Printer Friendly

A Roman Circus in Corinth.


During the 1967-1968 excavations of the Gymnasium area in Corinth, a long and narrow structure (the "Apsidal Building") was discovered. It is argued here that the structure represents the eastern meta and a portion of the spina of a circus, where chariot races were held. The circus appears to have been planned as an integral component of the Caesarian design of the city, constructed during the Augustan period, renovated in the late 1st century A.D., and refurbished as late as the 6th century. Furthermore, the circus was often the site of the equestrian contests of the Corinthian Caesarea festival and at times of the Panhellenic Isthmian Games.


When Colonia Laus Iulia Corinthiensis was founded in 44 B.C. by Julius Caesar, the new Roman colony was built on the site of the former Greek polis that had been destroyed by the consul Lucius Mummius in 146 B.C. (1) During the interim period of 102 years, the Greek city was deprived of its civic and political identity and was largely deserted until the arrival of the earliest Roman colonists, a combination of freedmen and veterans. (2) Although certain architectural elements of the Greek city were retained in the new foundation, the Roman architects and engineers brought with them a plan for the colonial city and proceeded over time to build new buildings, monuments, and structures of characteristic Roman type. (3) For example, the amphitheater at Corinth, located in the northeast corner of the planned colony of 44 B.C., was probably constructed during the Augustan period. (4) Later, under Vespasian, in the early 70s A.D., a second Roman colony was founded at Corinth, Colonia Iulia Flavia Augusta Corinthiensis, with subsequent planning and building activity. (5)

Another structure characteristic of a Roman city was the circus, the place where chariot races were held and where spectators watched the contests. (6) Most typically, four-horse chariots, quadrigae, and two-horse chariots, bigae, would compete on a closed racecourse, although other equestrian contests and sometimes athletic, gladiatorial, and dramatic events could be staged in a circus as well. (7) It is suggested here that Corinth had a circus and that it was planned from the earliest days of the colony, constructed as an integral component of the Augustan phase of the city, and then used throughout much of antiquity. (8) I also propose that the equestrian contests of the pentaeteric Corinthian Caesarea festival were held in this circus, and sometimes the Panhellenic Isthmian Games as well.


During the 1967-1968 excavations of the Gymnasium area in Corinth by the University of Texas, under the direction of James R. Wiseman, a long and narrow structure was discovered ca. 45 m to the south of a structure tentatively identified as the Gymnasium. (9) The structure, aligned in a roughly east-west direction and called at the time the "Apsidal Building," was excavated for a distance of about 19 m with an interior width of ca. 3.5 m; the west end of the structure was not found, presumably lying beyond the area excavated (Figs. 1-3). The maximum width of the excavated structure is 4.49 m and the east-west length of the curved eastern portion is 4.84 m. A hard-packed surface identified as that of a courtyard was found to the north, east, and south of the structure.


The curved eastern end is constructed of eight well-finished, curved marble slabs that are backed by a series of large reused poros blocks of various shapes, including a column fragment. The curved marble slabs on the exterior were clamped together. One surface of every marble block is a half-round; all but one of these half-round surfaces have been embedded in the ground although in one case the half-round is the present top surface. Approximately 2 m west of the east end of the structure, the curved end of the wall is continued by rectangular blocks that run in a straight line approximately east-west. A cross-wall of additional reused blocks joins the ends of the curved eastern end of the structure. (10) After an interval of 1-2 m, poros blocks, two courses high on the south, continue the east-west walls of the structure toward the west.

Three large pits cut into the bedrock were associated by the excavator with the original use of the structure. The pits, which are aligned roughly east-west, are spaced equally (ca. 6.0 m center to center) and are of similar dimensions. (11) Shallow channels lead toward the pits from different directions. To the south of the pits, a drain, approximately 0.50 m wide and 8 m long, covered with marble slabs, was excavated, beginning 11.5 m west of the east end of the structure and extending west into the western scarp (Fig. 3). The drain is associated with a later use of the structure.

A number of worked marble pieces, including a large cone (Figs. 4, 5), were found during excavation of the structure. The marble cone, found immediately to the north of the walls, is characterized by having its apex cut horizontally; the top horizontal surface has a central circular cutting similar to an empolion cutting, 0.09 m in diameter and 0.018 m deep. The bottom of the marble cone is broken off and thus the original size and form are not known. (12)


Several marble sculptural fragments were found in a 6th-century A.D. dump deposit above the Apsidal Building: a large marble egg-shaped object (S-2879; Fig. 6), a colossal marble hand (S-2877), and a third of the head of a young boy (S-2873) as well as most of his right arm (S-2876); a life-size terracotta comic mask (MF-12977) was also found. (13) The left leg from a statue of an athlete (S-2923), with an attached palm tree support, was discovered just outside of the dump. (14)


Three lead tablets were excavated immediately outside of the Apsidal Building: MF-12994, MF-68-271, and MF-68-272 (Fig. 3:A-C). (15) A heavy foundation wall, wall 16, was found approximately 18 m north of, and roughly parallel to, the Apsidal Building (Figs. 7-9). This wall, 1.2 m wide, was excavated for a total length of approximately 16 m. (16) At 6 and 10.5 m north of wall 16, where the West Waterworks structure begins, are robbing trenches (labeled as two pillaged trenches) for two walls that run parallel to wall 16 (Figs. 8, 9). (17) The space between these walls was filled with dumped material.


Twenty-five meters to the east of the eastern (excavated) end of wall 16, and on the same axis, Wiseman found evidence for a similar wall, 1.50-1.70 m wide; this stretch was excavated to a length of 8.15 m (Fig. 7). An additional lead tablet was found in this vicinity, approximately 23 m to the northeast of the east end of the Apsidal Building. (18) The wall continued under a farm road to the west, and its robbing trench was discovered to the east for another 7 m as cuttings in bedrock. A parallel wall was found approximately 7 m north of this wall. The northern wall, about a meter wide originally, is partially preserved and robbed. Wiseman suggested that these two parallel foundations may have been for a stoa, with a total width (including the walls) of 8.75 m, and that the building faced south onto an open hard-packed courtyard, which was explored for 19 m to the south of the foundations (Fig. 7). (19)

The only excavation undertaken to the west of the long and narrow structure (ca. 60 m to the west) was the early trench I, excavated by R. B. Richardson in 1896, in which a portion of a Late Roman wall made up of spolia was discovered (Fig. 7). In addition, to the south, Richardson dug a narrow north-south trench approximately 50 m long in which he found very little other than a series of blocks covering an area ca. 5 m wide at the north end. (20)

Although Wiseman did not positively identify the structure in the original excavation report, Charles K. Williams subsequently made the suggestion that the Apsidal Building could be a rural or outdoor shrine to Diana Nemorensis, possibly set up by the Roman freedmen when settling the Caesarian colony. (21) Williams envisioned an open-air temenos surrounded by a simple low wall and suggested that the original shrine to Diana, from the early days of the Caesarian colony, had been located somewhere near the Fountain of the Lamps (Fig. 16, below) and that it was rebuilt with its original component parts in the present location. He identified the truncated marble cone as a sacred log, or lignum, dedicated to Diana, of a type known from frescoes from Herculaneum and the House of Livia on the Roman Palatine. Fresh considerations of the excavated evidence, however, incorporating the newly understood principles and manifestations of the planning and design of the Caesarian colony at Corinth, indicate instead that the architectural features of the so-called Apsidal Building represent part of a circus facility.



