Geology and Architectural Conservatism in the Hellenistic Fountains of Corinth and Perachora
p. 51-64
Résumés
La géologie des terrasses naturelles sur lesquelles est installée la ville antique de Corinthe favorise l’essor d’un type de fontaine bien spécifique: la fontaine Peirènè en est l’exemple le mieux connu. Celle-ci est une réponse directe et évolutive aux conditions géologiques locales… L’imitation de Peirénè par plusieurs fontaines corinthiennes à l’époque hellénistique peut être interprétée comme la marque d’un conservatisme architectural délibéré, voire comme un souci d’antiquaire, dans la mesure où des éléments de la fontaine la plus ancienne étaient copiés même dans des environnements géologiques où ils ne présentaient plus aucun avantage fonctionnel. Parmi les exemples, on citera la fontaine Glaukè, le complexe de l’Asklépiéion de Corinthe, et surtout la fontaine de Pérachora qui applique le plan et les caractéristiques de Peirènè dans un contexte complètement inadapté. Nombre de particularités de la fontaine de Pérachora peuvent s’expliquer par la volonté des architectes de dupliquer autant que possible la forme et les techniques utilisées pour Peirènè, malgré leur inadaptation aux conditions géologiques locales.
The geology of the natural terraces on which the city of ancient Corinth stands led to the development of a distinctive type of fountain house, of which the fountain of Peirene is the best-known example. Peirene evolved in direct response to the local geological conditions, and as the city’s oldest and most celebrated water source, it became an architectural template for other Corinthian fountains. The imitation of Peirene in several Corinthian fountains of the Hellenistic period can be seen as deliberate architectural conservatism or antiquarianism, as elements of the older fountain were faithfully duplicated even in geological settings where they no longer conveyed any practical advantage. Examples include the fountain of Glauke and the fountain complex of the Asklepieion at Corinth itself, and above all the fountain house at Perachora, which attempted to transfer the design of Peirene to a completely inappropriate setting. Many of the puzzling features of the fountain at Perachora can be explained by the assumption that the builders were intent on duplicating as closely as possible the form and construction techniques used in Peirene, however unsuited they might be to the actual geological conditions of the site on which they chose to build.
Texte intégral
1On Sunday, the 2nd of July, 1911, William Dinsmoor Sr., then in his fourth season of work at the excavations conducted by the American School of Classical Studies at ancient Corinth, joined several other members of the excavation staff in an excursion across the gulf to Perachora.1 The party, which consisted of Dinsmoor and his wife, together with Carl Blegen, Allan Chester Johnson, and Clyde Pharr, hired a boat in modern Corinth and arrived at the small cove near the end of the Perachora peninsula by mid-morning. In the Heraion valley itself, twenty years before the beginning of British excavations at the site, they found a few traces of walls and large piles of Protocorinthian sherds turned up by the local villagers in their search for marketable antiquities, but little else in the way of ancient remains. After picnicking among the trees and investigating the chapel of Ayios Nikolaos, the party split up. Dinsmoor and his wife made their way to the small plain northeast of the Heraion, where they discovered a manhole leading to a series of long, narrow underground galleries lined with excellent Greek hydraulic cement.
2Dinsmoor’s notes and drawings constitute the earliest known description of the fountain house at Perachora. His tunnel-like subterranean galleries were in fact the reservoirs that lay behind a series of draw basins and a graceful Ionic façade, to be dated, probably, in the late IVth century or early IIIrd century BC (Fig. 1, top). None of this was known at the time, for it was not until British investigation of the western end of the building was renewed in the 1960s that the true purpose of the reservoirs was discovered and the fountain house as a whole received its first publication.2
3 Although Dinsmoor’s knowledge of the fountain house at Perachora was incomplete, he immediately noticed one of its most interesting features: the striking resemblance of its storage system to that of the fountain of Peirene at Corinth, which had been discovered and excavated a little more than a decade earlier (Fig. 1, bottom).3 In both cases the reservoirs take the form of a series of tunnel-like rock-cut chambers, constructed not by digging down from the surface but by cutting back into the face of a natural escarpment, in front of which were built the draw basins fed by the reservoirs and the decorative façade of the building itself.
Fig. 1 – Plans of the fountain house at Perachora (top) and the fountain of Peirene at Corinth (bottom), drawn to the same scale. After Tomlinson 1969, fig. 19 (top) and Corinth I.6, pl. II (bottom).

4The formal resemblance of the fountain house at Perachora to Peirene and other Corinthian waterworks has often been remarked. None of the published discussions, however, has fully examined the implications of that similarity, and none of them adequately brings out the differences which lie beneath the superficial resemblance, or the awkward practical compromises by which the resemblance was effected, or the inappropriateness of certain aspects of the design to the fountain’s physical setting in the Perachora plain.
5To understand what an aberrant example of its type the Perachora fountain house really is, one needs to keep in mind some important facts about the origin and development of the type at Corinth, and in particular the powerful influence exerted on that development by the local geological conditions. The city of ancient Corinth stands upon a series of natural terraces that descend northward from the citadel of Acrocorinth to the broad coastal plain bordering the Corinthian Gulf. These terraces have been interpreted as ancient shorelines elevated above sea level by the gradual uplift of the southern coast of the Corinthian Gulf. They are composed of relatively recent marine and lacustrine sediments deposited in cycles along the shore of this ancient sea, and as a result of the cyclical nature of the processes by which they were formed, the geological stratigraphy of each terrace, while not identical, is closely analogous: in each case the same types of strata with the same characteristic properties occur in the same relative positions.4
6Underlying the entire area is an enormous bed of clay or marl, most of it laid down in the Pliocene, some 10 to 20 million years ago. In the immediate vicinity of the ancient city this marl, composed of a distinctive dense white clay with an unusually high admixture of calcium carbonate, is typically found from two to ten meters below the modern surface. Overlying the marl and crowning each of the terraces is a series of younger strata, the shore deposits just mentioned, of middle to upper Pleistocene date, mostly less than half a million years old. This cap of harder rock varies considerably in depth and composition, ranging from a soft, fine oolitic limestone (the “poros” of archaeological reports) to coarser, more friable sandstones and pebbly conglomerates.
