Back to Home


Via Gabina Site 11 Italo-Megarian Bowls

Ten pieces of Via Gabina Site 11 pottery belong to a well-known Italian Hellenistic ware, the forms of which generally consist of thin-walled hemispherical moulded bowls, more or less correctly referred to as 'Italo-Megarian bowls'. These bowls imitate the form and decorative motifs of a Greek type originally produced in Athens and Corinth by the middle of the third c. B.C. They were subsequently imitated at Pergamum and at Antoch; at Pergumum they are first common in contexts dated c. 200-191 B.C., while at Tarsus they appear commonly only in the second quarter of the second c. B.C. (Moevs, 1980, p. 184).

The Italian hemispherical thin-walled moulded bowls are distinctive in fabric. The fine and hard, gritty and slightly micaceous redbrown fabric is consistently unslipped, the surfaces often darker and greyish. While the Italian ware recycles motifs from the Greek and Asian workshops, the range of motifs and their deployment is characteristic enough to be easily recognizable.

The Italian products are often referred to as 'Popilius' bowls, since a number of examples signed by C. Popilius are extant (10 examples of signed bowls and of a mould are illustrated by Moevs, 1980, pls. 17-18). Workshops of Popilius are known at Bevagna and Otricoli, both sites situated in Umbria, north of Rome on the Via Flaminia. Their geographical situation certainly tends to confirm that the establishment of these workshops was not earlier than the establishment of the highway by C. Flaminius, c. 220 B.C. (Verzar, pp. 121-2).

Besides C. Popilius, potters P. Lapius, L. Atinius and L. Quintius have signed bowls of this Italian type, but the sites of their workshops are not known. Moevs also discussed the potter Heracleides (Moevs, 1980, p. 194, n. 155), whom she associates with a workshop at Tivoli which has been identified by a dump of wasters and moulds (Moevs, 1980, p. 179; examples illustrated on pls. 22-23). The finds from Cosa also included two fragments of moulds (pl. 8, nos. 10-11, pp. 196-7), neither of high esthetic quality.

Moevs does not seem to discuss the implications of these moulds at Cosa for the production of Italo-Megarian ware. Whatever the provenance of the Italian examples, the fabric and decorative motifs are very consistent, suggesting that all these workshops were closely associated.

Moevs has written a detailed report on the sixty-eight sherds of Italo-Megarian ware from Cosa, discussing fabric, the internal and external evidence for dating, and the sequence suggested by stylistic criteria. These sixty-eight pieces are closely related in fabric and style to the Italo-Megarian bowls which appear at the Via Gabina.

The ware is particularly important at the Via Gabina, though the number of pieces is so small, because an example of this type in the construction for Period IC may be the latest closely datable ceramic evidence from that context. The stylistic arguments of Moevs are therefore of great interest, as they directly affect the dating of the earliest integrated architectural phase from the site.

The dating of the type in general has been very erratic, as Moevs demonstrates. It is absolutely clear, despite Moevs' hesitancy to embrace an early date for the Italian production, that the type does appear in Italy at about the same date that it does in Pergamum; that is, before the end of the third c. B.C. (Moevs, 1980, tomb at Castel d'Asso, pp. 177-8; Viterbo, tombs near Corchiano and at Musarna, p. 178). At Cosa, fragments nos. 4 and 6 are in a stratified context which would allow them to be dated as early as the end of the third c. B.C. (Moevs, 1980, p. 182), but on stylistic grounds, Moevs avoids placing them that early.

Moevs suggests that known signed Popilius bowls demonstrate a 'drastic change from a controlled to a dramatic naturalism'. Furthermore, she dates this change to the second quarter of the second c. B.C. on the basis of the Cosa finds (Moevs, 1980, p. 191).

It seems to me that Moevs' arguments from stylistic development are not strong enough to establish three separate periods of development (i.e., before 150, from 150-75, and post 75 B.C.). On the other hand, the evidence she presents allows me to be very comfortable with a date for this production between 200 and 100 B.C. Given the concentration of our own pieces in a specific area of the site, associated with the construction and occupation of Period IC, it seems reasonable to suggest that our pieces date c. 200-190, which is then the date for construction, allowing that the remaining pieces are in occupation of that period.

