Working fromJouffroy (1986), she notes (40) that 128 of 310 known public buildings in Republican Italy

vided the context for the culmination of a devel-

opmental process that stretched back chronologi- cally before the third century and extended geo- graphically farther afield than the region itself. The evolution of complex architectural genres is

unlikely to display a simple linear profile, so that the unique position of third and second century B.C. Campania as a prosperous and innovative cultural confluence offers all the more reason to look to it as the progenitor of the Roman bath's

peculiar form, itself clearly the product of multi-

ple influences.78

Finally, as earlier scholars emphasized, Campa- nia offers ready explanations for the chief charac- teristics of the Roman public bath: the communal

pools and the heated sequential spaces. While hard evidence for early Italic natural bathing practices is lacking, attention has focused especially on the monumental and confusing leisure com-

plex at Baiae. Although the visible remains date to the Augustan period and later, the hillsides at Baiae are riddled both with natural caverns and man-made tunnels and chambers fashioned for

sweating and bathing. That local populations had

long used such subterranean places in an unde-

veloped state is surely beyond doubt; certainly Medieval manuscript illustrations of Peter of Ebo- li's De Balneis Puteolanis-themselves possibly derived from now-lost classical depictions-clearly show people bathing in natural caverns.79 Later classical authors also commented explicitly on the existence of these natural phenomena and their use by locals.80

are located in Campania, strong testimony for that region's prosperity and architectural vitality.

79Yeguil 1996,142 (caverns) and 148-55 (illustrations). The manuscript depictions of the Balneum Gimborosum and Bal- neum Orti Donici (ibid., figs. 25-26) are particularly noteworthy; note also the Balneum Foris Cryptae, Balneum Cryptae Palumbariae seu Sybillae, Balneum Colme, and Balneum Spe- luncae (see the relevant entries in Di Bonito and Giamminelli 1992). Physical evidence for early human activity in the region, never mind the exploitation of the Campi Flegrei, is scat- tered, and the literary sources couched in myth; see Adinolfi 1982; Giacomelli and Scandone 1992, 16-23; and next note.

80Dio 48.50-51; Livy 41.16.3-4; Vitr. De arch. 2.6.2. Note also Celsus Med. 2.17.1; Strabo 5.2.9 and 5.4.5. In the latter notice, Strabo cites the historian Ephorus (ca. 405-330 B.C.) in reporting that the first inhabitants of the region, the Cim- merians, lived entirely underground. Ephorus may well here be transmitting local traditions about age-old human use of the caverns in the region and possibly reflecting practices of his own day. Note also the tradition that Odysseus's steersman Baios was buried here, giving rise to the name "Baiae" (e.g., Lycoph. Alex. 688-91; Strabo 5.4.6); the notice suggests at least an awareness of Baiae's attractions among the learned of the Hellenistic Age.

[AJA 105 422




The establishment of local bathing habits cen- tered on communal immersion in hot water would have been an obvious corollary of such conditions.

Furthermore, the putative use of caverns for sweat-

ing may well have generated the need for gradations of heated space, whereby an original, nonarti-

ficially-heated bathing ritual entailed exposure to steam in different degrees. In other words, the Cam-

pi Flegrei offered the natural conditions required to stimulate the development of existing technolo-

gies to fit specific, long-established needs. On a broad perspective, this order of events also makes the most sense, since the ancients rarely if ever modified technologies without a reason to do so. The Campanian desire to reproduce their natural

bathing habits artificially offers a likely stimulus for the extension of the Greek "annular" hypocaust into a full suspensura. The Greek-style hypocaust, used from the outset to heat both water and space, was thus elevated into the suspensura primarily out a desire not to heat communal pools (which it could do already), but to recreate the conditions of a steam-cavern. If this were so, then it was the established bathing habits of the Campanians that provided the crucial factor in bringing the Roman-style bath to its final fruition.

The reconstruction proposed above is necessar-

ily speculative. It is true, as some have pointed out, that hot pools and steam vents occur outside the Campi Flegrei, or that other possibilities exist for the evolution of the Roman sequence of bath-

ing rooms.81 What makes the case for Campania compelling, however, is its cumulative cogency, not

any one element of it. Thus, although the Campa- nians may have wanted to heat spaces in accor- dance with their established bathing practices, they did not the need hypocausts to do it; braziers would have sufficed, and did in certain facilities. However, the presence in the region of hypocausted Greek baths offered alternatives, which, if

adapted, proved far more effective, and new build-

ing technologies facilitated the entire endeavor.