7In spite of their variability, these upper strata, fine limestones and coarse conglomerates alike, share one important quality in contrast to the marl on which they rest. The upper layers of rock are porous; the dense, impermeable marl is not. Rain water and surface runoff percolating down through the porous bedrock cap cannot penetrate the surface of the marl: it accumulates on top of the deep clay bed, filling hollows and cavities and spreading laterally along the seam between the two strata to form a series of shallow subterranean reservoirs. The immediate hydrological effect of this process is to concentrate almost the entire available supply of groundwater at one particular point in the stratigraphic column. Once this pattern was recognized by the ancient inhabitants of the site, the location of the water- bearing horizon could be easily predicted and the resource reliably tapped at virtually any point within the city.
Fig. 2 – Catchment tunnel (left) and storage reservoir (right) of Peirene. Photos author.

8One consequence of these conditions has received considerable attention in the literature on Corinthian water supply. Wherever a cliff or ravine of sufficient depth cuts through the superimposed strata and exposes to the air the juncture between the marl and the limestone cap, the water that has accumulated at the seam may be released as a natural spring. This is the importance of the terraces on which the city is built, for such exposures are a ubiquitous feature of the low cliff faces separating one terrace from the next, and as a result springs and seeps of various sizes are to be found everywhere along their northern edges.5 Such natural springs ultimately lie behind many (although by no means all) of the developed water sources of the ancient city, even those that were much altered by later architectural embellishment.
9Just as geology has been paramount in determining the location of the city’s principal springs, so too has it influenced the manner of their exploitation and development. In order to increase the flow of water from these natural sources, the Corinthians at an early date began to dig tunnels back into the cliffs at the level of the joint between the marl and the overlying bedrock, in order to increase the surface area of the exposed joint. A horizontal tunnel following the plane of the joint guarantees the greatest amount of exposure with the least expenditure of effort, and because water may trickle in, in varying quantities, from nearly every point at which the joint is exposed, such tunnels are an extremely efficient means of increasing a given spring’s productivity (Fig. 2, left). They are easily dug through the soft marl and can be extended as far as desired with no specialized tools or training. Examples of these catchment tunnels or collecting galleries, among the most characteristic of Corinthian waterworks, are to found everywhere beneath the city, ranging from single short tunnels to networks of radiating corridors with multiple branches and sub-branches. The fountain of Peirene, the intensive development of which began in the Archaic period, if not earlier, is the best known and most impressive illustration of the practice: in its fully developed form it was served by a sprawling network of supply tunnels that extended for hundreds of meters behind the face of the terrace to the southeast and southwest, beneath and beyond the area later occupied by the forum of the Roman colony.6
10 The creation of these collecting galleries was the first step in the architectural development of a natural water source at Corinth, and in many cases the development stopped there.7 But where the demand for water was greater at times than the available flow, tunnel-like storage reservoirs could easily be contrived in the same manner, by cutting back into the cliff face to create a long, narrow chamber and lining it with hydraulic cement (Fig. 2, right). These chambers, which in form are simply upscaled versions of the catchment tunnels that preceded them, made convenient and capacious reservoirs, simple to construct and remarkably resistant to collapse. The unique qualities of the Corinthian marl, which is soft enough to be easily cut through, but also extremely dense and cohesive, ensured that the chambers would hold their shape without any additional support, and where a stucco lining was desired it could be applied directly to the firm marl walls.8 The addition of draw basins in front of the reservoirs, and in some cases a decorative architectural façade as well, completed the transformation of a natural spring in a cliff face into a man-made fountain house.
11The fountain of Peirene is the best preserved and most fully developed example of this development.9 More clearly than any other monument, it illustrates the way in which the plan of the traditional Corinthian fountain house evolved in direct response to a specific set of geological conditions, and it provides a repertoire of construction techniques and architectural forms that the builders of the city, Greek and Roman alike, continued to draw upon for centuries. But the significance of Peirene extends beyond its morphology. The most celebrated of Corinthian springs, regularly invoked by ancient authors as an exemplar of copious purity, it was well known outside the city’s borders and by the early Classical period had become so closely associated with Corinthian civic identity that Pindar could refer to Corinth, using an instantly recognizable metonymy, simply as “the city of Peirene.”10 The fountain was one of the first monuments to be restored after the foundation of the Roman colony, and its reputation continued to grow throughout the Roman period, when the power of poetic inspiration was added to its old virtues of abundant supply and superlative flavor.11 Peirene’s literary and historical preeminence and its role as a civic and cultural symbol, along with the obvious antiquity and distinctive physical form of the monumentalized spring, ensured that it would have a uniquely powerful influence on the development of other water sources in and around the city. In the creation of the canonical Corinthian fountain house, Peirene was not merely a typical example; it was the template.
