Observations on the Amphora Repertoire of Middle Punic Carthage

This publication sets out to define the amphora repertoire of Middle Punic Carthage (dated here 530–300 B.C.E.) as the first step to a description of trade relations between Carthage and its neighbors in this period. The complicated subject is potentially of great interest to historians as well as to archaeologists and might be especially welcome because little that is relevant to this topic has appeared in English. Bechtold uses amphora data from the excavation at Bir Messaouda Site 2, where four relevant subperiods were defined by director Roald Docter. These are Early Punic (EP)/Middle Punic (MP) (530–480 B.C.E.), MP I (480–430 B.C.E.), MP II.1 (430–400 B.C.E.), and MP II.2 (400–300 B.C.E.) (2). Her results suggest that while EP Carthage depended largely on Phoenicia and Iberia, a more cosmopolitan MP Carthage already traded almost exclusively with the non-Punic central Mediterranean (22–3).

Unfortunately, the relevant new data is exiguous. Bechtold derives percentages of local vs. imported amphoras from 31 small archaeological contexts (section 4). These show that imported amphoras rose from 10–15% at the beginning of this period to 20–30% at the end, much lower percentages of imports than in the preceding EP period (75); 165 wall sherds of amphoras from these stratified MP contexts allow her to identify and date the occurrence of various fabrics (tables 1–3). Docter hypothesized that the contexts were unnaturally small because Carthage collected garbage intensively from the very beginning of the MP period (25 n. 80). Since there were no stratified rims, handles, or bases (diagnostic sherds) in these contexts, identifying amphora types from stratified evidence was impossible. Bechtold publishes 71 possibly relevant diagnostic amphora sherds from later contexts of the site in the catalogue (section 8). The catalogue is organized in terms of 37 fabrics that can be linked with these types, and for these fabrics, she also publishes cross-sections in color (at 8x magnitude) (pls. 1–5).

This relatively small number of unstratified amphora fragments reveals a range of imported types and suggests highly differentiated trade in the MP period, but these types had to be dated by external criteria and could be related to the dated stratified contexts only by fabric. However, it is helpful that earlier Punic amphora types, which had been studied by Docter (Archaische Amphoren aus Karthago und Toscanos: Fundspektrum und Formentwicklung. Ein Beitrag zur phönizischen Wirtschaftsgeschichte [Amsterdam 1997]), are eliminated from Bechtold’s MP repertoire. Also eliminated are well-known North African cylindrical types and Rhodian amphoras from the last half-century of the Late Punic (LP) period, but the catalogue still includes amphora types belonging to the third century B.C.E. and later (the LP, rather than the MP, period).

Bechtold begins by identifying all MP and LP amphoras published and illustrated from excavations at Carthage since the beginning of the UNESCO project in 1972 (section 1). Most of this material has not previously been analyzed in any depth, and Bechtold does not mend this omission. Here, she explicitly omits typologically EP amphoras, local LP amphoras, and Rhodian amphoras (4 n. 11). The remaining MP and LP amphora types are tabulated (table 6). The author’s own evidence shows that many of the Graeco-Italic and Punic types she discusses in this section are not relevant to MP Carthage. Bechtold’s summing up of this material (section 2) rightly lists many caveats.

In section 3, Bechtold presents identifications for 87 numbered fabric classifications organized by 16 Mediterranean-area provenances. In section 8, she publishes fabric descriptions for 37 of these. The number of fabric samples assigned to each of the 87 fabrics ranges widely, and it is not clear that these numbers are representative. The data suggest that Sardinia was an important source of Punic-style amphoras, that Carthage was supplied by Graeco-Italic amphoras from a relatively large number of production centers ranging from Magna Graecia to Corcyra, and that the range, if not the number, of imports from northern Aegean production sites was broad.

Tables 1–3 correlate fabric with context date; when the context date is derived from stratigraphy, the designation is in bold. Corinthian A occurs only in EP/MP and MP I (530–430 B.C.E.). Sicilian/South Italian wine in Graeco-Italic amphoras occurs only in MP II (430–300 B.C.E.). Bechtold’s evidence (four sherds [table 1B]) implies that local North African Punic amphora fabrics do not appear before MP II (430–300 B.C.E.). Ramón Torres earlier pointed out that there is little evidence for amphora production at Carthage between 575 and 400 B.C.E. and certainly between 520 and 450 B.C.E. (Las ánforas fenicio-púnicas del Mediterráneo central y occidental [Barcelona 1995] 283), but I did not find this noted in Bechtold’s text. Pie charts demonstrate that the chronological distribution of the stratified MP amphora fabrics bears no relation to the expected chronological distribution of the unstratified amphora types in her study group (figs. 11, 12), but this only serves to confirm that the two populations are not comparable.

Section 6 is an overview of amphora studies in Sardinia, Sicily, and environs (51–74). Bechtold’s evidence suggests that Punic Sicily was quite independent of Carthage (76). Section 7 presents Bechtold’s conclusions (75–7), most of which I have summarized briefly above. Here, she adds that most imported amphoras in the MP period contained wine; amphoras from the northern Aegean are particularly characteristic of MP II.1 (430–400 B.C.E.) and may be related to the contemporary increase in Attic black glaze (77).

I should summarize and comment on Bechtold’s views on the important Corinthian B amphora. Her survey of previously published amphoras at Carthage revealed only four examples from the German excavations (table 6), but Wolff (“Carthage and the Mediterranean: Imported Amphoras from the Punic Commercial Harbor,” CahÉtAnc 19 [1986] 141) noted 230 sherds with a fine soft beige fabric from the Punic commercial harbor, mostly from the fourth and early third centuries B.C.E. (12, 123 n. 504).

Bechtold identifies Corinthian B with Gassner’s rim form 5 from the mid fifth to fourth centuries B.C.E. (3 [fig. 1.7], 100). Gassner’s rim forms 1–7 all have a range of production areas (23). Bechtold’s table 2.A.4 indicates that her Fabrics 7 and 15 are produced in Corcyra/Corfu (30). Cross-sections of sherds of these rather similar buff fabrics are illustrated on plate 3. The two Corfiote fabrics total 29 of 165 (17.6%) of her stratified MP sherds, a disproportionately large number, which suggests that Corinthian B is by far the most important MP import at Carthage. Fabric 7 is stratified in MP I (480–430 B.C.E.) and Fabric 15 in MP I and II (480–300 B.C.E.) (30). Bechtold catalogues only one rim of Corinthian B/Gassner rim form 5 (cat. no. 39, fig. 21) in Fabric 15, and it is from a Roman context (100). Diagnostic sherds in these fabrics include Gassner’s rim forms 1 and 3a (the earliest Corinthian B production had the same form as Gassner’s rim form 1, the “top-shaped” western Greek amphora) (Gassner, Materielle Kultur und kulturelle Identität in Elea in spärtarchaisch-frühklassischer Zeit: Untersuchungen zur Gefäss- und Baukeramik aus der Unterstadt (Grabungen 1987–1994) [Vienna 2003] 183–84).

Bechtold implies that amphora imports from Corcyra increase from 480–430 B.C.E. to 430–300 B.C.E. (35), although table 2.A.4 (30) hardly supports this statement. She states that Corinthian B amphoras from Corcyra, equally with amphoras from around Rhegion, are the most common import at Carthage in 430–300 B.C.E. (35). She suggests that the most important trade route for Carthage at this time is therefore from Athens/Corinth by way of Corcyra, along the Ionian coast of Calabria and past Sicily to Carthage (48).

In the 1990s, I catalogued 10 stamped handles of Corinthian B amphoras from the National Museum of Carthage, with permission from director Abdelmajid Ennabli. Six have fabrics roughly comparable to Bechtold’s Corfiote Fabrics 7 and 15, while four appear in a quite different reddish-orange fabric. Bechtold’s study (supported by Wolff’s observations) indicates that the Corinthian B amphoras at Carthage are uniformly produced around Corcyra; the examples in the museum suggest, however, that this is an oversimplification.

Flaws in this work include lack of rigor in methodology and lack of clarity in the presentation of the results. The work is not reasonably self-contained; for example, designations such as “Ramón T-4.1.1.2” are used as if they are intrinsically informative. The bibliography is heavily weighted to material published since 2000. Bechtold’s study cannot be ignored by amphora experts who are interested in MP (and LP) Carthage, but it will not be loved.

Joann Freed
Department of Archaeology and Classical Studies
Wilfrid Laurier University
Waterloo N2L 3C5
Canada
jfreed@wlu.ca