I owe the original idea of the identification of the Apsidal Building as a circus meta and spina to Williams. (22) The eastern end of the excavated structure is of an appropriate general size and shape for the low wall of a meta, or turning post, of a Roman circus. This feature at Corinth is surrounded on three sides--north, east, and south--by a series of layers of hard-packed road metal. (23) The two lines of parallel blocks extending to the west (Fig. 3: walls 1, 2) could reasonably define the spina of a circus. Furthermore, many of the finds associated with the so-called Apsidal Building can be related to the functions of a meta or spina or to the use of the overall facility as a circus. For comparisons of design, the western turning post (meta secunda) of a circus at Lepcis Magna has been excavated and published and serves as a close parallel (Fig. 10). (24)


The large truncated marble cone described earlier, found above the foundations and near the north wall of the structure, should be interpreted as a part of the turning post itself; the empolion cutting at the top of the cone would have received a finishing element. (25) At least one additional member of the cone should be restored below the preserved piece, and the stone egg may have rested on top (see below). (26) The three lead tablets can also be related to the functioning of the circus; a number of lead tablets have been excavated in circus contexts and proven to be curse tablets. (27) Curse tablets positioned at the starting gates and turning posts of circuses would, by means of "aggressive magic," have served to initiate some form of misfortune on the charioteer and chariot team. (28)

The equally spaced pits within the meta and spina, one of which (pit 2) was excavated with a portion of an amphora still in situ, and the associated channels could be interpreted as a series of connected water basins, hydraulic facilities sometimes found within the spina of circuses. (29) An attractive alternative interpretation is that the pits and inlaid amphoras were receptacles for trees or bushes. (30) Such features, including amphoras used as planting pots, are known from Greek and Roman gardens. (31) The north-south cross-wall at the western end of the meta was built, during the later use of the structure, with a series of reused blocks, including a column fragment, partially covering and filling pit 1.

Immediately to the west of the meta are two other series of blocks, east-west foundations (walls 1 and 2), that continue for a maximum of 12 m (Fig. 3). These blocks have presumably been reset into the original foundation trenches of the walls from the spina. Originally the blocks are likely to have been stacked two to three courses high (see below, Fig. 14), similar to the situation at Lepcis Magna. (32)


The parallel foundation walls excavated to the north and northeast of the spina and meta may well be the foundations for a spectator gallery. It is likely that wall 16 was the foundation for a podium wall. It is customary that circuses have spectator facilities built along the sides of the racecourse and, although the evidence currently indicates their presence only to the north of the circus at Corinth, it would be expected that the gallery also continued around the south and west sides. The 5 m wide series of blocks found at the northern end of the southern portion of trench I (Fig. 7) may have been part of the northern spectator facilities of the circus.

Several objects discovered in the 6th-century dump fill above the east-west foundations of the spina could have had a connection with the circus. (33) The large stone egg could be interpreted as one of the ova commonly used as counters for the laps of the chariot races; eggs were also used to decorate the top surface of the metae. (34) It is curious that a terracotta comic mask (MF-12977) was found in the dump fill. Theatrical performances were often an important aspect of circus activities, and pantomimes were known to have been included in circus programs from the 2nd century until the 6th century. It is conceivable, therefore, that this mask was used in such performances in the circus in Corinth. (35)


The structure identified here as a circus had a long and complicated history and, in addition, was largely pillaged after its period of use. At least four phases can be recognized: construction in the mid- to late Augustan period; renovation in the late 1st century A.D.; reworking in the 6th century; and abandonment later in the 6th century.

The date of the original construction of the circus would appear to be mid- to late Augustan based on several pieces of evidence from the area of walls 1 and 2 and from the pits found within the spina and meta.

Wall 2 (spina)

The original foundation trench of the southern east-west wall of the spina was located beneath wall 2, which was built in the 6th century. The blocks of wall 2 sit on top of earlier layers of sediments that are Late Roman, Early Roman, and mid- to late Augustan in date (Fig. 11). Both east-west walls are associated with the stone drain covered with marble slabs, which was probably built during the 6th-century refurbishing of the structure (Figs. 3, 12). When the blocks of wall 2 were installed, the earlier foundation trench was not dug to its original depth so that the original stratigraphy remains. The earliest level excavated at the bottom of the trench of this wall is described as hard red earth dating to the early 1st century A.D. (36)


Wall 1 (spina)

A trench dug immediately to the north of wall 1, which was also built in the 6th century, revealed material from the early foundation trench, 1.15 m wide, dating to the Early Roman period. (37)

Pits (spina, meta)

The lowest stratigraphic level found at the bottom of pit 3 dates to the mid- to late Augustan period (Figs. 3, 12). (38) A stratigraphic level from the bottom of pit 2, although not precisely datable, can be dated generally to the Roman period. (39) Pit 1 was partially filled with a large column fragment but did not have datable levels.


Wall 16

The heavy foundation wall 16, ca. 18 m north of the spina (Fig. 7), must, on the basis of pottery found to the south of the wall, have been in place by the 2nd century A.D. (40) To the north of wall 16 and extending for a distance of half a meter are several stratigraphic levels that together suggest a dumped fill, probably Roman, but not precisely datable (Figs. 8, 9). (41) The stratigraphic level that covers wall 16 probably dates to the early 6th century. (42)

Curved Slabs of the meta

It is clear from the foundation trench cut into the hard-packed floors to the north, east, and south of the curved marble slabs of the meta that the slabs, mostly laid on bedrock, had been resituated late in antiquity. The foundation trench continues to the outside of some of the poros blocks north and south of the curved end, indicating that the marble slabs and the poros blocks at the eastern end had been replaced or reset at one time. The resetting of the curved marble slabs at the east end of the meta can be dated to the 6th century A.D., allowing us to date the latest use of the entire structure to this time. (43) From another part of the same reutilized foundation trench, at the east end of the meta, comes additional dating evidence to corroborate the 6th-century date. (44) A late-4th-century coin found on the racecourse floor, in hard-packed road metal within a meter of the east end of the meta, indicates a use at that time, or following in the early 5th century. (45) Another 4th-century coin was found on the road metal ca. 20 m to the east. (46)

Important for this discussion is the fact that the curved exterior marble blocks of the meta were reset late in the history of the structure. The earlier meta had undergone renovation and repair, including refurbishing of the eastern exterior of the curved east end, probably as a result of wear and the rising elevation of the road levels. Based on the evidence for dating the road surfaces near the meta, the facility seems to have remained in use into the 6th century, although continuous use cannot be proven. The excavator has suggested that the structure went out of use during the 6th century, after which time a deep garbage dump covered much of the structure. This dumped fill would have been deposited later in the 6th century, after the resetting of the exterior marble slabs of the meta. (47)


It may be useful to summarize the phases of development thus far documented for the circus in Roman Corinth. The original construction took place during the mid- to late Augustan period. The form of the circus would have included the spina, two metae, starting gates (carceres), an arena, and perhaps little else (Fig. 13). (48) The spina and the metae were constructed by digging foundation trenches into the bedrock, into which the foundation blocks were set; the trench for these foundations was 1.15 m wide and 1.2 m deep. Several courses of stone above the foundation level would then have been required to create a low wall for both features (Fig. 14). The north-south cross-wall, ca. 4 m west of the curved end of the meta, suggests that the meta was separated from the spina by a space. The three large pits were probably dug directly into bedrock at the same time, as part of the initial construction of the metae and spina, and may have been intended for the planting of trees or bushes. The total width of the arena in the Augustan phase may have been ca. 55 m (ca. 25.5 m to the north of the spina and 29.5 m to the south) as there is no evidence for any formal spectator accommodation at this time and there appears to have been space for a racecourse of this width (Fig. 13). The carceres are likely to have been built at the east end of the facility (see below).