12I return now to the fountain house at Perachora, which bears a remarkable resemblance in plan to fountains of this distinctive Corinthian type, above all in its long, narrow underground reservoirs cut out of the local clay. That it belongs to the same architectural tradition that produced Peirene and other fountains at Corinth itself cannot, I think, be doubted, and its appearance has misled some scholars into treating it as a typical representative of the type.12 A closer look, however, reveals important differences in substance beneath the superficial similarities of form, and shows just how costly and cumbersome it was for the builders of the fountain to create this impression.
13The Perachora fountain house does not occupy the site of a natural spring, and therefore cannot be a product of the kind of local, geologically directed architectural evolution that we see at Corinth. In fact, the plain in which the fountain stands is completely dry, so the water to supply its reservoirs had to be hauled up from a depth of 30 m below ground through three great rectangular shafts, each 12–15 m long and over a meter wide. Underground tunnels linking the shafts and a staircase some 60 m long to provide access to them completed the system, one of the most elaborate and expensive supply systems associated with any known Greek fountain house.13
14After going to such trouble to obtain the water, one might have expected the builders to place the point of use right next to the supply shafts. But instead they chose to construct the fountain house itself nearly 200 m away, and to transport the water to it in a small stuccoed runnel at ground level (Tomlinson 1969: 202–203). What was the reason for this curious decision? What made this particular spot in the middle of the plain the most attractive site for the fountain house? The answer is that it was the only place anywhere in the vicinity that offered a natural scarp of soft clay covered by a harder rocky cap into which the long, narrow storage reservoirs could be dug (Fig. 3). In other words, it offered a setting superficially similar to that of Peirene and other fountains at Corinth, and allowed the builders of the Perachora fountain house to duplicate—or to attempt to duplicate—the tunnel-shaped, rock-cut reservoirs with which they were familiar from the fountains in the city (Fig. 4).
Fig. 3 – The site of the Perachora fountain house. Photo B.A. Robinson.

Fig. 4 – Storage reservoirs of Peirene (left) and the fountain house at Perachora (right). Photos author.

15R. A. Tomlinson, in his publication of the Perachora fountain, points out that by constructing the reservoir chambers in this way, the builders were able to use the stratum of harder rock overlying the clay as a natural ceiling; he describes the spot as “an ideal position” for the fountain house, and implies that this is sufficient explanation for the choice of the site and the form of the building.14 The practice is certainly common at Corinth itself, where the rocky cap above the marl was regularly used in this way, and where the economy and convenience offered by this mode of construction were important motivating forces behind the evolution of the plan in the first place. But does it really offer the same advantages at Perachora? Closer scrutiny suggests precisely the opposite: that in this case, on this particular site, the architectural form chosen by the builders, and the manner of construction associated with it, is astonishingly inconvenient and inefficient, and that it required a significantly greater expenditure on materials and labor than would have been entailed in other, more practical designs.
Fig. 5 – Storage reservoir of the fountain house at Perachora. Photo author.

Fig. 6 – Double-apsidal cistern at Perachora. Photo author.

16Remember that the principal advantage of this type of construction at Corinth was its extreme simplicity. The reservoirs and supply tunnels of Peirene and other Corinthian fountains required virtually no labor beyond the simple cutting of the chambers into the marl. The strength and density of the Corinthian marl ensured that no additional support for the walls or ceiling was necessary, and the waterproof stucco lining could be applied directly to the surface of the marl itself. It is precisely these advantages that the site at Perachora fails to offer. The poorly cemented clay into which the Perachora reservoirs are cut is miserable stuff: loose, powdery, and full of shells. Unlike the firm, well-cemented marl at Corinth, it cannot hold a shape without additional support, and the crumbly texture of the matrix, together with the brittle, projecting shell fragments, made it impossible to prepare a surface smooth and solid enough to accept a stucco lining of the reservoirs directly.
17In order, therefore, to prevent the collapse of the chambers, and to provide a suitable surface for the application of the stucco, the builders of the Perachora fountain house were forced to adopt a highly labor-intensive alternative: they lined the rock-cut sides of each of the reservoirs with walls of ashlar masonry, half a meter wide and four courses high, built of carefully mortared blocks of limestone (Fig. 5). The entire exposed surface of these walls was then systematically scored with the tip of a pick in order to ensure better adhesion of the stucco lining.
18It should be immediately obvious that, once the construction of such walls becomes necessary, the decision to build rock-cut, tunnel-shaped reservoirs is no longer a particularly efficient or economical one. The whole point of a rock-cut reservoir is that one can simply cut it out of the rock and then walk away. If, on the other hand, masonry walls are going to be required in any case, it makes little sense to construct them inside a series of narrow underground galleries, where the work is made more difficult by the close quarters and the poor lighting. Far better to employ the “cut-and-cover” method, constructing the walls in a large open pit or trench dug from the surface, or even entirely above ground. The advantage noted by Tomlinson, that the ledge of harder bedrock overlying the friable clay could be used as a natural ceiling, and that the necessity of constructing a roof could thereby be avoided, is not a compelling counterargument, since any savings in materials and labor thus effected were more than offset by the number of separate tunnels that had to be dug in order to provide the total desired storage capacity, and by the necessity of lining each side of each individual rock-cut chamber with masonry walls, thereby greatly increasing the amount of stone required. No special mathematical or architectural training is needed to perceive that three narrow tunnels lined with stone walls on both sides require nearly three times as much stone as a single large reservoir of the same total capacity. In fact, it’s hard to imagine any other plan in which the volume of masonry in proportion to the volume of available storage space for water would have been more wasteful.