Verzar illustrates an example of the type from Arezzo (fig. 1, p. 132, INR.63.583). While the moulded motifs include running spiral and fret borders and vertical acanthus leaves and spikes, all of which also occur at Cosa, the vertical rim which rises directly to an everted lip is quite different from the Cosa and Via Gabina examples, which have a hollow 'neck' above the topmost moulded border before the lip of the rim turns outward. The form copies metallic prototypes (Moevs, 1980, p. 185). Moevs suggests that the typical hollow of the Cosa rim is earlier rather than later, in contrast to the sequence of development among producers of the 'Megarian' bowl in Greece and Asia Minor (p. 193).

The Via Gabina pieces are all in identical fabric. Nine of the ten come from a very restricted area of the site. In terms of decoration, they are so similar that there is some question about how many examples are actually represented. The pieces include three rims, five wall sherds, and two rosette base medallions. The variety among the rims demonstrates that a minimum of three examples of the type are represented.

The pieces which Moevs illustrates from Tivoli show motifs which appear on the Via Gabina sherds: borders of running spirals, tendril scroll, running leaf and rosettes, running oak leaf and acorns, fret, fret with amphoras; median designs of jewel-ribbed nymphaea caerulea alternating with curly spikes or acanthus leaves, imbricate petals, nymphaea nelumbo alternating with acanthus, curly spikes. Generally, the style of the Tivoli fragments illustrated is caught between what she characterizes as 'controlled' versus 'dramatic' naturalism. They vary from very spontaneous to very rigid and lifeless compositions, yet there is no evidence to suggest that they are not all nearly contemporary.

In the following descriptions, I have followed Moevs' terminology wherever possible:

Three rims; Moevs notes that variations in rim diameter are typical of Megarian bowls (1980, p. 186 and 195):

d291 p15            VG11/76            H19 R E 5

Rim and wall, rd 14.0

Italo-Megarian bowl.

Rim is everted above long hollow throat; the curving throat joins the body to form an offset shoulder below which the moulded decoration begins. The bowl has a double border preserved: the upper zone is made up of running spirals; the lower zone, though difficult to read, is evidently a flowered scroll. Each of these zones is slightly depressed and is underlined by a narrow groove.

Fairly hard and fine gritty medium brown fabric with a slightly lighter core (color varies between 2.5YR 6/6 light red and 5/4 reddish brown). This, the most complete example of the type from Via Gabina Site 11, was described as 'badly moulded' by the registrar.

The rim and throat of this bowl is very comparable to no. 6 at Cosa; this is important because of the likely early dating of the rim type.

The upper border of running spirals is one of the most common border motifs among the finds at Cosa, occuring on nos. 4-6, 8-9, 24-28, 41, 54, 60 and 67. The motif is traditional on Greek Megarian bowls (Moevs, 1980, p. 192). Nos. 4 and 6 are the only pieces at Cosa for which a date at the very beginning of the third c. B.C. might be argued on stratigraphic grounds, although Moevs prefers to date them in the second quarter of the second c. B.C. for stylistic reasons (Moevs, 1980, p. 183).

Whereas no. 6 has a fret for its second border, d291 has a rather poorly drawn flowered scroll with arched vine and tendrils, evidently similar to nos. 25-28 at Cosa. Moevs links the flowered scroll to Heracleides (whom she furthermore links to the workshop at Tivoli), and says that this border was not in the repertory of Popilius and Lapius. She therefore places this border in the last quarter of the second c. B.C.

d233            VG11/77            I19 A 2

Rim, rd 12.0

Everted rim above a slightly shorter hollow throat than d291; no lower moulded decoration is preserved.

Hard and fine medium red-brown fabric (2.5YR 6/6 light red) with a slightly darker surface.

d36 p38            VG11/77            H18 3

Rim chip, rd 10.0 (?)