81 Supra, n. 72. Various reasons for the development for the sequence of rooms can be speculated, such as the desire to make use of leftover gases from a single heated room (caldarium) led to the addition of a second, lesser-heated room (tepidarium); or the structural advantages of using a series of chambers as mutually-buttressing barrel-vaults; or by appealing to the "natural" orderliness of Roman architecture. All are essentially "accidental" explanations and are not particularly convincing: it is not immediately clear why a series of chambers intended to buttress barrel vaults should be sequentially heated, nor why leftover gases had to be used instead ofvented (as they were in Greek hypocausted baths), nor how "natural" orderliness generates heated spaces. It

A combination of factors generated the suspensura, not any single requirement or circumstance.

Campania provides all the conditions conducive to the development of the Roman public bath, and at the very time suggested by the literary sources as critical for the appearance of that type of facili-

ty. Here was a wealthy and vibrant place, standing at a cultural crossroads not only of Italy but of the wider Mediterranean, displaying signs of architectural innovation, and boasting the local natural conditions to stimulate the move toward the de-

velopment of the Roman-style bath. All of these factors combined make the case for Campania, despite its necessarily circumstantial nature, the

strongest on available evidence; unless a more via- ble alternative can be found, it must remain our best option.82

My reassertion of the centrality of Campania in the development history of the Roman-style public bath, it must be stressed, does not entail the abandonment or even the less vigorous pursuit of other lines of inquiry. The region in the third and second centuries B.C., as a crucible of cultural interaction, merely represents the final stage of an

evolutionary sequence that surely has roots extend-

ing much farther afield and deeper into the past. Whereas earlier scholars attributed the Roman bath in its entirety to the region, it is argued here that Campania offered the context for its final

appearance. The Roman public bath did not "orig- inate" anywhere-it was generated by various circumstances, all of which are locatable in and around Campania. Narrowly focused claims that the origins of Roman baths lie specifically with the Greeks or with the private baths of Italy ought therefore to be replaced in favor of a more em-

bracing search for diverse precursors. Previous

unitary claims have all been found wanting as ex-

planations for the appearance of the Roman bath in and of themselves although, in most instances, threads of influence can be detected. It is here

argued that conditions in third- and second-cen-

seems far more likely that the bathers' need for gradations of heat came first, and the sequence of heated rooms re- flected that need.

82 One line of inquiry that might throw more light on this question is that of third- and second-century water supply and irrigation systems in Campania. It is true that most simple baths did not need aqueduct water and could function effectively from wells or cisterns (see Fagan 1999a, 73 n. 106), but nevertheless investigation of Campanian hydraulics in the crucial period could yield pertinent results. For some recent work on the subject of Campanian water supply (mostly dealing with later periods), see De Haan andJan- sen 1996; Ohlig 2000 (non vidi).

423 2001]



GARRETT G. FAGAN

tury B.C. Campania brought those threads togeth- er. Even if Campania was not necessarily the birth-

place of every facet of the Roman public bath, it was surely the nursery of its final form.

CLASSICS AND ANCIENT MEDITERRANEAN

STUDIES AND HISTORY

PENN STATE UNIVERSITY

108 WEAVER BUILDING

UNIVERSITY PARK, PENNSYLVANIA 16802-5500

[email protected]

Works Cited

Adams, J.-P. 1994. Roman Building: Materials and Tech- niques. London: Batsford.

Adinolfi, R. 1982. I Campi Flegrei nella preistoria. Naples: Massimo.

Aleshire, S. 1989. The Athenian Asklepieion: The People, their Dedications, and the Inventories. Amsterdam: J.C. Gieben.

. 1991. Asklepios at Athens: Epigraphic and Prosopo- graphic Essays on the Athenian Healing Cults. Amster- dam: J.C. Gieben.

Allen, H.L. 1974. "Excavations at Morgantina (Serra Orlando) 1970-1972." AJA 78:368-83.

Anderson,J.C. 1997. Roman Architecture and Society. Balti- more:Johns Hopkins University Press.

