(  )  Amphoras Project

Imports of Knidian Wine to Athens, Corinth, and Delos

Slightly modified from a talk given by PMW Matheson and CG Koehler at the Christmas meetings of the American Institute for Archaeology in December of 1990. The figures have not been updated. During the talk additional slides were shown illustrating the shape of Knidian amphoras; the sequence of shapes is illustrated here by Fig. 1, the photograph of stamped Knidian amphoras on display in the wine shop at the Stoa of Attalos.

Imports of Knidian Wine to Athens,
Corinth, and Delos

Our starting point for examining the imports of Knidian wine at Athens and Corinth is the obvious usefulness of transport amphoras as indicators of trade in wine. The AMPHORAS Project, which is computerizing Virginia Grace's records on ancient Greek wine jars, now has after five years enough examples in its databases to extract statistics that begin to show trends of imports and exports for several sites. The Corinth data are complete, with about 2000 stamped handles on record, though for material excavated in the Athenian Agora we are working here with 11,500 (about three-quarters) of its 15,000 catalogued stamps.

Knidian series

Fig. 1
Series of Knidian amphoras found in the excavations of the Athenian Agora, datable to Period III on the far left, then Period IV (time of the phrourarchy), Period V (after the destruction of Corinth), and finally Period VI (time of the duoviri). Taken from Amphoras 1979 fig. 64.
Note the rolled rim; the handles growing sharper and rising from their attachent at later dates, curved downward at the top earlier; the body narrowing as the height of the jars increases, and the characteristic ring round the toe.

We'll be looking at the exports of Knidos, which was of course a major producer of wine in the Hellenistic period, using the totals collected by Dr. Grace for different types of Knidian stamps. These individual readings are broken down into Periods,[1] based largely on major historical events which affect the archaeological record.
Knidian jars with their distinctive ringed toe begin in the 3rd century BCE The picture of the series of Knidian jars in Fig. 1 gives a good idea of their shape evolved. The practice of stamping began in Period III, which extended from about 220 or a little earlier, to 188 BCE, a date pegged by a dated deposit at Pergamon. The events of this period include upheavals affecting the balance of power between Antiochus III (the Seleucid) and Attalos I (the Eumenid) involving Rhodes and Macedon and the final intervention of Rome.
Period IV A ends in 167 with the Roman victory at Pydna over Perseus of Macedon, which resulted in much reorganization of Greek states, including the handing over to Athens of Delos and the exile of the Delians. Delos became a free port, and Rhodes lost most of her possessions on the mainland of Asia Minor. In 146 the Romans again interfered drastically with the amphora trade by sacking Corinth (thus incidentally providing us with another useful archaeological context date and the end of Period IV B).
1. The periods of Knidian export discussed here are:
Period III: ca 220–180 BCE,
Period IV: ca 180–146 (IV A = 188–167 when phrourarchs are named in the stamps (under Rhodian domination), IV B = the rest of Period IV),
Period V: 146–ca 108,
Period VI 108–ca 80 when the duoviri are named in the stamps (under Roman domination).

The jars becomes taller and thinner in shape throughout Period V (146–108) into Period VI B (98–88). In 88 the success of Mithridates in wiping out Roman traders throughout the Aegean, and the Roman reprisals against Greeks who supported him, produced a number of destruction contexts (for example, at Delos, and two years later the Sullan destruction at Athens) and brought the Knidian production of wine to a virtual stop for a few years thereafter. All this is reflected in the figures for Knidian stamped amphora production, with the addition of a further date, supplied by internal stamp evidence, of 108, when the practice of adding a new pair of officials to the amphora stamps suggested to Dr. Grace the introduction at that date of Roman tax collecting measures in the wine trade. [2] As the graph below shows, this was also a time of phenomenal increase in the production of Knidian wine.
2. For a full discussion of Virginia Grace's chronology of the Knidian series of amphoras, see EAD 27 1970 p318–322.
Knidian amphoras continue to be made in some quantities after the end of Dr. Grace's period VI C but the stamps gradually cease to be as informative and we aren't going to include them here.
The general idea is to illustrate the importance of looking at the whole of the export of a commodity, rather than at just the pattern of import of that commodity at a given site. As you will see, a very misleading impression of the relative quantities of wine imported by a city like Corinth is obtained from looking only at the Knidian jars found there.