In the 2nd century A.D. a spectator gallery, with a maximum width of ca. 8 m, (49) was constructed on the north side of the circus, probably extending along the south and west sides as well (Figs. 7, 8, 13). These spectator accommodations would have diminished the total width of the racecourse proper, at least in this area, from ca. 55 to 39 m. (50) The race-course width to the north of the spina would have been restricted to ca. 18 m. The spectator capacity of the facility can be estimated at approximately 12,000. (51)

Over time the racecourse surface was raised as layers of road metal were created on the track. In the 6th century, when the meta and spina were renovated, the original trenches for the foundation walls were reutilized but the refitted blocks were not set fully into the trenches (Figs. 3, 11). By this time the foundation walls (including wall 16) for the spectator facility to the north had gone out of use and been covered (Fig. 9).


It is not known whether carceres were located at the east or west end of the circus. If they were located at the eastern end of the facility, more space would have been available for the total length of the arena. Given topographical constraints, the arena could have been as long as 388 m if the carceres were located at its eastern end, but only ca. 330 m if they were located at its western end, since, in circus design, the space required from the carceres to the meta secunda is greater than the space required from the closed end of the circus to the meta prima (Fig. 13). (52)

Although from the 2nd century on the available arena width between the eastern meta and the spectator facility to the north was only ca. 18 m, thus implying the narrower space associated with a meta prima rather than a meta secunda, other evidence suggests that this was not the case. The location of the foundations for the spectator facilities that are only roughly parallel to the arena and some 35 m east of the eastern meta indicates that the arena continued some distance to the east and did not turn to follow the semicircular end of the circus. The curved end of the arena should thus be found at the western end of the circus. The orientation of the spectator facilities to the north of the circus suggests that the northern spectator wall flared slightly to the north by approximately 1[degrees] 30', providing more space for the starting chariots from the carceres in the east to the meta secunda. (53) The projected location of the carceres is near that of the major cardines that approach the center of the city from the north. For instance, there would have been a major roadway (likely to be cardo IV west) leading north from the Theater to the area of the Asklepieion and the Gymnasium. There is ample space in the plan of the city for this possibility and the arrangement would thus be similar to examples elsewhere in the Roman world. (54)

The identification of the eastern meta and spina of a Roman circus in Corinth provides an important part of the plan of the Roman colony of 44 B.C. (55) As a result of research carried out by the Corinth Computer Project since 1988, it has been possible to reconstruct a "drawing board plan" of the Caesarian colony, covering 240.6 hectares and divided into four centuries, or quadrants, each of 240 iugera (Figs. 15, 16). (56) The urban area is arranged lengthwise with respect to the north-south axis of the colony and probably had as its dominant unit an area measuring 1 x 2 actus (1 actus = 120 feet or 35.4 m). The primary north-south orientation of the grid of the Roman city is ca. 3[degrees] west of north (N3[degrees] 3' 46" W), based on the orientation of the east curb of the Lechaion Road as the cardo maximus of the colony. The principal surveyed cardines of the city are oriented generally 2-3[degrees] west of north; the decumani are generally 2-3[degrees] south of west. The average orientation of the east-west foundations of the meta of the circus are 2[degrees] 47' 51" south of west, which is within 16' from the principal east-west orientation of the Roman colony. Typically the metae and spina of a circus are oriented obliquely to the axis of the arena in order to provide sufficient space to the right of the meta secunda for the starting chariots. This appears to be the case at Corinth, where there is, as a result of the oblique planning, slightly more room to the north of the spina, in front of the attested spectator facilities, than there is to the south.


The location of the meta as a part of the spina of the circus falls close to the middle of a 2-actus north-south dividing line within the organization of the city (Figs. 15, 16). It is known that a major east-west roadway, decumanus III north, lay 3 actus to the south of the meta, and I have suggested elsewhere that the northern east-west limes of the city was located ca. 1 actus to the north of the meta. (57) Decumanus IV north would then have been located immediately to the south of the circus and decumanus V north immediately to the north (Fig. 16). Within the 2-actus space projected as the maximum potential width of the circus, there is slightly more room to the south of the axis of the meta than there is to the north, 38.2 m and 32.6 m, respectively, although it is clear that the entire width of the 2-actus space was not utilized. There can be little doubt that the orientation and location of the circus were planned with respect to the layout of the Roman city since the total width of the circus would fit into the 2-actus block, including room for spectator facilities on the retaining walls to the north, and presumably also the south.

Without the spectator facilities included in the planning, the available space for the racecourse, as noted above, would have been ca. 55 m (Fig. 13). Furthermore, space remains in the urban plan for an east-west Roman roadway to the north of the circus, decumanus V north (Fig. 16). (58)


The general topography of the northwest quadrant of the Roman city is flat, providing ample room for a circus structure with an overall total length of ca. 400 m and width of 65 m. The resulting length of the arena would be 388 m and the width 43 m (Fig. 13). The Roman circus was located to the south and west of the Asklepieion and directly south of the Gymnasium and the Fountain of the Lamps (Fig. 16). (59) The northwest quadrant of the Roman city appears to have been filled with civic and religious structures. Although technically the Asklepieion and the Gymnasium were outside the limites of the Roman city, the Odeion, the Theater, a bath to the north of the Theater, and the circus would have been within the city proper.

Other facilities were constructed in the area of the eastern meta (Figs. 7, 17, 18). The area immediately to the north, identified by Wiseman as the West Waterworks, includes a large cistern and a bronze-casting pit and was possibly an industrial installation. This complex, 15 m in its north-south dimension, appears to have had its southern wall built up against the already existing spectator gallery on the north side of the racecourse (the northern element of which is now represented by a robbing trench). Further to the east and immediately south of the east-west Gymnasium stoa lies the rubble concrete foundation of the so-called Domed Building. It appears to have been built up against the existing spectator gallery further to the east and was also presumably used for industrial purposes.


Wiseman dated the construction of all of these facilities to the 1st-2nd centuries A.D., suggesting that they remained in use until the 4th century. (60) These buildings were built on available land immediately to the north of the circus, where, as just noted, the decumanus V north is likely to have passed. Another structure in the area, the Fountain of the Lamps, approximately 70 m north of the circus, had a long history in the Greek and Roman periods as a bath; it was heavily utilized in the 4th and 5th centuries and possibly into the 6th century. (61) Elements identified as being from a Late Roman wall, built in the late 4th or early 5th century A.D., were excavated between the Fountain of the Lamps and the circus (Figs. 7, 17). (62)

The location of the Corinth circus with respect to the Caesarian city plan--at the northern limit of the city and tucked next to the northern east-west city limes (Fig. 15)--is worthy of mention. For comparison, the amphitheater is located at the northeast corner of the colonial plan, approximately 1,200 m to the east. Given that the excavated meta and spina appear to be aligned with the orientation of the Roman colony, and that the typology of the circus design as well as the dating evidence of the spina suggests a mid-late Augustan date for the original construction, it seems likely that the circus was a part of the early planning of the colony. (63) The eastern meta clearly had a long life in this location since its exterior curved marble slabs were reset into the earlier layers of road metal in the 6th century.

To the west of the eastern meta, sufficient space exists for the circus to extend some 290 m before intersecting with the Greek circuit wall that defines the northern limit of the city; there are no other apparent topographical restrictions. The only excavation undertaken to the west of the meta, as noted above, was that by Richardson in 1896 (Fig. 7: trench I). (64) To the north, there is approximately 45 m of space before the long east-west stoa of the Gymnasium. The restored plan of the north-central area of the Roman city ca. A.D. 150 is shown in Figure 18.