Benedum,J. 1967. "Die Balnea Pensilia des Asklepiades von Prusa." Gesnerus 24:93-107.

Blake, M.E. 1947. Ancient Roman Construction in Italy from the Prehistoric Period to Augustus. Carnegie Institution of Washington Publication 570. Washington, D.C.: Carnegie Institution of Washington.

Boethius, A. 1978. Etruscan and Early Roman Architecture. 2nd ed. Harmondsworth: Penguin.

Boudon, V. 1994. "Le r6ole de l'eau dans les prescriptions medicales d'Asclepios chez Galien et Aelius Aris- tide." In L'eau, la sante et la maladie dans le monde grec, edited by R. Ginouves, A.-M. Guimier-Sorbets, J. Jouanna, and L. Villard, 157-68. BCH Suppl. 28. Par- is: de Boccard.

Brodner, E. 1992. Die romischen Thermen und das Antike Badewesen: Eine kulturhistorische Betrachtung. 2nd ed. Darmstadt: Wissenschaftliche Buchgesellschaft.

Broise, H. 1994. "La pratique du bain chaud par immersion en Sicile et dans la peninsule italique a l'epoque hellenistique." Xenia Antiqua 3:17-32.

Broise, H., and V. Jolivet. 1991. "Le bain en Etrurie i l'epoque hellenistique." In Les thermes romains: Actes de la table ronde organis&e par l'Ecole franfaise de Rome (Rome, 11-12 novembre 1988), Ecole francaise de Rome, 79-95. CEFR 142. Rome: Ecole francaise de Rome.

Bruno V.J., and R.T. Scott. 1993. Cosa. Vol. 4, The Houses. University Park: Penn State University Press.

Burford, A. 1969. The Greek Temple Builders at Epidau- ros: A Social and Economic Study of Building in the Askl- epian Sanctuary, During the Fourth and early Third Centuries B. C. Liverpool: Liverpool University Press.

Carducci, C. 1940. Tibur: Regio IV, Sabini et Samnium. Italia romana: municipi e colonie, ser. 1, v. 3. Rome: Istituto di studi romani.

Cerchiai, L. 1995. I Campani. Biblioteca di archeologia 23. Milan: Longanesi.

Cohen, A. 1997. The Alexander Mosaic: Stories of Victory and Defeat. Cambridge: Cambridge University Press.

Crova, B. 1953. "Le terme romane nella Campania." Atti dellVIII covengo nazionale di storia dell'architettura:271-88.

Cultrera, G. 1938. "Siracusa - rovine di un antico stabil- imento idraulico in contrada Zappala." Nsc:261-301.

Dalcher, K. 1994. Das Peristylhaus 1 von Iaitas: Architektur und Baugeschichte. Studia letina 6. Zurich: E. Rentsch.

De Haan, N. 1996. "The Significance of Roman Private Baths for the Development of Public Baths." Paper read at the Second International Conference on Ancient Baths, 22-27 April, Varna.

. 1997. "Nam nihil melius esse quam sine turba lavari: Privatbader in den Vesuvstadten." Papers of the Nether- lands Institute in Rome 56:205-26.

. 2001. "Si aquae copia patiatur: Pompeian Pri- vate Baths and the Use of Water." In Water Use and Hydraulics in the Roman City, edited by A.O. Koloski- Ostrow, 41-9. AIA Colloquia and Conference Papers 3. Dubuque: Kendall/Hunt.

De Haan, N., and G.M.C. Jansen, eds. 1996. Cura Aquarum in Campania: Proceedings of the Ninth Interna- tional Congress on the History of Water Management and Hydraulic Engineering in the Mediterranean World. Bulle- tin Antieke Beschaving Suppl. 4. Leiden: BABesch.

DeLaine,J. 1988. "Recent Research on Roman Baths." JRA 1:11-32.

. 1989. "Some Observations on the Transition from Greek to Roman Baths in Hellenistic Italy." Med- itArch 2:111-25.

.. 1992. "New Models, Old Modes: Continuity and Change in the Design of Public Baths." In Die romische Stadt im 2. Jahrhundert n. Chr: Der Funktionswandel des offentlichen Raumes: Akten des Kolloquiums Xanten 1990, edited by H-J. Schalles, H. von Hesberg, and P. Zank- er, 257-75. Cologne: Rheinland.