Narrow hollow corniced slightly everted rim; below the rim a single pointed leaf of moulded vegetal decoration survives; cp. no. 9, 'a straight ivy garland divided into segments with leaves turned in opposite directions, of which two ends and terminal round-petalled rosettes are preserved' (Moevs, 1980, p. 218). Moevs believes that this ivy garland has a Pergamene origin; a closely related garland appears on the early issues of Pergamene cistophoroi coins, which first appeared around 190 B.C. (Moevs, 1980, p. 192). Because of this association, she allows that this garland can appear on Italo-Megarian bowls dated to the second quarter of the second c. B.C.

Slightly micaceous hard medium orange-brown fabric (close to 2.5YR 6/6 light red) with minute black inclusions.

The following three small wall sherds all bear a fret or Greek key pattern:

d542            VG11/79            F14 unstratified Wall sherd, wall thickness 0.3

Remains of three leaves of a border, probably a straight ivy garland of the type discussed immediately above, placed over (?) a fret with an amphora/vase: nos. 21 and 22 at Cosa show that the amphoras alternate with a square of four 'checkers', no. 5 shows that the fret can be placed upside down; stars (nos. 6 and 50) and other motifs also are attested as combined with this type of fret.

Hard, slightly micaceous red-brown fabric (slightly redder than 2.5YR 5/6 red) with some minute white calcareous inclusions.

d309 p183            VG11/76            H19 R A 4

Wall sherd, angled and curved up for probable long corniced rim. Poorly moulded Greek key pattern runs along over raised ovals. A line of closely spaced square beads runs along below the Greek key pattern.

Hard fine brown fabric (between 2.5YR 6/4 light reddish brown and 6/6 light red), turning black (7.5YR 3/2 dark brown) towards exterior surface. Minute white and mica inclusions.

d156 p138            VG11/77            H19 N CBS (sic) 'on top of block 1'

Wall sherd.

Greek key pattern runs along over the odd blob or two; a double groove runs either above or below.

Slightly micaceous orange brown fabric (2.5YR 6/6 light red).

The following two wall sherds show well-known motifs of this type:

d298 p42            VG11/76            G20 S B 3

Wall sherd, thickness 0.2 mm.

Moulded ribbed vertical long leaf with trace of vertical corded line parallel.

Fine hard medium brown fabric with minute white and micaceous inclusions, darkened exterior surface.

d657            VG11/79            H16 C 3

Wall sherd.

Moulded imbricated leaf decoration from calyx near base of bowl. Medium hard pink to light brown fabric (between 2.5YR 6/6 and 6/8 light red) with light grey core (2.5YR 6/2 pale red), minute mica. Moevs makes a good deal of the outlining of imbricated leaves, I think mistakenly. She describes them as 'rigid' in association with the leaf which barely shows up on no. 25, pl. 2, where it must be the last leaf of the rosette calyx, otherwise entirely lost (Moevs, 1980, p. 194). She associates this with the eclecticism of Heracleides, with the workshop at Tivoli, and with the last quarter of the second c. B.C. at the earliest (Moevs, 1980, p. 203)

Note that this outlining appears on pl. 20, no. 2, Rome, Villa Giulia Museum; what does she think of it?

The following two sherds incorporate basal medallions:

d297 p149            VG11/76            F20 1

Base sherd.

Moulded central rosette surrounded by pattern of alternating nymphaea nelumbo and acanthus leaves.

Hard orange fabric, dull brown surface (varying from 2.5YR 6/8 light red to 5YR 5/4 reddish brown), with minute white quartz and brown inclusions.

There is no doubt that this bowl is an example of 'dramatic naturalism'.

The rosette seems to be more or less identical to the rosette on mould no. 10, pl. 8 at Cosa, where the intention of outlining each petal is also expressed but not consistently carried out.