The size of Roman circuses varies considerably. At Lepcis Magna, for instance, the spina was 231 m long and 6.2 m wide between the faces of the outer walls, and the overall length of the arena was 450 m. (65) At Lepcis the distance between the meta prima to the curved end of the arena is much shorter (ca. 50 m) than the distance from the meta secunda to the carceres (ca. 160 m). This spacing has to do with the nature of the start of the race: a considerable distance between the carceres and the meta secunda is needed to allow the starting teams to run an equal distance and have enough room to clear the barrier. (66)

Given the known widths of starting gates in Roman circuses, one should allow ca. 2.7-3.1 m on the arena floor for each chariot team. (67) At Lepcis Magna the 64 m wide arena (32 m for half) provided room for approximately 12 chariot teams. The 55 m wide arena at Corinth in the Augustan period would have allowed approximately eight chariot teams to compete. This figure includes some room for the retaining walls and spectator areas, north and south. As noted above, however, by A.D. 100 the width of the track at Corinth was reduced to ca. 39 m by the construction of spectator facilities, which may have limited the number of teams to six.

Typologically, the circus design at Corinth is similar to that of other circuses constructed in the Augustan period. The general characteristics are the relatively short length and narrow width of the arena, the narrow spina, and the shallow area reserved for spectators. For comparison, the circus found at Bovillae, approximately 11 miles south of Rome on the Via Appia, is similar in some respects, with an arena about 328.5 m long and 60 m wide (Fig. 19). (68) The excavated spina is 2.5 m wide, leaving a track width of approximately 28.75 m per side. The circus at Bovillae is dated to the Augustan period, or not much later, on the basis of construction materials, and Humphrey suggests that "in the early first century A.D. a circus with an arena length of ca. 330 m and an overall length of 350 m was considered minimally adequate." (69)


From recent excavations it is now known that at Caesarea Maritima a very small circus facility was part of the Herodian palace complex of the 1st century B.C. (Fig. 20). (70) The length of the arena is at least 265 m and its width 50.35 m. Other relatively small facilities for equestrian contests have been identified in the Roman East, for instance at Neapolis (Nablus) (71) and Gerasa in the 2nd century A.D. (Fig. 20). (72) The length of the arena at Gerasa is 245 m and the width is 50 m; the circus functioned with only five starting gates, implying that there would have been only five chariot teams entered in a race. (73)


At Corinth the size and shape of the circus suggests a hybrid form based on the narrow width of the western and eastern examples and a length more closely comparing with the western example at Bovillae. As noted above, there probably would have been no more than six starting gates after the spectator stands were built in the 2nd century, although there may originally have been room for eight gates.

No other circuses are known in Roman Achaea besides those originally Greek hippodromes that were used in the Roman period. A Roman circus is documented at Gortyn in Crete (administratively a part of Roman North Africa); another must have existed at Nikopolis in Epirus, and a Late Roman example is known in Thessaloniki. (74) Several circuses are known from Roman Asia Minor. (75)


Chariot races were a part of two agonistic festivals associated with Corinth, the Panhellenic Isthmian Games and the Caesarea. (76) The two appear to have had some relationship with each other since it is known that the same agonothetes presided over both festivals, and the Isthmian Games (trieteric) preceded the Caesarea (pentaeteric). (77) It has often been assumed that all the contests were held at the Sanctuary of Poseidon at Isthmia, although convincing epigraphical evidence indicates that both the Isthmian and the Caesarea contests were sometimes held together in Corinth. (78) Gebhard notes that the celebration of the Isthmian Games is likely to have remained in Corinth until the principate of Nero and that, from their founding in the Augustan period until the 2nd century, the Caesarea games were also usually held there. (79) Such an arrangement would have required a stadium and circus for the agonistic contests and a theater for the musical contests of these festivals. Although a Roman stadium has not been discovered in the city plan, the Roman circus is now identi.ed, with the east meta only 275 m north of the orchestra of the Theater (Fig. 15). (80) The equestrian events of the Caesarea, and possibly the Isthmian contests at times, presumably took place in the circus. It is also possible that other events connected with the two festivals could have been held in this area. (81)

A. B. West has suggested that the Caesarea festival at Corinth was probably founded in 30 B.C. following the battle at Actium and the victory of Octavian, although this date is not secure. (82) The earliest inscription at Corinth that mentions the Caesarea is a partial victor list on a herm that dates to A.D. 3. (83) Suetonius mentions that many Caesareas were founded all over the Roman world in honor of Octavian and it is clear that the festival at Corinth was one of them. (84) It thus seems likely that the Caesarea games were first held at Corinth in 30 B.C. and that earlier equestrian contests were held in a less formal predecessor of the circus. As discussed above, although the construction of the circus cannot be dated before the mid-late Augustan period, the original plan of the Caesarian colony apparently included a space for the circus, and we can thus assume that the races were held in the same area in a less formal structure.

The equestrian events of the Caesarea at Isthmia are known from a series of inscriptions. I-2740, for example, includes lists of musical, athletic, and equestrian victors in the Caesarea of spring A.D. 127. (85) Included are equestrian events (for the most part these are restored terms), including the armed chariot ([TEXT NOT REPRODUCIBLE IN ASCII]), line 65; two-horse chariot for young horses ([TEXT NOT REPRODUCIBLE IN ASCII]), line 65; horse race for young horses ([TEXT NOT REPRODUCIBLE IN ASCII]), line 65; two-horse chariot for full-grown horses ([TEXT NOT REPRODUCIBLE IN ASCII]), line 70; four-horse chariot for young horses ([TEXT NOT REPRODUCIBLE IN ASCII]), line 70; horse race for full-grown horses ([TEXT NOT REPRODUCIBLE IN ASCII]), line 75; four-horse chariot for full-grown horses ([TEXT NOT REPRODUCIBLE IN ASCII]), line 75; armed chariot race ([TEXT NOT REPRODUCIBLE IN ASCII]), line 75; mounting and dismounting chariot race ([TEXT REPRODUCIBLE IN ASCII]), line 85. (86)

That a circus would be constructed within the limites of the Roman city of Corinth is of great importance for the understanding of the religious and cultural development of the city and has many implications. Furthermore, games in honor of Octavian and Poseidon, held within the city of Corinth, would have drawn interest from around the Greek and Roman world. The equestrian events alone of both of these festivals must have attracted a large audience of spectators.


The circus has been shown above to be an integral part of the Roman city plan of Corinth. From what is known of the archaeological history of a portion of the spina and the eastern meta of the structure, it appears that the circus had a long life in the northwest quadrant of the city, beginning in the mid- to late Augustan period, and continuing into the 6th century. Furthermore, the location and orientation of the structure suggest the utilization of the Caesarian plan of the Roman city. The circus's best architectural parallels are found in the west at Bovillae outside Rome as well as in the eastern examples at Caesarea Maritima, Gerasa, and Neapolis. The facility in Corinth is likely to have been the site of the equestrian contests of the Corinthian Caesarea and, on occasion, of the Panhellenic Isthmian Games. Use of the facility into the 6th century suggests that chariot racing continued to be held in Corinth during the later Empire. (87) Further excavation in this area of the site would elucidate the history of the structure as well as provide more details of its physical characteristics and appearance.

(1.) I thank John H. Humphrey, David R. Jordan, Guy D. R. Sanders, Kathleen Warner Slane, Charles K. Williams II, and James R. Wiseman for assistance and helpful discussions concerning the subject of this article. In particular, I would like to thank Slane for the pottery identification and description, and Williams for assistance with the stratigraphic analysis. I am also grateful to Wiseman for supplying the photographs for Figs. 1, 2, 4, and 6. The drawings were produced by the Corinth Computer Project of the University of Pennsylvania Museum of Archaeology and Anthropology. I thank all of the students who have helped me both in Corinth and in Philadelphia, and in particular Timothy Demorest, David Pacifico, and Nicholas L. Stapp for their assistance in preparing the final drawings. Finally, I thank the anonymous Hesperia reviewers for their comments and suggestions on an earlier version of this article.