1993. "Roman Baths and Bathing."JRA 6:348-58. - . 1997. The Baths of Caracalla: A Study in the Design,

Construction, and Economics of Large-scale BuildingProjects in Imperial Rome. JRA Suppl. 25. Portsmouth, RI:Jour- nal of Roman Archaeology.

. 1999. "Benefactions and Urban Renewal: Bath Buildings in Roman Italy." In Roman Baths and Bath- ing: Proceedings of the First International Conference on Roman Baths Held at Bath, England, 30 March-4 April, 1992. 2 pts., edited byJ. DeLaine and D.E.Johnston, 67-74. JRA Suppl. 37. Portsmouth, RI: Journal of Roman Archaeology.

DeLaine, J., and D.E.Jonhston. 1999. Roman Baths and Bathing: Proceedings of the First International Conference on Roman Baths Held at Bath, England, 30 March-4 April, 1992. 2 pts.JRA Suppl. 37. Portsmouth, RI: Journal of Roman Archaeology.

De Rossi, G.M. 1979. Bovillae. Forma Italiae: Topografi- ca antica, regio 1, v. 15. Florence: Olschki.

Di Bonito, R., and R. Giamminelli. 1992. Le Terme di Cam- pi Flegrei: Topografia storica. Milan: Jandi Sapori.

Di Capua, F. 1940. "Appunti su l'origine e sviluppo delle terme romane." RendNap 20:81-160.

Ecole franqaise de Rome. 1991. Les thermes romains: Actes de la table ronde organisee par 1'Ecole francaise de Rome (Rome, 11-12 novembre 1988). CEFR 142. Rome: Ecole francaise de Rome.

[AJA 105 424




Eschebach, H. 1970. Die Stddtbauliche Entwicklung des antiken Pompeji mit einem Plan 1:1000 und einem Exkurs: Die Baugeschichte der Stabianer Thermen nach H. Sulze. Heidelberg: F.W. Kerle.

. 1979. Die Stabianer Thermen in Pompeji. Berlin: de Gruyter.

Fabricotti, E. 1976. "I bagni nelle prime ville romane." CronPomp 2:29-111.

Fagan, G.G. 1996. "Sergius Orata: Inventor of the Hy- pocaust?" Phoenix 50:56-66.

. 1999a. Bathing in Public in the Roman World. Ann Arbor: University of Michigan Press.

---. 1999b. "Interpreting the Evidence: Did Slaves Bathe at the Baths?" In Roman Baths and Bathing: Pro- ceedings of the First International Conference on Roman Baths Held at Bath, England, 30 March-4 April, 1992. 2 pts., edited byJ. DeLaine and D.E. Johnston, 25-34. JRA Suppl. 37. Portsmouth, RI: Journal of Roman Archaeology.

Farrington, A. 1999. "The Introduction and Spread of Roman Bathing in Greece." In Roman Baths and Bath- ing: Proceedings of the First International Conference on Roman Baths Held at Bath, England, 30 March-4 April, 1992. 2 pts., edited byJ. DeLaine and D.E.Johnston, 57-65. JRA Suppl. 37. Portsmouth, RI: Journal of Roman Archaeology.

Fracchia, H. 1999. "Western Lucania, Southern Samni- um, and Northern Apulia: Settlement and Cultural Changes, Fifth-Third Centuries B.C." Paper read at the Third E. Togo Salmon Conference, 1-2 October, Hamilton.

Fraenkel, E. 1960. Elementi Plautini in Plauto. Florence: "La nuova Italia" Editrice.

Frederiksen, M. 1984. Campania. London: British School at Rome.

Fulford, M., and A. Wallace-Hadrill. 1998. "Unpeeling Pompeii." Antiquity 72:128-45.

Futrell, A. 1997. Blood in the Arena: The Spectacle of Roman Power. Austin: University of Texas Press.

Gatti, G. 1934. "Saepta Iulia e Porticus Aemilia nella forma severiana." BullCom 62:135-44.

Giacomelli, L., and R. Scandone. 1992. Campi Flegrei, Campania Felix: II Golfo di Napoli tra storia ed eruzioni. Naples: Liguori.