The nymphaea nelumbo motif here appears only on no. 31 at Cosa; it is relatively common elsewhere, appearing as an inhabited space on the glazed bowl signed by Popilius in the Vatican (pl. 18), another Vatican bowl (inv. 15462, pl. 20, no. 6 and pl. 21, no. 1) and in pl. 22, nos. 3, 5 and 6 from the Tivoli dump of moulds and wasters. Theories for sequence of the motif suggest that it appears early, on a bowl in the style of Quintius (Boston Museum inv. 03.858). The suggestion that the motif evolves from naturalism to styization, ending in the nelumbo framing figurative scenes, does not seem to me to have any firm basis (Moevs, quoting Byvanck-Quarles, 1953, p. 7) Moevs dates no. 3 from Tivoli with her no. 31, to the first quarter of the first c. B.C., while she dates no. 6 with the glazed Popilius bowl.

The acanthus leaves here are most similar to the long swirly flanking leaves on no. 7 at Cosa. Moevs quotes a parallel from Ostia (Arena, 113-4, no. 10, figs. 19-21).

The style of the leaves and of the calyx as a whole have been characterized as a turgid and dramatic style to be associated with Pergamum. (Moevs, 1980, following Byvanck-Quarles, 1953, who dates this trend to the beginning of the second half of the first c. B.C. The style was part of the repertoire of Popilius himself (Moevs, 1980, pl. 17, no. 9, Tarquinia, inv. R.C. 5592), and therefore its introduction in Italy cannot be 'late'; Moevs suggests, again, the second quarter of the second c. B.C.

d706            VG11/79            I19 E' 3

Base sherd.

Moulded rosette in disk, alternating pattern of rising double corded line and moulded oval ?

Hard and fine light orange-pink fabric (2.5YR 6/6 to 6/8 light red) with a browner (5YR 6/4 light reddish brown) exterior surface.

Bibliography for further reference:

Arena, M. S., "Su alcuni frammenti di ceramica italo-megarese conservati nell'Antiquarium di Ostia", RSL 35 (1969), pp. 101-121. Edwards, G. R., "Hellenistic Pottery", Small Objects from the Pnyx: II, Hesperia Suppl. 10 (Princeton, 1956), 79-112.

Jones, F. F., "Bowls by Popilius and Lapius", Record of the Art Museum, Princeton University 17, 1 (1958), pp. 21-40.

Marabini Moevs, Maria Teresa, "Italo-Megarian Ware at Cosa", Memoirs of the American Academy in Rome 34 (1980), pp. 163-227 and plates.

Ohlenroth, L., "Zu den mittelitalischen Reliefbechern (Popilius-Ware)", Germania 33 (1955), pp. 43-45.

R. Paribeni, "Scarichi di una fabbrica di vasi fittili in localita

S. Anna (Tivoli)", NSA (1927), pp. 374-8.

Schafer, J., Hellenistische Keramik aus Pergamon, Deutsches Archaologisches Institut Pergamenische Forschungen 2, Berlin, 1968.

Thompson, H. A., "Two Centuries of Hellenistic Pottery", Hesperia 3 (1934), pp. 311-480.

Verzar, Monika, "Archaologische Zeugnisse aus Umbrien", in Hellenismus in Mittelitalien, pt. I, ed. Paul Zanker, Kolloquium in Gottingen vom 5. bis 9. Juni 1964, Gottingen, 1976, pp. 116-142.

As of January 5, 1992, I am very much bemused by all this. The problems are as follows:

1) Motives are not studied separately in any organized way

2) Past dating has been based largely on wishful thinking, particularly on stylistic structures which may not hold water.

3) I don't believe the general type can last for more than 100 years: shouldn't all known pieces just be taken back to the earliest possible date unless there is good reason to make them later?

4) Moevs is determined to fit each piece into a dating framework which is hardly going to stand up under the strain--does she illustrate the pieces which are actually found in dated contexts?

5) There is no reason why a number of decorative schemes cannot be contemporary. It seems obvious that the exuberant naturalism school was dominant at Cosa in the second c. B.C.: nothing else is particularly demonstrable.

6) Via Gabina motives and Cosa motives are very close.

7) The Tivoli workshop is close to the VG: is this relevant?

8) The influence of other workshops is important for both date and cultural cross influence.

9) I should read over the stratigraphic evidence from Cosa again.


All text and images copyright © 2002 by Walter Widrig and Rice University. Last updated June 2005 by