References to the Corinth excavation notebooks are abbreviated as "Corinth NB" throughout.

(2.) For the destruction of Corinth, see Paus. 2.1.2; Strabo 8.6.23; Livy Per. 52. For the damage to the city, see Wiseman 1979, pp. 491-496. For discussion of the planning of the colony, see Romano 1993. For the colonists, see Strabo 8.6.23 (freedmen); Plut. Caes. 57 (veterans). See also the recent study of the early colony at Corinth by Walbank (1997). For activity during the interim period, see Williams 1978, pp. 21-22; Romano 1994, pp. 62-64.

(3.) From the archaeological record one must assume that there was a partial and selective physical destruction of the buildings and structures of the Greek city. See Wiseman 1979, pp. 491-496.

(4.) Welch 1999, pp. 133-140. Welch suggests that the date of the structure is based on its construction techniques, which are most similar to those of other amphitheaters built during the late 1st century B.C., Sutrium in Etruria and Carmo in Baetica. Furthermore, she points out that, for both social and political reasons, an amphitheater would have been a necessity in a Roman colony founded by Julius Caesar. The location of the amphitheater at Corinth in the northeast corner of the colony is similar to the locations of amphitheaters in other Roman colonies, just inside or outside the city limits.

(5.) The Flavian colony is known from numismatic, epigraphical, and (recently) archaeological evidence. See Corinth VI, p. 26, no. 93; Corinth VIII.3, p. 42, no. 82; Romano 2000; 2003, pp. 291-299. The second Roman colony is likely to have introduced new colonists into the city.

(6.) The Lex Coloniae Genetivae, the colonial charter of Urso, a foundation of Julius Caesar in 44 B.C., indicates that the aediles were to organize games in the circus or gladiatorial contests in the forum. Each aedile was to spend not less than 2,000 sesterces from his own money, to be supplemented by 1,000 sesterces of public money. See Crawford 1996, vol. 1, pp. 423-424. It is possible that the same law was a part of the colonial charter of the Caesarian foundation at Corinth.

(7.) Humphrey 1986, pp. 71-72; Beacham 1999, pp. 1-44.

(8.) It is very likely that Greek Corinth had a hippodrome although its location has not been discovered. It must remain a possibility that the site of the Roman circus could at an earlier date have been the site of the Greek facility although Humphrey (1986, p. 12) reminds us that "there is no Roman example of a Classical Greek hippodrome having been converted into a canonical--i.e. monumental--Roman circus."

(9.) Wiseman 1969, pp. 64-72, pl. 20:b, c. For the excavations in the Gymnasium area, see also Wiseman 1967a, 1967b, 1972. I thank Professor Wiseman for discussing with me elements of this structure and for permission to cite unpublished information from the excavation notebooks.

(10.) One of the blocks, "ag" in Fig. 3, appears to have had an earlier use as a starting-line block from the dromos of a gymnasium or stadium.

(11.) Wiseman (1969, p. 69) gives the following dimensions: for pit 1, Diam. 1.65 m, D. 0.88 m; pit 2: Diam. 1.65 m, D. 0.79 m; pit 3: Diam. 1.45 m, D. 0.80 m. There is no indication of any hydraulic cement from within the pits.

(12.) A-767: max. p.L. 1.38 m, Diam. (top) 0.28 m, max. p.Circum. 1.42 m. Williams (1987, p. 386) describes the shaft of the marble cone as "pick-dressed" and notes the presence of two "shallow boss-like protrusions" that are visible on one side of the shaft, 0.78 m below the horizontal top of the cone.

(13.) S-2879: p.L. 0.384 m, max. Diam. 0.19 m. For photographs of these objects, see Wiseman 1969, pls. 21:c, 22:a-c, 24:c. The excavator found four additional fragments of terracotta masks nearby: MF-12949 and MF-12972, from a 3rd-century A.D. deposit in a bronze-casting pit; MF-12967, from a 1st-2nd-century deposit from test trench 8, west of a bronze foundry; and MF-12968, from a Late Roman deposit from the east end of the Gymnasium south stoa. See Wiseman 1969, pp. 103-104, pls. 22:j-l, 32:c.

(14.) Wiseman 1969, p. 72, pl. 21:c.

(15.) MF-12994 (Fig. 3:A) and MF-68-271 (Fig. 3:B) were found in layers of hard-packed road metal to the east and northeast of the eastern end of the structure (see below, nn. 23, 44). MF-68-272 (Fig. 3:C) was found in the foundation trench for the marble slabs at the northeast side of the curved east end of the structure (Corinth NB 379, pp. 144-149; NB 418, pp. 149-152, 169-172). MF-12994 is illustrated in pl. 21:a in Wiseman 1969. See also Jordan 1985, pp. 166-167.

(16.) Wiseman 1969, p. 68, fig. 2, although this drawing does not show the entire length of the excavated wall.

(17.) Corinth NB 453, p. 189.

(18.) This lead tablet (MF-68-273) largely disintegrated during excavation. See Corinth NB 350, pp. 30-32; Wiseman 1969, p. 70, n. 10.

(19.) Wheel ruts were found at the bottom of test trench 1 of trench "Gym South 1," to the south of these foundations; Wiseman 1967b, pp. 420-421. The elevation of these ruts is 54.150 m. For comparison, in test trench 9, at the east end of the Apsidal Building (Corinth NB 425, p. 71), the elevation of the hard floor surface dating to the 4th century is 54.363-54.358 m.

(20.) Richardson 1897, pp. 457-458. Wiseman (1967b, pl. 92) reproduces the plan by W. B. Dinsmoor of the 1896 trenches.

(21.) Williams 1987.

(22.) Charles K. Williams II (pers. comm.). Williams suggested this possibility to me in 1978 although, as noted above, he later (1987) published the structure as the shrine to Diana Nemorensis.

(23.) The daily excavation record (Corinth NB 425, pp. 71-87) describes a series of hard-packed floors, up to 0.50 m deep, several of which date to the 4th century.

(24.) The example at Lepcis Magna measures ca. 6.9 m (N-S) and 5.8 m (E-W). At Lepcis Magna the meta is detached from the spina as appears to also be the case in Corinth.

(25.) The marble cone should probably be associated with the meta in its latest use.

(26.) Humphrey (1986, pp. 46-47, fig. 27) restores the cones at Lepcis Magna as 4.06 m tall (4.75 m, including the crowning egg).

(27.) See Heintz 1998. Wiseman (1969, p. 70, pl. 21:a, b) describes and illustrates one of the lead tablets, MF-12994, currently under study by David Jordan. Wiseman writes that "the script is Greek; a few of the letters are 'printed' but most of the writing is cursive." This curse tablet is associated with pottery lot 4881, dating to the late 4th-5th century. It was found at an elevation of 54.521 m. MF-68-271 (elev. 54.554 m) is associated with pottery lot 5555, dating to the 6th century, and MF-68-272 (elev. 54.518 m) with pottery lot 5568, dating generally to the Roman period.

(28.) At Antioch, for example, five of the 13 curse tablets found in the city were located in drains along the central barrier and turning posts of the circus; Heintz 2000. I am grateful to David Jordan for the information that several curse tablets from Carthage and Hadrumentum curse race horses named Korinthos (which is spelled Corentus in the Hadrumentine example), suggesting that the horses may have come from Corinth, or at least that Corinth was a good source of horses. Although this coincidence does not necessarily presuppose the existence of a circus or chariot racing at Corinth, it is suggestive. For examples of horses named Korinthos, see Audollent 1904, pp. 311-312, no. 235, line 27; 1906, p. 386.