19Ironically, an excellent example of a more rational approach to the problem of underground water storage at Perachora can be found only a few minutes’ walk to the west, in the Heraion valley just east of the harbor. There a large, well-preserved cistern with apsidal ends, excavated in the 1930s and published in 1969, has been tentatively dated to the IIIrd century BC, more or less contemporary with the fountain house in the upper plain.15 The cistern consists of a single large chamber constructed in the “cut-and-cover” technique, within a trench dug down from the surface; the sides, like those of the fountain house reservoirs, were lined with masonry and sealed with hydraulic cement, and the roof, at ground level, was supported on a central row of piers (Fig. 6). This cistern, I submit, does everything that the fountain house reservoirs do, and does it better and more economically. Its construction required less in the way of materials and labor; the finished structure, like the fountain house reservoirs, lay entirely below ground level, and so was equally stable and well-insulated; it was much easier to clean and maintain; and, finally, it held essentially the same amount of water (ca. 300 cubic meters vs. ca. 332 cubic meters for the fountain house tunnels combined). Moreover, if the builders at Perachora needed a precedent for the successful use of this kind of storage facility in a fountain house, they had only to look to the neighboring city of Megara, where the so-called fountain house of Theagenes, constructed perhaps in the first quarter of the Vth century BC, was served by a similar reservoir, consisting of a single large chamber divided in two by a partition wall and covered with a roof supported on polygonal piers.16
Fig. 7 – Plan of the fountain of Glauke at Corinth. Corinth I.6, fig. 126; courtesy of the Trustees of the American School of Classical Studies at Athens.


Fig. 8 – The fountain of Glauke at Corinth. Corinth I.6, fig. 144; courtesy of the Trustees of the American School of Classical Studies at Athens.
20In light of all this, it’s hard to believe that the architects at Perachora adopted the traditional Corinthian plan for their fountain house because it was the easiest or most convenient way in which to build it. On the contrary: in this case at least, the architectural form forced upon the builders a manner of construction that was neither easy nor convenient. Instead of being a simple, practical, and inexpensive response to the local geological conditions, the design was here imposed upon a set of very different conditions at considerable cost.
21How then are we to account for its form? The choice can hardly be accidental, and it would be uncharitable to suggest that it was the product of ignorance or bull-headedness on the part of a group of Corinthian architects and workmen who knew only one way to build a fountain house and could conceive of no other. The late IVth and early IIIrd centuries BC saw a great deal of activity at Corinth involving waterworks of very similar sorts, and the Corinthians involved in the construction of the Perachora fountain house will have been experienced men: they knew the practical advantages of the Corinthian system, knew also why it was particularly well-suited to conditions at Corinth, and could not have failed to appreciate how ill-suited it was to Perachora. I suspect, therefore, that this had very little to do with practical considerations and everything to do with appearances — specifically, with a desire to reproduce the distinctive plan of a traditional Corinthian fountain house regardless of the obstacles, and a willingness to tolerate a certain amount of inefficiency and additional cost in order to create a building that imitated the appearance of its models as closely as possible.
22As it happens, we can see the very same kind of reasoning at work at Corinth itself, in two of the city’s best-known fountains, which also happen to be roughly contemporary with the fountain house at Perachora. The first is the fountain of Glauke, situated in the center of the city, ca. 200 m west of Peirene and just northwest of the Roman forum. Its general plan is very similar to that of Peirene and the Perachora fountain house: a series of narrow, roughly parallel, rock-cut storage reservoirs cut back into a natural scarp and fronted by a group of draw basins and an architectural façade, the latter now missing (Figs. 7, 8). The resemblance to Peirene would have been even greater in its original state, when Glauke formed part of a continuous ridge of rock that extended westward from the hill occupied by the Archaic temple; its peculiar, blocky appearance today is the result of later Roman quarrying, in the which the ridge on all sides of the fountain was cut away, leaving it an isolated cube of stone.17
23From a glance at the plan one might easily conclude that Glauke was a product of the same kind of on-the- spot architectural evolution that produced Peirene: the exploitation of a natural spring in the cliff face, gradually augmented with catchment tunnels, reservoirs, and draw basins in order to increase and store the natural flow of water at the site. That this is not in fact the case has been known since the original publication of the fountain in 1910, although the misunderstanding is still found frequently in modern scholarship, including standard handbooks on ancient water supply and urban development.18 The truth is that Glauke is an entirely artificial construction, at least in the sense that there never was a natural spring at this site, and all of the water that supplied the fountain had to be brought in from elsewhere. The entire installation is cut out of the limestone crown of the ridge, well above the marl stratum that determines the position of the local water table, and the reservoirs were filled instead by a pipeline that drew its supply from an entirely different source hundreds of meters to the south.19 In its architectural form, Glauke faithfully follows the model established by fountains like Peirene, but the success of the mimicry conceals the fact that in this case the traditional form has lost almost all connection with the geological conditions that originally helped to shape and define it.
Fig. 9 – Plan of the reservoirs in the lower courtyard of the Asklepieion ( “Lerna”) at Corinth. Corinth XIV, plan C; courtesy of the Trustees of the American School of Classical Studies at Athens.