SS imports

Fig. 2
Exports from KNIDOS

This graph represents the stamped amphora handles from Knidos found at sites all over the Mediterranean, in the greatest quantities at Alexandria, Athens, and Delos, but also at sites yielding smaller numbers from Italy to Asia Minor, and from Macedonia and the Black Sea to Africa. The total number of Knidian stamps known to us from this time span is of the order of 40,000, but we have limited our statistical sample to a mere 29,000 of these, omitting those which could not be allotted to one of the Periods shown.
The number of stamps which can be assigned to each Period has been divided by the number of years in the Period to give a unit of "stamps per year", or SAH (stamped amphora handles) / yr, as an index for comparison. Knidian jars were regularly stamped on both handles, so the number of jars this represents would be about half. On the other hand, we know that Knidian wine was also occasionally shipped in unstamped containers (a few have been found in the Agora from these Periods) and we have no way of knowing what percentage of all the jars made have turned up so far in excavations, so all these figures are very much lower than the actual quantities involved. The stamps, however, allow us to consider the relative proportions in which the wine was exported and imported by various states.

SS imports

Fig. 3
Imports into ATHENS: Agora

Here you see a similar graph showing the numbers of Knidian stamped amphora handles found in the Athenian Agora and datable by the type of the stamp to Periods in Dr. Grace's chronology. It will be apparent that these two graphs show similar ups and downs. This should not surprise us, for two reasons:
1. The finds from the Agora represent a large proportion, over one fifth, of the datable Knidian stamps from the period, and

2. the ancient economy differed from ours in that, as Tarn puts it, demand normally outran supply, and if you could get a thing you could certainly sell it. [3] Personal taste exercised by consumers in accordance with a fashion in wines was likely to be a minor factor in the import pattern of any state, at least in comparison with economic and political factors. As Athens would be likely to want to import in proportion as Knidos was able to export, the variations between the two graphs should provide us with Athens' ability to import at any given time.
3. Tarn 1936 p218. Cf. Glotz 1926 p363
In Periods III and IV A, the scale of imports of Knidian wine at Athens exactly mirrors that of the total export of the wine. In IV B, the period between Pydna and the sack of Corinth, export of wine at Knidos is slightly down, but Athenian imports are slightly further down: she is no longer importing quite the same relative quantity of Knidian wine, and the trend continues more markedly in the second half of the 2nd century. Period VI, however, shows a resurgence in Athens' imports and by 88, when the Mithridatic Wars bring wine export and import alike to a temporary halt, Athens is again importing the same proportion of the wine shipped from Knidos as she had been in the first third of the 2nd century.
Here we must enter a caveat on the figures shown for Athens. Some uncatalogued stamps, and the final third of the catalogued ones, have been counted in the bulk figures for export from Knidos but are not represented in the graph for Athens. Thus the import figures for well attested periods may be slightly low, and not too much significance can be attached to the slight apparent diminution in the Knidian wine trade with Athens in the second half of the 2nd century, though it does actually fit nicely with the general economic depression of that period at Athens, known from other sources.
Both these graphs draw on a substantial quantity of finds, so the next question is, Can we get an accurate impression of the imports of a site which yields hundreds, rather than thousands, of samples?

Ker imports

Fig. 4
Imports into ATHENS: Kerameikos
Here is a graph of the datable Knidian handles from the finds from the German excavations at the Kerameikos. Even though our computer database is not yet complete for this site, and the graph presented is based on only 684 samples, the similarity of the graphs is striking. The rate of import in Periods III to V follows exactly the same pattern as at the Athenian Agora. Although the difference here between Periods VI A and VI B is greater than that at the Agora, it does find a parallel from another Athens group of Knidian jars of the early first century found in the American and Greek excavations of the the north and south slopes of the Acropolis and the Pnyx (which contains a mere 140 samples). The smaller size of these samples does not seem to have prevented the typical Athens pattern of Knidian imports from emerging.

Corinth imports

Fig. 5
Imports into CORINTH
It is interesting now to to turn to the graph of Corinthian imports of datable Knidian stamps for the same period. The sample is a good deal smaller again than the one just seen for the Kerameikos finds (though larger than our sample from the slopes of the Acropolis and the Pnyx), but the pattern clearly reflects the political and economic differences between Athens and Corinth. The graph suggests a fairly slow start to importing Knidian wine and Corinth's Knidian imports peak in Period IV A. The falling off in IV B is again similar to Athens, and then in 146 she was sacked by the Romans. Her imports for Period V are very low, there is a brief and comparatively small revival in Period VI A, and then in VI B her imports drop off at the time when other sites reach their maximum rate of import of Knidian wine, a time which is also the peak period of export for Knidos.

When we first started to work on the Corinth stamps, we were struck by the fact that such a large proportion of them can be dated to the time after the destruction in 146 B.C.[4] It appeared that Corinth was importing Knidian wine in the second half of the 2nd century at a rate almost comparable to the period before 188, and that her rate of import at the very end of the century was nearly equal to her previous "high" in Period IV A. And certainly the graph for Periods V and VI does show an echo of the pattern common at other sites, which suggests that some activity continued at the site even during the first half of the interim period before the refounding of Corinth as a Roman colony in 44 B.C.[5] What was not apparent without the comparative data from Knidian exports and imports at other sites was just how faint that echo is: compared with the enormous rise in exports from Knidos in Period VI and the corresponding expansion in imports at Athens, the Corinthian rate of increase is exceedingly slight, and the graph for Periods V and VI does indeed reflect the severity of the blow dealt by Mummius in 146.
4. Williams 1978 p21, n29.
5. A Roman colony was founded on the site of Corinth in 44 BC by Gaius Julius Caesar. To quote Strabo: Now after Corinth had remained deserted for a long time, it was restored again, because of its favorable position, by the deified Caesar, who colonized it with people that belonged for the most part to the freedmen class.” Strabo, Geography, 8.6.23. c.23 AD.
It is perhaps pushing such a small sample too far to note as significant the relative rates of import in Periods VI A and VI B. But if you will look for a moment at the graph for the datable Knidian stamps from the French excavation at Delos for the same period you will see what almost amounts to a reversed image.

Delos imports

Fig. 6
Imports into DELOS
Where Corinth is thriving in the late 3rd to early 2nd century, Delos is inactive, at least as far as the import of Knidian wine goes. Delos still shows comparatively little activity in Period IV, though her imports are increasing, and they expand again in Period V when Athens, and even Knidos, are decreasing, and Corinth has "bottomed out." From this time on, Delos' rate of import rivals Athens'

Note also that Delos' imports in Period VI A are distinctly lower than in VI B, where Knidos' exports are very close to the same in these Periods, and Corinth shows greater imports for VI A than for B. Modern historians have noted that times of trading prosperity at Corinth were likely to be times of adversity for Delos, and vice versa (though we won't go so far as to agree with Mommsen[6] that the Romans sacked Corinth primarily in order to favor Delian trade). But here one might also see the possible effect of some more local phenomenon, a temporary increase in the habitation of the site of Corinth by people whose salient characteristic, for us at any rate, was that they drank Knidian wine and left behind them their empties bearing the names of pairs of the special dating officials, Εὐκράτης and Κλευπόλις, Κράτερος and Νικασιβοῦλος, which give that little extra height to the Corinth graph in Period VI A.
6. The apparently needless cruelty of Mummius in Corinth, by no means characteristic of him, is explained by Mommsen as due to the instructions of the senate, prompted by the mercantile party, which was eager to get rid of a dangerous commercial rival. Encyclopaedia Britannica 11th ed. (1911) vol 18 p966–967.
It is entirely possible that we have over-stretched the historical conclusions to be drawn from our comparison of the very various amounts of data available from different sites, and entirely certain that our graphs will continually shift slightly in outline as more data already collected by Dr. Grace are processed and new data from excavations are added. But we offer these preliminary figures as an indication of Knidian exports and of the significance of the relative imports by Corinth and Athens.



Amphoras 1979
Amphoras and the Ancient Wine Trade 2nd ed. American School of Classical Studies at Athens, prepared by Virginia R. Grace (Princeton 1979)

Glotz 1926
Glotz, G. Ancient Greece at Work (London 1926)

EAD 27 1970
Grace, Virginia R.; Savvatianou-Petropoulakou, Maria "Les timbres amphoriques grecques" (in l'Îlot de la Maison des Comédiens), Explorations Archéologique de Délos XXVII (Paris 1970)

Tarn 1936
Tarn, W.W. Hellenistic Civilization (London 1936)

Williams 1978
Williams, II, C.K. "Corinth 1977, Forum Southwest," Hesperia 47 (1978)