(29.) Cf., e.g., the western meta at Lepcis Magna (Fig. 10). In reference to the barrier of the Circus Maximus, Humphrey (1986, p. 294) observes: "in the High Empire, and possibly at the moment when the euripus was first built, certain monuments were converted into fountains."

(30.) I am grateful to one of the anonymous Hesperia reviewers for suggesting this idea to me.

(31.) See Gleason 1994, p. 16. Such planting pots, known as ollae perforatae to the Romans, have been discovered at Pompeii and Jericho; see Messineo 1984 for examples. At the circus of Lepcis Magna, Humphrey (1986, p. 38) mentions that plants or trees could have been included on the barrier.

(32.) Humphrey 1986, p. 39, fig. 18.

(33.) It is not of course certain that the objects associated with the dump fill come from the circus.

(34.) See, e.g., Humphrey 1986, pp. 260-262, for discussion of the literary and archaeological evidence for the eggs from the Circus Maximus.

(35.) MF-12977: Wiseman 1969, pp. 71, 104, fig. 3, pl. 22:a. For comparanda for the mask, see Agora VI, pp. 20-23, 59-62, pls. 11-15. For the role of theater in the circus, see Cameron 1976, pp. 193-229. For mime and pantomime in the Roman theater, see Beacham 1991, pp. 117-153.

(36.) Pottery lot 5530. Corinth NB 418, p. 95; NB 433 (elev. 53.908-53.558 m). The soil is described as reddish brown. Approximately half of the pottery is pre-Roman, mostly Hellenistic. The other half is Early Roman, and includes two unidentified red-slipped plate fragments of the late 1st century B.C., two casserole fragments (one floor and one rim), one cooking-ware fragment, and two plain-ware bases.

(37.) Pottery lot 5575. Corinth NB 425, p. 80; NB 433 (elev. 54.644-53.636 m). The soil is described as light brown and hard. The pottery includes Greek black glaze, coarse wares, unfired Corinthian, a micaceous water jug fragment, and an Early Roman lekane rim fragment. This trench was excavated as test trench 13.

(38.) Pottery lot 5561. Corinth NB 418, pp. 131, 136, 189; NB 425, p. 43; NB 433 (elev. 54.634-53.373). The soil is described as sandy red. The pottery includes a handful of Hellenistic cooking-pot sherds, an unidentified Roman amphora fragment, a Roman cooking-pot rim, two fragments of thin-walled mid-late Augustan cups, and a Roman casserole handle.

(39.) Pottery lot 5533. Corinth NB 418, pp. 104, 107, 111; NB 433 (elev. 54.168-53.810 m). The soil is described as hard gray-brown to red-brown. The pottery includes a red-slipped plate like that found in pottery lot 5530 (see n. 36, above), one micaceous sherd that is not from a water jar, and three fragments of an amphora. The amphora has been tentatively identified as either a North Italian Dressel 6 or Spanish Dressel 7-11 amphora. It had probably been used for collecting water within pit 2, possibly for a tree or bush, as its neck and foot had been cut off. North Italian Dressel 6 amphoras date to the 1st century A.D. and the Spanish Dressel 7-11 type from the Flavian to the early Antonine period, until about A.D. 150. It would therefore be safe to say that this amphora was set in place in pit 2 in the 1st or 2nd century. Since the pits were probably dug when the structure was first constructed, the amphora was presumably a replacement for an earlier one.

(40.) Pottery lot 5902. Corinth NB 425, pp. 130, 140, 165-166; NB 433 (elev. 54.921-54.781 m). The soil is described as dark reddish brown. The pottery includes two Italian Sigillata fragments (one of which is a foot of a Haltern 8 vessel), a battered fragment of Eastern Sigillata B, a possible gray-ware fragment, a local red-slipped fragment, a fragment from a 1st-century A.D. white-slipped pitcher, two Italian thin-walled wares, a thin-walled (probably) local cup fragment, a local small round-mouthed pitcher, and a lekane fragment. Two black micaceous water jars can be dated to the late 1st century, and a gritty brown wheel-ridged sherd, possibly from an amphora, dates to the late 1st or beginning of the 2nd century. Two joining fragments with red slip derive from the late 1st, 2nd, or possibly even 3rd century.

(41.) Pottery lot 5899. Corinth NB 425, pp. 135-136, 139-140, 142, 148, 178-179; NB 433 (elev. 54.914-54.596 m). The soil is described as reddish brown, with and without stones. Almost all of the pottery is Classical and Hellenistic, including a fragment of a Corinthian B transport amphora, a fragment of a Corinthian A transport amphora toe, and several pre-Roman cooking-pot fragments. Possibly Roman are a probable bulbous unguentarium fragment, an Eastern Sigillata A fragment, and a possible burned Sigillata fragment. A piece of poros stucco, some waterproof cement, and a piece of bronze were also found.

(42.) Pottery lot 5898. Corinth NB 425, pp. 125, 128, 132, 137, 158, 160, 163, 176; NB 453, pp. 22-23; NB 433 (elev. 55.166-54.850 m). The soil is described as light brown to red-brown. This level includes some Classical pottery, a fragment of an Early Roman red-ware plate, an Early Roman pitcher handle, an Early Roman pitcher foot (cooking fabric), a fragment of a grooved basin rim, and two fragments of an Early Roman water pipe. The Late Roman material includes a Late Roman C form 3B (mid- or second half of the 5th century), three fragments of African Red Slip (3rd or 4th century), a Hayes 104C rim (525-600; see Hayes 1972, pp. 160-166), a cooking-pot handle, a folded-rim bowl fragment, an imported Late Roman Aegean micaceous cooking-pot fragment (5th-early 6th century), and possibly Palestinian amphora sherds. Distinctive 6th-century amphora sherds are missing.

(43.) Pottery lot 5557. Corinth NB 418, pp. 166-167; NB 433 (elev. 54.577-54.513 m). The soil is described as light brown and rocky. The finds include a fragment of a Late Roman micaceous water jar, an Early Roman double-rolled handle, several fragments of a Late Roman amphora (probably Palestinian), and a fragment of a Late Roman water pipe.

(44.) Pottery lot 5555. Corinth NB 418, pp. 149, 152, 166; NB 433 (elev. 54.687-54.566 m). The soil is described as light red-brown to hard brown with pebbles. The pottery includes two handles from Late Roman micaceous water jars, two fragments of Late Roman amphoras of type 2, two fragments of a Palestinian amphora, a toe from a Gaza-type amphora, a cooking-pot rim (probably 6th century), a fragment of 3rd-century African Red Slip ware, and a probable 6th-century African Red Slip ware fragment. Also contained in this pottery lot is a lead curse tablet, MF-68-271.

(45.) Corinth NB 418, p. 174: a test trench through the hard road metal, elev. 54.566-54.484 m. The coin (68-1428) is datable to Theodosius I, Valentinian II, or Arcadius, A.D. 383-388; I thank Orestes Zervos for this information.

(46.) Corinth NB 350, p. 87: a hard-packed area, elev. 54.889 m. Coin 66-563, of Constantius II or Constans, A.D. 341-346, was found in this area. I thank Zervos for this identification.

(47.) Wiseman 1969, p. 71. As noted above (n. 13), the terracotta mask MF-12977 and several marble fragments were discovered in this dump.

(48.) It is by no means certain that the spina of the circus was continuous from the east end to the west end.

(49.) The width of the spectator gallery is based on the dimensions of the excavated foundations on the north side of the circus.