24The same can be said of the reservoirs in the peristyle courtyard with dining rooms that forms part of the sanctuary of Asklepios at the northwestern edge of the city (Figs. 9, 10). Since the 1930s these waterworks have generally been identified with the spring of Lerna mentioned by Pausanias, and although that identification is probably incorrect, for convenience I will continue to use the name here.20 At Lerna a series of five long, narrow rock-cut reservoirs was dug back into the exposed scarp created by the construction of the lower courtyard. Although the chambers were cut through marl just beneath the harder limestone cap, the joint between the strata at this elevation is entirely dry today, and it was evidently not a significant aquifer in antiquity either, to judge from the fact that no catchment tunnels were created at this level to tap the local groundwater. Instead, two of the reservoirs seem to have been simple cisterns, supplied through openings in the ceilings by rainwater and surface runoff, while the other three, like the reservoirs of Glauke, were filled with water transported to this site from an entirely different source hundreds of meters to the south.21
25The Lerna reservoirs can be dated with some confidence to a major expansion of the Asklepieion in the late IVth to early IIIrd century BC, and although the date of Glauke has been a matter of some debate, it too probably fits most comfortably into the early Hellenistic period.22 Both installations are therefore roughly contemporaneous with the fountain house at Perachora, and like it, they clearly look back to an architectural template established centuries earlier, while at the same time transferring it to a physical setting very different from the one for which it had originally been developed. While Glauke and Lerna have often been lumped together with Peirene as “typical” Corinthian fountains, it is perhaps more appropriate to view them as examples of intentional archaism, carefully designed to recall the architectural form of Peirene and other venerable installations constructed on the sites of natural springs.23
Fig. 10 – Reservoir II in the lower courtyard of the Asklepieion ( “Lerna”) at Corinth. Photo author.

26The image of the grotto in the classical world—a shaded hollow beneath a rocky overhang, haunted by nymphs and refreshed by copious streams of cool, clear water—is perhaps best known through its appearance in Roman poetry, and it played an important role in the design of many imperial Roman nymphaea. But all of the ideal grotto’s characteristic features, both physical and metaphorical, were well established by the Hellenistic period, and in Peirene the Corinthians had long possessed one of the most famous and evocative instantiations of that ideal.24 It is no surprise, then, to find the Hellenistic builders at Corinth and Perachora attempting to reproduce this effect in a series of newly-created, artificial fountains, patiently coaxing the illusion of natural springs in mossy caves out of the dry rock and powdery clay that nature had given them to work with. In all three of the fountains discussed here, Glauke, Lerna, and Perachora, we can see the same conservative, antiquarian impulses at play, in the deliberate imitation of venerable construction techniques and architectural forms that had come to define the traditional image of a Corinthian fountain. The careful observance and faithful perpetuation of these forms, even in circumstances where they conferred no practical advantage, demonstrate clearly the dominant position occupied by Peirene in the physical and cultural landscape of Corinth, and the powerful and lasting effect that the accidents of geology had on the architectural development of the city’s hydraulic resources.
Bibliographie
Des DOI sont automatiquement ajoutés aux références bibliographiques par Bilbo, l’outil d’annotation bibliographique d’OpenEdition. Ces références bibliographiques peuvent être téléchargées dans les formats APA, Chicago et MLA.
Format
- APA
- Chicago
- MLA
Bibliography
COULTON 1964: J. J. Coulton, The Stoa by the Harbour at Perachora, ABSA 59, 1964, p. 100–131.
CORINTH I.1: H. N. Fowler, R. Stillwell, Introduction, Topography, Architecture (Corinth I.1), Cambridge, Mass., Harvard University Press, 1932.
CORINTH I.6: B. H. Hill, The Springs: Peirene, Sacred Spring, Glauke (Corinth I.6), Princeton, American School of Classical Studies at Athens, 1964.
CORINTH VII.6: I. McPhee, E. G. Pemberton, Late Classical Pottery from Ancient Corinth: Drain 1971-1 in the Forum Southwest (Corinth VII.6), Princeton, American School of Classical Studies at Athens, 2012.
CORINTH XIV: C. Roebuck, The Asklepieion and Lerna (Corinth XIV), Princeton, American School of Classical Studies at Athens, 1951.
CORINTH XX: C. K. Williams, N. Bookidis (eds.), Corinth: The Centenary, 1896–1996 (Corinth XX), Princeton, American School of Classical Studies at Athens, 2003.
10.2307/498957 :DE WAELE 1933: F. J. de Waele, The Sanctuary of Asklepios and Hygieia at Corinth, AJA, 37, 1933, p. 417–451.
DE Waele 1935: F. J. de Waele, The Fountain of Lerna and the Early Christian Cemetery at Corinth, AJA, 39, 1935, p. 352–359.
DELBRÜCK, VOLLMÖLLER 1900: R. Delbrück, K. G. Vollmöller, Das Brunnenhaus des Theagenes, AM, 25, 1900, p. 23–33.
10.4324/9781315771809 :DIXON 2014: M. D. Dixon, Late Classical and Early Hellenistic Corinth, 338– 196 B.C., London and New York, Routledge, 2014.
ELDERKIN 1910: G. W. Elderkin, The Fountain of Glauce at Corinth, AJA, 14, 1910, p. 19–50.
GRUBEN 1964: G. Gruben, Das Quellhaus von Megara, AD, 19, 1964, A’, p. 37–41.
HAYWARD 2003: C. L. Hayward, Geology of Corinth: The Study of a Basic Resource, in Corinth XX, p. 15–42.
HELLNER 2004: N. Hellner, Die Krene von Megara, AM, 119, p. 163–220.
HODGE 1992: A. T. Hodge, Roman Aqueducts and Water Supply, London, Duckworth, 1992.
KAVVADAS et al. 2003: M. J. Kavvadas, A. G. Anagnostopoulos, V. N. Georgiannou, M. E. Bardanis, Characterisation and Engineering Properties of the Corinth Marl, in T. S . Tan, K. K. Phoon, D. W. Hight, and S. Leroueil (eds.), Characterisation and Engineering Properties of Natural Soils, Lisse, 2003, vol. 2, p. 1435–1460.