Ginouves, R. 1959. L'etablissement thermal de Gortys d'Arcadie. Paris: J. Vrin.

-. 1962. Balaneutike: Recherches sur le bain dans l'antiquite grecque. BEFAR 200. Paris: de Boccard. --. 1994. "L'eau dans les sanctuaires medicaux." In L'eau, la sante et la maladie dans le monde grec, edited by R. Ginouves, A.-M. Guimier-Sorbets,J.Jouan- na, and L. Villard, 237-46. BCHSuppl. 28. Paris: de Boccard.

Ginouves, R., A.-M. Guimier-Sorbets,J.Jouanna, and L. Villard, eds. 1994. L'eau, la sante et la maladie dans le monde grec: Actes du colloque organise a Paris (CNRS et Fondation Singer-Polignac) du 25 au 27 novembre 1992 par le Centre de recherche "Archeologie et systemes d 'information " et par I'URA 1255 "Medecine grecque. "BCH Suppl. 28. Paris: de Boccard.

Goodburn, R. 1979. Chedworth Roman Villa. London: National Trust.

Gros, P. 1996. L'architecture romaine du debut du IIIe siecle av. J.-C. a la fin du Haut-Empire. Vol. 1, Les monuments publics. Paris: Picard.

Gruen, E.S. 1984. The Hellenistic World and the Coming of Rome. 2 vols. Berkeley: University of California Press.

Harvey, P.B.,Jr. 1986. "Historical Topicality in Plautus." CW 79:297-304.

Heinz, W. 1984. Romische Thermen: Badewesen und Bade- luxus im romischen Reich. Munich: Hirmer.

Hilton Turner,J. 1947-1948. "Sergius Orata, Pioneer of Radiant Heating." CJ43:486-7.

Houston, G.W. 1992. "The OtherSpas of Ancient Campa- nia." In The Two Worlds of the Poet: New Perspectives on Vergil, edited by R.M. Wilhelm and H.Jones, 356-70. Detroit: Wayne State University Press.

loppolo, G. 1992. Le terme del Sarno at Pompei: Iter di un 'analisi per la conoscenza, il restauro e la protezione sis- mica del monumento. Rome: L'ERMA di Bretcschneider.

Jackson, R. 1988. Doctors and Diseases in the Roman Em- pire. Norman: University of Oklahoma Press.

Jouffroy, H. 1986. La construction publique en Italie et dans l'Afrique romaine. Strasbourg: AECR.

Koloski-Ostrow, A.O. 1990. The Sarno Bath Complex. Rome: L'ERMA di Bretschneider.

-. Forthcoming. "The City Baths in Pompeii and Herculaneum." In Pompeii and the Ancient Settlements under Vesuvius, edited by P. Foss andJJ. Dobbins. Lon- don: Routledge.

Kunze, E., and H. Schleif. 1944. IV Bericht iiber die Aus- grabungen in Olympia. Berlin: de Gruyter.

Ladstatter, G. 1995. "Das sog. 'Griechische Hypocaus- tenbad' in Zeusheiligutm von Olympia: Eine Neube- trachtung in Verbindung mit fruihen italischen Ther- menarchitektur." Paper read at the conference "Constructions publiques et programmes edilitaires en Grece du IIe S. av.J.-C. au Ier S. ap.J.-C.," 14-17 May, Athens.

Lafon, X. 1991. "Les bains prives dans l'Italie Romaine au IIe siecle av. J.-C." In Les thermes romains: Actes de la table ronde organisee par 'Ecole franfaise de Rome (Rome, 11-12 novembre 1988), Ecole francaise de Rome, 97- 114. CEFR 142. Rome: Ecole francaise de Rome.

Lambrinoudakis, V. 1994. "L'eau medicale a Epidaure." In L'eau, la sante et la maladie dans le monde grec, edited by R. Ginouves, A.-M. Guimier-Sorbets, J. Jouanna, and L. Villard, 225-36. BCH Suppl. 28. Paris: de Boc- card.

Lamprecht, H.-O. 1996. Opus Caementitium: Bautechnik derRomer. 5th ed. Dusseldorf: Beton.

Ling, R. 1981. Review of Die Stabianer Thermen in Pompeji, by H. Eschebach. JRS 71:213-4.

Maiuri, A. 1950. "Scoperta di un edificio termale nella Regio VIII, Insula 5, Nr. 36." Nsc: 11 6-36.

Mallwitz, A. 1972. Olympia und seine Bauten. Munich: Prestel.

Manderschied, H. 1988. Bibliographie zum romischen Badewesen unter besonderer Beriiksichtigung der offentlichen Thermen. Munich: Drucken.

Mau, A., and F.W. Kelsey. 1907. Pompeii: Its Life and Art. Rev. ed. New York: MacMillan.

Merten, E.W. 1983. Bider und Badegepflogenheiten in der Darstellung der Historia Augusta. Bonn: Habelt.

Moore, T.J. 1998. The Theatre of Plautus. Austin: Univer- sity of Texas Press.

Netzer, E. 1999. "Herodian Bath-Houses." In Roman Baths and Bathing: Proceedings of the First International Confer- ence on Roman Baths Held at Bath, England, 30 March-4 April, 1992. 2 pts., edited by J. DeLaine and D.E.

2001] 425



G.G. FAGAN, THE GENESIS OF THE ROMAN PUBLIC BATH

Johnston, 45-55.JRA Suppl. 37. Portsmouth, RI:Jour- nal of Roman Archaeology.

Nielsen, I. 1985. "Considerazioni sulle prime fasi dell'evoluzione dell'edificio termale romano." Anal- Rom 14:81-112.

. 1993. Thermae et Balnea: The Architecture and Cul- tural History of Roman Public Baths. 2 vols. 2nd ed. Aar- hus: Aarhus University Press.

Ohlig, C. 2000. De Aquis Pompeiorum: Die Herkunft, Zulei- tung und Verteilung des Wassers im Castellum Aquae. Cir- cumvesuviana 4. Amsterdam: J.C. Gieben.

Orlandini, P. 1960. "Gela: Impianto greco di bagni pub- blici presso l'Ospizio." NSc: 181-211.

Pasquinucci, M. 1987. Terme romane e vita quotidiana. Modena: F.C. Panin Editore.

Pugliese Carratelli, G., ed. 1991. Storia e civiltd della Cam- pania. Vol 1, L'evo antico. Naples: Electa.

Richardson, L.,jr. 1988. Pompeii: An Architectural History. Baltimore:Johns Hopkins University Press.

Salmon, E.T. 1967. Samnium and the Samnites. Cambridge: Cambridge University Press.

Sgobbo, I. 1929. "Terme flegree ed origine delle terme romane." AttiCStR 1:186-94.

Tagliamonte, G. 1996. I Sanniti: Caudini, Irpini, Pentri, Carricini, Frentani. Biblioteca di archeologia 25. Mi- lan: Longanesi.

Talbert, R.J.A., ed. 2000. Barrington Atlas of the Greek and Roman World. Princeton: Princeton University Press.

Tuck, S.L. 2000. "A New Identification for the 'Porticus Aemilia."'JRA 13:175-82.

Weber, M. 1996. Antike Badekultur. Munich: C.H. Beck. Welch, K. 1994. "The Roman Arena in Late-Republican

Italy: A New Interpretation." JRA 7:59-80. . Forthcoming. "Explaining the Origin of the

Roman Basilica." Wikander, 0. 1996. "Senators and Equites VI: Caius

Sergius Orata and the Invention of the Hypocaust." OpRom 20:177-82.

Wilson, A.J.N. 1966. Emigration from Italy in the Republi- can Age of Rome. Manchester: Manchester University Press.

Wilson, R.J.A. 1990. Sicily under the Roman Empire: The Archaeology of a Roman Province, 36 BC-AD 535. Warm- inster: Aris and Phillips.

Yegul, F. 1981. Review of Die Stabianer Thermen in Pompeji, by H. Eschebach. AJA 85:506-7.

. 1992. Baths and Bathing in ClassicalAntiquity. Cam- bridge, Mass.: MIT Press.

. 1996. "The Thermo-Mineral Complex at Baiae and de Balneis Puteolanis." ArtB 78:137-61.

Zanker, P. 1998. Pompeii: Public and Private Life. Cam- bridge: Harvard University Press.

426



the genesis of the roman public bath: recent … genesis of the roman public bath: recent approaches...

Documents