(50.) In Fig. 8, the sarcophagus to the south of wall 16 is Greek (pottery lot 5923). Covering the sarcophagus is a deposit, pottery lot 5909, dating to the 1st century A.D. See Corinth NB 453, pp. 45-50.

(51.) This estimate is based on the formula of 0.5 [m.sup.2] per spectator (Romano 1993b, p. 22). Hansen (1996, pp. 25-29) discusses minimum space requirements for individuals standing and sitting in areas of assembly and suggests a smaller figure of 0.4 [m.sup.2].

(52.) Humphrey (1986, p. 334) states that "the preferred orientation for a circus [in North Africa] seems to be that the starting gates should be at the west, the semicircular end at the east."

(53.) This arrangement is similar to the hypothetical plan illustrated by Humphrey (1986, p. 22, fig. 5:c).

(54.) See, e.g., Ribera 1998 for the recently discovered circus at Valentia (Hispania Tarraconensis). There, in the reconstruction of the circus, the carceres are only ca. 150 m to the east of the forum and immediately to the south of the decumanus maximus, very close to the center of town. Other examples of Roman circuses, especially later ones, e.g., in Thessaloniki, were constructed close to a major thoroughfare and next to a palace; they are likely to have crossed earlier city streets.

(55.) See Romano 1993, pp. 13-21, figs. 3-5, 7.

(56.) The Corinth Computer Project of the University of Pennsylvania Museum of Archaeology and Anthropology was directed from 1988 to 1997 by the author. The project was carried out under the auspices of the American School of Classical Studies at Athens, Corinth Excavations, C. K. Williams II, Director.

(57.) Romano 1993, p. 20.

(58.) This roadway was put out of use by the construction of the West Waterworks and the Domed Building (see below).

(59.) For discussion of the Gymnasium and the Fountain of the Lamps, see Wiseman 1972.

(60.) Wiseman 1967a, pp. 21-23; 1967b, pp. 413-416; 1969, pp. 67-69.

(61.) Wiseman 1969, pp. 77-78.

(62.) This wall is also known as the Epistyle Wall. See Wiseman 1967b, pp. 410-413; 1969, pp. 87-92; 1972, pp. 5-7; Gregory 1979. The placement of this wall would have limited the circus at its western end (Gregory 1979, p. 265, fig. 1). Since the wall was probably built in the early 5th century, and the circus continued to function into the 6th century, the wall would presumably have respected the circus on the west, as we know that it does to the north. The circus, therefore, cannot have extended all the way to the Greek city circuit wall since there needs to have been room inside the Greek perimeter for the construction of the Late Roman wall. See, however, Slane and Sanders 2005, p. 293, for a proposed new course for the line of the wall.

(63.) In summarizing the location of the North African circuses, Humphrey (1986, pp. 333-334) writes that "circuses were usually found on the out-skirts of towns, outside the built-up areas. In most cases they were built outside the regular street grid. Only at Carthage does the circus seem to follow the orientation of the grid and to lie within the area allocated to the grid; it was however at the extreme edge of the city, as is shown by the fact that the Late Roman city wall passed immediately around it." The situation at Corinth is very similar to that at Carthage.

(64.) See n. 20, above. The open space to the west provides adequate room for the circus to be accommodated within the Late Roman wall; see n. 62, above.

(65.) Humphrey 1986, pp. 36-47.

(66.) Humphrey 1986, pp. 19-24; see also Humphrey 1996.

(67.) Humphrey (1986, pp. 48-49) records the width of the starting gates at Lepcis Magna as ca. 2.7 m, and Patrich (2001, p. 275) reports the width of the carceres at Caesarea Maritima as 3.1 m.

(68.) For a summary of the evidence at Bovillae, see Humphrey 1986, pp. 561-566.

(69.) Humphrey 1986, p. 566. According to the plan of Bovillae published by Dobosi and reproduced by Humphrey (p. 563, fig. 268), the spina appears to lie parallel to the axis of the circus, although Humphrey comments that this is not proven.

(70.) Porath 1995; see also Patrich 2001. The facility was definitely used for equestrian contests because elements of the starting gates have been found. According to Porath (p. 17), excavation in the facility revealed "two parallel subterranean walls 1.7 m apart ... on the central longitudinal axis of the arena. [The walls] may have supported a vault and served as a dividing device for the length of the track."

(71.) Porath 1995, p. 25.

(72.) Humphrey 1996, pp. 125-129.

(73.) Ostrasz 1989.

(74.) Humphrey 1986, pp. 523-524 (Gortyn); 525 (Nikopolis); 625-631 (Thessaloniki).

(75.) Humphrey 1986, pp. 525-528.

(76.) For a general consideration of Roman festivals in Greece, see Spawforth 1989.

(77.) See Geagan 1968, p. 71.

(78.) Corinth VIII.2, pp. 63-66, no. 81; and more recently, Gebhard 1993.

(79.) Gebhard 1993, p. 87, n. 35. See Gebhard, Hemans, and Hayes 1998, pp. 416-417, for the chronology of the return of the Isthmian Games to the Isthmus.

(80.) For the Theater in Roman times, see Corinth IX.3, pp. 4-8.

(81.) Porath (1995, pp. 26-27) suggests that the multipurpose facility in Caesarea Maritima was used for dramatic, athletic, and equestrian contests. The 6th-century circus program from Oxyrhynchus (Cameron 1976, frontispiece) included six chariot races interspersed with processions and performances by singers, athletes, and mimes; P Oxy. 2707 in Rea 1968. If the same were true for Corinth, several types of events, athletic as well as equestrian, of the Caesarea and Isthmian Games could have been held there. There is no evidence to suggest that there would have been faction racing at Corinth.

(82.) For a history of the Caesarea at Corinth, see the discussion in Corinth VIII.2, pp. 64-66.

(83.) Corinth VIII.1, pp. 14-18, no. 14 (I-751). The herm, found in 1917 in the area of the Gymnasium, records a partial list of athletic, equestrian, and musical victors in the Isthmia Caesarea. West (Corinth VIII.2, pp. 64-66) explains that if the festival was established in 30 B.C., then the celebration of A.D. 3 would be the ninth of the series.

(84.) Suet. Aug. 59.

(85.) Biers and Geagan 1970; see esp. p. 84, n. 10, for evidence that the Caesarea and the Isthmian Games were held together.

(86.) The inscription I-2740 was found in the Great Bath on the Lechaion Road, reused in a 6th-century latrine. The original location of the inscription is not known, but its place of reuse is only 400 m southeast of the eastern meta of the Roman circus. In addition to the herm noted above (n. 83), another herm (I-70-39) was found in the lowest levels of the pool of the Fountain of the Lamps, 115 m northwest of the eastern meta. This herm is likely to date to the mid-1st century A.D. and includes a list of athletic officials and victors; see Wiseman 1972, p. 20. Herms were an important element of Roman circuses, and were often set up between the starting gates; see Humphrey 1986, pp. 138-151.

(87.) For references to chariot racing in the eastern Empire, see Cameron 1976, pp. 203-213.


Agora VI = C. Grandjouan, Terracottas and Plastic Lamps of the Roman Period (Agora VI), Princeton 1961.

Audollent, A. 1904. Defixionum tabellae, Paris.

--. 1906. "Rapport des 'tabellae defixionum' recemment decouvertes a Sousse (Tunisie)," Bulletin archeologique du Comite des travaux historiques et scientifiques II, pp. 378-387.

Beacham, R. C. 1991. The Roman Theatre and Its Audience, London.

--. 1999. Spectacle Entertainments of Early Imperial Rome, New Haven.