10.1016/0025-3227(88)90034-5 :KERAUDREN, SOREL 1987: B. Keraudren, D. Sorel, The terraces of Corinth (Greece): a detailed record of eustatic sea-level variations during the last 500,000 years, Marine Geology, 77, 1987, p. 99–107.
LANDON 1994: M. E. Landon, Contributions to the Study of the Water Supply of Ancient Corinth, Ph.D. diss., Univ. of California, Berkeley, 1994.
LANDON 2003: M. E. Landon, Beyond Peirene: Toward a Broader View of Corinthian Water Supply, in Corinth XX, p. 43–62.
OLESON 1984: J. P. Oleson, Greek and Roman Mechanical Water-Lifting Devices: The History of a Technology (Phoenix Supplement 16), Toronto, Dordrecht, D. Reidel, 1984.
PAYNE 1940: H. Payne, Perachora I: The Sanctuaries of Hera Akraia and Limenia, Oxford, Clarendon Press, 1940.
PFAFF 2003: C. A. Pfaff, Archaic Architecture, in Corinth XX, p. 95–140.
ROBINSON 1962: H. S. Robinson, Excavations at Corinth, 1960, Hesperia, 31, 1962, p. 95–133.
ROBINSON 2005: B. A. Robinson, Fountains and the Formation of Cultural Identity at Roman Corinth, in D. N. Schowalter and S. J. Friesen (eds.), Urban Religion in Roman Corinth: Interdisciplinary Approaches (Harvard Theological Studies 53), Cambridge, Mass., Harvard Divinity School, 2005, p. 111–140.
ROBINSON 2011: B. A. Robinson, Histories of Peirene: A Corinthian Fountain in Three Millennia, Princeton, American School of Classical Studies at Athens, 2011.
ROBINSON 2013: B. A. Robinson, Playing in the Sun: Hydraulic Architecture and Water Displays in Imperial Corinth, Hesperia, 82, 2013, p. 341–384.
ROUX 1958: G. Roux, Pausanias en Corinthie (Livre II, 1 à 15), Paris, 1958.
SANDERS, MIURA, KVAPIL 2014: G. Sanders, Y. Miura, L. Kvapil, A Re-examination of Some of the South Stoa Wells at Corinth, Black Sea Studies, 16, 2014, p. 65–81.
SIDDALL 2006: R. Siddall, Hydraulic Mortars in Antiquity: Analyses from Fountains in Ancient Corinth, in A. Brauer, C. Mattusch and A. Donohue (eds.), Common Ground: Archaeology, Art, Science, and Humanities: Proceedings of the XVIth International Congress of Classical Archaeology, Boston, August 23–26, 2003, London, 2006, p. 208–211.
TÖLLE-KASTENBEIN 1990: R. Tölle-Kastenbein, Antike Wasserkultur, Munich, C.H. Beck, 1990.
10.1017/S0068245400014568 :TOMLINSON 1969: R. A. Tomlinson, Perachora: The Remains outside the Two Sanctuaries, ABSA, 64, 1969, p. 155–258.
TOMLINSON 1976: R. A. Tomlinson, The Perachora Waterworks: Addenda, ABSA, 71, 1976, p. 147–148.
10.4324/9780203412909 :TOMLINSON 1992: R. A. Tomlinson, From Mycenae to Constantinople: The Evolution of the Ancient City, London, Routledge, 1992.
VITA-FINZI, King 1985: C. Vita-Finzi, G. C. P. King, The seismicity, geomorphology, and structural evolution of the Corinth area of Greece, Philosophical Transactions of the Royal Society A: Mathematical, Physical, and Engineering Sciences, 314, 1985, p. 379–407.
WESTAWAY 1996: R. Westaway, Quaternary Elevation Change of the Gulf of Corinth in Central Greece, Philosophical Transactions of the Royal Society A: Mathematical, Physical, and Engineering Sciences, 354, 1996, p. 1125–1164.
WILLIAMS 1987: C. K. Williams II, The Refounding of Corinth: Some Roman Religious Attitudes, in S.A. Macready, F.H. Thompson (eds.), Roman Architecture in the Greek World, London, 1987, p. 26–37.
WILLIAMS, ZERVOS 1984: C. K. Williams II, O. Zervos, Corinth, 1983: The Route to Sikyon, Hesperia, 53, 1984, p. 83–122.
WISEMAN 1967: J. Wiseman, Excavations in Corinth: The Gymnasium Area, 1966, Hesperia, 36, 1967, p. 402–428.
WISEMAN 1969a: J. Wiseman, Excavations in Corinth: The Gymnasium Area, 1967–1968, Hesperia, 38, 1969, p. 64–106.
WISEMAN 1969b: J. Wiseman, Ancient Corinth: The Gymnasium Area, Archaeology, 22, 1969, p. 216–225.
WISEMAN 1970: J. Wiseman, The Fountain of the Lamps, Archaeology, 23, 1970, p. 130–137.
WISEMAN 1972: J. Wiseman, The Gymnasium Area at Corinth, 1969–1970, Hesperia, 41, 1972, p. 1–42.
WISEMAN 1979: J. Wiseman, Corinth and Rome I: 228 B.C.–A.D. 267, in ANRW, II.7.1, p. 438–548.
Notes de bas de page
1 Dinsmoor’s notes from the trip are preserved in Corinth notebooks 67: 44–47 (original pencil notes) and 69: 3–13 (fair copy in ink). I first became interested in the peculiarities of the Perachora fountain house nearly 25 years ago, and tried out some of these ideas in a brief talk at the meeting of the Archaeological Institute of America in 1993. I am grateful to Sophie Bouffier and the HYDRΩMED project for the opportunity to revisit the topic in greater detail here.
2 Tomlinson 1969: 201–218; 1976. In the absence of useful stratigraphic evidence, the date of the fountain house rests entirely on the architectural style of the façade. Tomlinson, comparing it with the Ionic stoa in the harbor at Perachora, concluded (1969: 218) that “both seem to be part of the same general redevelopment of the Heraion at the end of the fourth century BC”; in this he followed the date proposed for the stoa by J. J. Coulton (Coulton 1964: 24–129). Coulton’s stylistic analysis depended heavily on comparisons with the South Stoa at Corinth, once placed in the 330s BC, but now dated to ca. 300 BC. or even later: see, e.g., Corinth VII.6: 14–19; Sanders, Miura, Kvapil 2014; Dixon 2014: 121–127. Dating late Classical and Hellenistic architecture solely on the basis of style is a precarious practice at best, but it now seems that the fountain house at Perachora is as likely to belong in the early IIIrd century BC as in the late IVth.
3 On the discovery and excavation of Peirene, see Landon 2003: 45–46; Robinson 2011: 68–77.
4 For the debate over the structure and date of the Corinthian terraces and the geological processes that produced them, see Vita-Finzi, King 1985; Keraudren, Sorel 1987; Westaway 1996. A convenient summary for non-specialists can be found in Hayward 2003: 16–17.
5 The pattern is clearly visible in published maps of the city’s water sources: see Landon 2003: 47, fig. 3.1; Robinson 2011: pl. A.
6 Corinth I.6: 54–63, pl. III; Robinson 2011: 11–17, pl. D. Such tunnels are not, of course, unique to Corinth; they can be found throughout the Mediterranean basin, wherever the local geological conditions support them. But Corinth, where the rock-cut tunnel remained for centuries the basic unit of water collection, storage, and transport, is rightly considered the type-site for this form of hydraulic technology.
7 A good example of such a source, augmented with tunnels for catchment but apparently no further architectural elaboration, is the spring known today as the Baths of Aphrodite, situated at the southern end of a large embayment in the face of the city’s lower terrace, close to the line of the Lechaion Road (Corinth I.1: 91–91; III.2 269–271; Robinson 1962: 120–130; Landon 1994: 222–234).
8 The exceptional strength and cohesiveness of the Corinthian marl is due to an unusually high concentration of calcium carbonate, which acts as a cementing agent, bonding the clay particles together and preventing the kind of disintegration and collapse seen in other, less well-cemented marls (Kavvadas et al. 2003).
9 For more detailed descriptions of the forms and phases of the fountain, see Corinth I.6: 1–115; Robinson 2011.
10 Pindar, Ol. 13.61 (464 BC). Peirene’s literary and mythological reputation is discussed in detail in Robinson 2011: 27–64.
11 On the restoration of the fountain and its appeal to the Roman colonists of Corinth, see Robinson 2005: 116–127; 2011: 175–200.
12 E.g., Tölle-Kastenbein 1990: 26–28, fig. 9, where the reservoirs are wrongly described as catchment tunnels (Sickerstollen).
13 Payne 1940: 11–13; Tomlinson 1969: 195–203; 1976. The nature of the apparatus used to raise water from the shafts is uncertain. Tomlinson’s original proposal of large-diameter bucket-chain wheels had to be amended in light of subsequent excavation around the mouths of the shafts, and his reconstruction has not been widely accepted (cf. the criticisms in Oleson 1984: 237–240).
14 Tomlinson 1969: 205, citing the reservoirs of the “Lerna” system adjacent to the Asklepieion at Corinth as a parallel. The comparison is apt; I shall return to it below.
15 Payne 1940: 14, pl. 4: a; Tomlinson 1969: 157–164, pls. 45–47 ( “the double-apsidal cistern”).
16 Delbrück, Vollmöller 1900; Gruben 1962; Hellner 2004.
17 On the quarrying of the ridge, see Williams, Zervos 1984: 97–98; Hayward 2003: 25.
18 E.g., Tölle-Kastenbein 1990: 26–27, where the reservoirs are mischaracterized as catchment tunnels (Sickergalerien); Hodge 1992: 25, 406 n. 4; Tomlinson 1992: 76.
19 These facts were stated clearly in Elderkin 1910: 39–40 (reprinted in Corinth I.6: 222, with an additional note by W. B. Dinsmoor: 227–228); see also Williams 1984: 99; 1987: 34; Landon 2003: 48 n. 21; Pfaff 2003: 133; Robinson 2005: 128–132; 2013: 345.
20 The identification of the waterworks in the courtyard west of the Asklepieion with the spring of Lerna described by Pausanias (2.4.5) was proposed when the site was excavated in the 1930s (De Waele 1933: 432; 1935: 353–354); it was accepted by Roebuck in the final publication (Corinth XIV) and continues to be taken for granted in most of the Corinthian literature. The excavation of the so-called Fountain of the Lamps a short distance to the west in the 1960s and 70s, however, brought to light another candidate for the name. No conclusive epigraphical evidence has been found at either site, and the matter cannot be decided on topographical grounds, since both installations lie in the quarter of the city indicated by Pausanias, somewhere beyond the theater, close to the gymnasium and the temples of Zeus and Asklepios. Neither complex conforms very closely in its architectural details to Pausanias’s description, for both lack the surrounding κίoνες to which he specifically refers. The courtyard adjacent to the Asklepieion was indeed enclosed by colonnades in its original Hellenistic phase, but these had been destroyed to the level of the foundations long before Pausanias’s visit (Corinth XIV: 90–91), and there is no good evidence for the Roman reconstruction of the northern colonnade tentatively suggested by Roebuck. The walled precinct surrounding the pool in front of the Fountain of the Lamps, on the other hand, seems never to have had a colonnade at all (Wiseman 1972: 22, where a colonnade is considered but found unlikely; 1979: 511). Wiseman counters this objection with the suggestion that Pausanias may have been referring instead to the stoas of the gymnasium on the terrace above the spring to the south, but this idea is undermined by the precision of Pausanias’s language: he clearly states that the columns stood not near the spring or above the spring, but περὶ αὐτὴν. Attempts to explain away the contradiction by denying the plain meaning of the Greek (Roux 1958: 127) or by asserting that Pausanias did not actually visit the site himself (Wiseman 1979: 511–512) are unworthy of serious consideration. The question remains open, and the problems cannot be easily dismissed, but at present the Fountain of the Lamps seems the more likely candidate, if only because the waterworks there were probably still functioning at the time of Pausanias’s visit, while most of those in the courtyard next to the Asklepieion seem to have gone out of use by the beginning of the 2nd century A.D. (Corinth XIV: 106; Wiseman 1969a: 75).
21 For the reservoirs, see Corinth XIV: 102–106. The supply system for reservoirs III–V was an underground aqueduct, which has been traced for over 100 m to the south (Corinth XIV: 105–106; Wiseman 1967: 416– 417; 1969b: 218; 1970: 131–132; 1972: 72–75). The fact that the walls of this tunnel were sealed with waterproof stucco shows clearly that it was used for transportation, not for catchment. There is indeed a natural spring at the site of Lerna, one that was augmented in antiquity with catchment tunnels of the characteristic Corinthian type, but it lies at a lower elevation, below the level of the reservoirs under discussion here (Corinth XIV: 106 –110, Plan E; Landon 1994: 247–264). After the construction of the courtyard, the water from this spring was conducted through stone drainage channels beneath the floor for use outside the city walls. The only installation at courtyard level to draw upon the local water table was the small spring house in the southeast corner (Corinth XIV: 96–99), which had a capacity of perhaps 10 m3; the large reservoirs at courtyard level, with a cumulative capacity of ca. 340 m3, had no connection to this source. The important distinction between these two groups of waterworks has often been overlooked.
22 The early excavators assumed, on no grounds other than its formal resemblance to Peirene, that Glauke too had its origins in the Archaic period. This notion has proved remarkably tenacious, although there was never any evidence to support it, and good reason to believe otherwise. In the 1980s Williams summarized a number of features that made such an early date unlikely and tentatively suggested that Glauke was an Early Roman construction (Williams 1984: 98–100; 1987: 34–35). He allowed for the possibility of a Hellenistic first phase, however, and a pre-Roman origin is supported by the presence in the fountain of Greek stucco, which usually lacks the crushed terracotta particles that are a ubiquitous feature of Roman hydraulic mortars at Corinth (Siddall 2006). The current consensus is that the fountain is most likely a creation of the late IVth–early IIIrd century BC, perhaps a product of the same building program that produced, inter alia, the South Stoa and the Asklepieion courtyard with its waterworks (see, e.g., Landon 2003: 48 n. 21; Robinson 2005: 128–131; 2013: 350). For a partial overview of Corinthian building activity during this period (not including Glauke), see Dixon 2014: 110–131.
23 On Glauke’s antiquarian appeal to the Roman colonists at Corinth, see Williams 1987: 34–35; Robinson 2005: 132–138; 2013: 349–350. That the same archaizing qualities were also appreciated by Corinthians of the Hellenistic period seems not unlikely.
24 On Hellenistic grottos, see Robinson 2011: 166–173, focusing especially on the so-called Cyclopean Fountain at Corinth, but with a useful collection of other evidence, from miniature votive fountains of terracotta in Magna Graecia to Alexandrian descriptions of the elaborate artificial nymphaea of the Ptolemies.
Auteur
-
Mark Landon
Department of Classics Mount Holyoke College
Le texte seul est utilisable sous licence Licence OpenEdition Books. Les autres éléments (illustrations, fichiers annexes importés) sont « Tous droits réservés », sauf mention contraire.
Peupler et habiter l’Italie et le monde romain
Stéphane Bourdin, Julien Dubouloz et Emmanuelle Rosso (dir.)
2014
Archéologie au présent
Les découvertes de l’archéologie préventive dans les médias
Catherine Dureuil-Bourachau
2015
Sarta Tecta
De l’entretien à la conservation des édifices. Antiquité, Moyen Âge, début de la période moderne
Charles Davoine, Maxime L’Héritier et Ambre Péron d’Harcourt (dir.)
2019
Gérer l’eau en Méditerranée au premier millénaire avant J.-C.
Sophie Bouffier, Oscar Belvedere et Stefano Vassalo (dir.)
2019
Le village de la Capelière en Camargue
Du début du ve siècle avant notre ère à l’Antiquité tardive
Corinne Landuré, Patrice Arcelin et Gilles Arnaud-Fasseta (dir.)
2019
Les métaux précieux en Méditerranée médiévale
Exploitations, transformations, circulations
Nicolas Minvielle Larousse, Marie-Christine Bailly-Maitre et Giovanna Bianchi (dir.)
2019
L’Homme et l’Animal au Maghreb, de la Préhistoire au Moyen Âge
Explorations d’une relation complexe
Véronique Blanc-Bijon, Jean-Pierre Bracco, Marie-Brigitte Carre et al. (dir.)
2021