Biers, W. R., and D. J. Geagan. 1970. "A New List of Victors in the Caesarea at Isthmia," Hesperia 39, pp. 79-93.

Cameron, A. 1976. Circus Factions: Blues and Greens at Rome and Byzantium, Oxford.

Corinth = Corinth: Results of Excavations Conducted by the American School of Classical Studies at Athens

II = R. Stillwell, The Theatre, Princeton 1952.

VI = K. M. Edwards, Coins, 1896-1929, Cambridge, Mass., 1933.

VIII.1 = B. D. Meritt, Greek Inscriptions, 1896-1927, Cambridge, Mass., 1931.

VIII.2 = A. B. West, Latin Inscriptions, 1896-1926, Cambridge, Mass., 1931.

VIII.3 = J. H. Kent, The Inscriptions, 1926-1950, Princeton 1966.

IX.3 = M. C. Sturgeon, Sculpture: The Assemblage from the Theater, Princeton 2004.

XX = Corinth, the Centenary: 1896-1996, ed. C. K. Williams II and N. Bookidis, Princeton 2003.

Crawford, M. H., ed. 1996. Roman Statutes (BICS Suppl. 64), 2 vols., London.

Geagan, D. 1968. "Notes on the Agonistic Institutions of Roman Corinth," GRBS 9, pp. 69-80.

Gebhard, E. R. 1993. "The Isthmian Games and the Sanctuary of Poseidon in the Early Empire," in The Corinthia in the Roman Period (JRA Suppl. 8), ed. T. E. Gregory, Ann Arbor, pp. 78-94.

Gebhard, E. R., F. P. Hemans, and J. W. Hayes. 1998. "University of Chicago Excavations at Isthmia, 1989: III," Hesperia 67, pp. 405-456.

Gleason, K. 1994. "To Bound and To Cultivate: An Introduction to the Archaeology of Gardens and Fields," in The Archaeology of Garden and Field, ed. N. F. Miller and K. Gleason, Philadelphia, pp. 1-24.

Gregory, T. E. 1979. "The Late Roman Wall at Corinth," Hesperia 48, pp. 264-280.

Hansen, M. H. 1996. "Reflections on the Number of Citizens Accommodated in the Assembly Place on the Pnyx," in The Pnyx in the History of Athens, ed. B. Forsen and G. Stanton (BICS 55), Helsinki.

Hayes, J. W. 1972. Late Roman Pottery, London.

Heintz, F. 1998. "The Archaeological Contexts of Circus Curses," JRA 11, pp. 337-342.

--. 2000. "Magic Tablets and the Games at Antioch," in Antioch: The Lost Ancient City, ed. C. Kondoleon, Princeton, pp. 163-167.

Humphrey, J. H. 1986. Roman Circuses: Arenas for Chariot Racing, Berkeley.

--. 1996. "'Amphitheatrical' Hippo-Stadia," in Caesarea Maritima: A Retrospective after Two Millennia, ed. A. Raban and K. G. Holum, Leiden, pp. 121-129.

Jordan, D. R. 1985. "A Survey of Greek De.xiones," GRBS 26, pp. 151-197.

Messineo, G. 1984. "Ollae perforatae," Xenia 8, pp. 65-82.

Ostrasz, A. A. 1989. "The Hippodrome of Gerasa: A Report of Excavations and Research, 1982-1987," Jerash Archaeological Project, 1984-1988 2, Paris, pp. 51-77.

Patrich, J. 2001. "The carceres of the Herodian Hippodrome/Stadium at Caesarea," JRA 14, pp. 269-283.

Porath, Y. 1995. "Herod's 'Amphitheatre' at Caesarea: A Multi-Purpose Entertainment Building," in The Roman and Byzantine Near East: Some Recent Archaeological Research (JRA Suppl. 14), ed. J. H. Humphrey, Ann Arbor, pp. 15-27.

Rea, J., ed. 1968. The Oxyrhynchus Papyri 3, London.

Ribera, A. 1998. "The Discovery of a Monumental Circus at Valentia," JRA 11, pp. 318-337.

Richardson, R. B. 1897. "Excavations at Corinth," AJA 1, pp. 455-480.

Romano, D. G. 1993a. "Post 146 B.C. Corinth and the Planning of the Roman Colony of 44 B.C.," in The Corinthia in the Roman Period (JRA Suppl. 8), ed. T. E. Gregory, Ann Arbor, pp. 9-30.

--. 1993b. Athletics and Mathematics in Archaic Corinth: The Origins of the Greek "Stadion" (Memoirs of the American Philosophical Society 206), Philadelphia.

--. 2000. "A Tale of Two Cities: Roman Colonies at Corinth, Colonia Laus Iulia Corinthiensis, Colonia Iulia Flavia Augusta Corinthiensis," in Romanization and the City: Creations, Transformations, and Failures. Proceedings of a Conference Held at the American Academy in Rome to Celebrate the 50th Anniversary of the Excavations at Cosa, 14-16 May 1998 (JRA Suppl. 38), ed. E. Fentress, Portsmouth, R.I., pp. 83-104.

--. 2003. "City Planning, Centuriation, and Land Division in Roman Corinth: Colonia Laus Iulia Corinthiensis and Colonia Iulia Flavia Augusta Corinthiensis," in Corinth XX, pp. 279-301.

Romano, I. B. 1994. "A Hellenistic Deposit from Corinth: Evidence for Interim Period Activity (146-44 B.C.)," Hesperia 63, pp. 57-104.

Slane, K. W., and G. D. R. Sanders. 2005. "Corinth: Late Roman Horizons," Hesperia 74, pp. 243-297.

Spawforth, A. J. S. 1989. "Agonistic Festivals in Roman Greece," in The Greek Renaissance in the Roman Empire: Papers from the Tenth British Museum Classical Colloquium, ed. S. Walker and A. Cameron, London, pp. 193-197.

Walbank, M. E. H. 1997. "The Foundation and Planning of Early Roman Corinth," JRA 10, pp. 95-130.

Welch, K. 1999. "Negotiating Roman Spectacle Architecture in the Greek World: Athens and Corinth," in The Art of Ancient Spectacle (Studies in the History of Art 56; Center for Advanced Study in the Visual Arts, Symposium Papers 34), ed. B. Bergmann and C. Kondoleon, New Haven, pp. 124-145.

Williams, C. K., II. 1978. "Pre-Roman Cults in the Area of the Forum of Ancient Corinth" (diss. Univ. of Pennsylvania).

--. 1987. "Laus Julia Corinthiensis et Diana Nemorensis?" in [TEXT NOT REPRODUCIBLE IN ASCII] 2, Athens, pp. 384-389.

Wiseman, J. R. 1967a. "Excavations at Corinth: The Gymnasium Area, 1965," Hesperia 36, pp. 13-41.

--. 1967b. "Excavations at Corinth: The Gymnasium Area, 1966," Hesperia 36, pp. 402-428.

--. 1969. "Excavations in Corinth: The Gymnasium Area, 1967-1968," Hesperia 38, pp. 64-106.

--. 1972. "The Gymnasium Area at Corinth, 1969-1970," Hesperia 41, pp. 1-42.

--. 1979. "Corinth and Rome I: 228 B.C.-A.D. 267," ANRW II.7.1, pp. 438-548.

David Gilman Romano

The University of Pennsylvania Museum of

Archaeology and Anthropology



COPYRIGHT 2005 The American School of Classical Studies at Athens (ASCSA)
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2005 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Romano, David Gilman
Geographic Code:4EUGR
Date:Sep 22, 2005
Previous Article:Capital C from the Argive Heraion.
Next Article:Sella cacatoria: a study of the potty in archaic and classical Athens.

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters