Bryn Mawr Classical Review 95.03.33

Gary Reger, Regionalism and Change in the Economy of Independent Delos, 314-167 B.C. Berkeley: University of California Press, 1994. Pp. 396, xvii; 36 tables. ISBN 0-520-08460-8. $55.00.

Reviewed by Richard Hamilton (

R.'s book is an extended, meticulous and intelligent argument against the common idea that the Delian economy was affected by a "universal price-setting market"; in fact he doubts even the concept of an "Aegean market" and seeks "local explanations for local phenomena" using a genial combination of reason, careful extrapolation and sophisticated analysis of the mounds of data supplied by the priests' monthly ledgers during the island's independence (314-167 BC).

R. begins with a look at those who benefited economically from their relations with Delos and finds that almost without exception they come from nearby: loans are to people from Delos or the Kyklades; craftsmen, as far as we can tell, were mostly Kykladic. Proxeny decrees, on the other hand, involve a widespread net of Greek cities, but these reflect parochial Delian political worries rather than long-distance economic connections. Of the sixteen decrees honoring citizens of Hellespontine or Black Sea cities five are clearly non-commercial. Only one is commercial, and here the honorand's city of origin "had nothing to do with the source of his goods". The same is true of the Rhodian and Chian decrees: the ties are religious and political not economic.

R. then shows how easily Delos, given its size, population and location, could have satisfied its need for grain (which supplied 70% of the ancient diet) from nearby islands. "Roughly half of the regular annual aggregate Delian demand for grain could have come out of the surpluses of Apollo's immediate neighborhood" since Delos itself (pop. 2600-9100) could supply 230-350 people with grain, Rhenea about 725, Mykonos (pop.1625) somewhere from 1960 to 5660. [This conclusion is hardly, as claimed, "inescapable" since the worst-case would put the total for Delos at 230+500+335, which could be as little as 12%, but, since the best case (350+700+4035) is 196%, it seems reasonable enough.] R. then goes on to argue that "the Kyklades must normally have been more or less self-sufficient in grain" given a population of about 37,000 and grain for 50-100,000 if we assume cultivation of 20-40% of the land. Admittedly, modern data from the 1930's suggests cultivation of only 6-8% (i.e., enough grain for 15-20,000), but R argues (1) recent study of Melian countryside has found it 17-58% arable (14% cropped in 1971); (2) terracing in Delos (probably "in place by the fifth or fourth century B.C.") would have made it 70% arable; (3) in 329 B.C. Skyros (total area 20,900 hectare) produced 9600 medimnoi of wheat and 28,800 of barley; and Lemnos (total area 47,600 ha) produced 56,750 medimnoi of wheat and 248,525 of barley, suggesting cultivation of 12-44% of the land by my calculations (R. gets 20-50% but does not explain how, nor does he footnote the source for this information until five pages later). When grain was scarce, the Kykladic islanders could go a bit farther afield: "Lemnos, Lesbos, Samos, Khios, Kos and Rhodos, lying but a short sail east of Delos, were remarkably productive". Why then has Delos been considered a great center of grain trade? R. argues that Delos' central location and relatively high need for imported grain made it a local (but not international) distribution center; when international politics of late 3rd C. became destabilized, Delos developed as a transshipment point around the Kyklades especially for soldiers' grain, for instance, in the 230's Demetrios II sent a grain buyer to Delos, in the 190's the Nesiotic League sent grain purchasers, and about that time the Histaians sent sitonai to borrow money and buy grain, presumably because of a local shortage. Eventually a permanent local sitonia fund was created. On the other hand, Delian prices seem not to fluctuate with Egyptian prices or even in response to a shortage along the Euripos.

R. then turns to the data in the priests' accounts, namely the prices of olive oil, firewood and pigs, which he studies both short-term and long-term. Oil, wood and pigs may seem odd bedfellows, but R had no choice: "the majority of the thousands of prices recorded in the inscriptions are useless for economic analysis". Olive oil is said to show [though I cannot see it] spring and fall adjustments, which can be explained by the sailing season and the seasonal cycles of the olive. More persuasive is the explanation for the general lack of variation (42 of 63 prices, 67%, do not vary): calculations suggest it would take only 2-4 shiploads to satisfy annual Delian demand and once these had arrived prices would tend to be set across the whole year. Firewood is said to show spring declines and fall increases linked to seasonal change of temperature (though again I doubt the pattern; year-end activity is more evident). [Here the lack of variation (31 of 64, 48%) is less pronounced, though from 224 it becomes noticeably more so (6 of 24 25% before; 25 of 40 63% after)]. "Monthly pig prices show less patterned behavior" and considerable variation (only 18/83, 22%, do not vary). Since pigs are not a staple like wood and oil, demand was extremely sensitive to price, resulting in a "sawtoothed price curve with a periodicity of about six months" because of the pig's two-litter/year cycle. Notably, prices tended to rise at the Thesmophoria: "an increase in temple demand from one to five pigs apparently delivered a substantial boost to the price."

Long-term price histories are clearer. Oil shows high but declining prices from 304 to 279 and then a long period of fluctuating but steady prices, contrary to the usual assumption of a steady decline. The decline can be explained by Delian attempts, once independent, to find non-Athenian sources for imports (for oil mostly Rhodes); high local prices should have encouraged local Kykladic plantings which would gradually supplant the Rhodian oil (Rhodian amphora handles drop 50% in the period after 275) -- at the same time the Nesiotic League "first tied the Kyklades together into a region focused on Delos". Firewood prices fluctuate considerably but at a lower average level before 218 than after, with unusually low prices in 250. Three possible causes: a law (ID 509) controlling sale of wood required its value to be declared before sale and no possibility of resale ("sellers would therefore prefer to set prices high"); economic expansion in the late 3rd C. meant a rising demand for wood for construction and probably for all wood; deforestation on neighboring islands. Pig prices are very high in 302/1 (when Demetrios Poliorketes and 10,000 troops were in the area) then fluctuate around a low mean before 200 and a higher mean thereafter. Surprisingly, pig prices moved in the same direction as wood prices from year to year (after a delay): presumably the pigs would feed amid the trees ("pannage"); and rising wood prices would increase wood harvesting, reducing the acreage available for pannage. (There are no significant relationships between pigs and oil or oil and wood.) Study of the relationship between military activity in the area and price fluctuations confirms that only firewood was drawn from a relatively broad trade network: "Delos was a Kykladic backwater, very unlike its subsequent incarnation under the Athenians in the late second and early first centuries B.C."

The following chapter, "The Rent Histories of Estates and Houses," confirms the findings of the previous chapter (once the estates are properly grouped by presence or absence of vines): from 270 BC on oil prices predict rents for vineless estates very well; the same is not true of wood or pig prices (which we have seen together move independently of oil) but is confirmed by barley prices, with which "estate rents move in exact conjunction". [There is a problem with the chart here, which does not record the disjunctive rise of rents in 269, but presumably this error is not repeated in the statistics.] Thus, as barley and oil prices were set by the local market so too must have been rents, though the relationship between oil and estates, which lacked olive trees, is considerably less obvious than barley, which we can assume was their main crop (not herding as earlier scholars surmised). The well-known drop in prices in 290 is best explained as a combination of a post-independence rise in rents once non-Athenians (or Athenian sympathizers) were allowed to rent and a subsequent series of defaults culminating in a sacred law (ID 503) with stricter requirements for guarantors, which considerably reduced the appeal of the estates (and increased the seizure of private property for default). Wine prices seem to have played a part in the collapse of rents after 220, not because of an international price decline as others have argued but because of cheap foreign competition.

At last R. can write a partial economic history of Delos under independence: 314-290 high estate rents and declining oil prices; 272-212 commodities and house prices level, estate prices gradually declining; 207-167 house rents, wood and pig prices rise while oil remains steady and estates, except for those without vines, continue to fall.

R.'s anti-Aegean market stance turns what could be a numbing series of charts and statistical print-outs into a virtual narrative that impels the reader on to the denouement. Yet the data are not compromised, just efficiently presented, with supporting detail relegated to seventy pages of appendices. Linear regression is effectively deployed where useful (and only there); extrapolation is done with the widest parameters; and those for whom Durbin-Watson and the F-statistic are not daily fare have numerous graphs to view (though I missed one on wood and rents, whose non-significant relationship would have provided a valuable contrast). I am sorry that daily wages did not fit into R.'s argument since his good sense might have produced some interesting comparanda, but I was delighted by the range of topics that did result, such as population, per capita grain consumption, cultivatable acreage, yield, rainfall, sailing season, biennial production of olives, cultic requirements, birth cycle of pigs, coppicing, pruning olive trees.

R.'s book is important for the questions it asks, the many ways it addresses them, and the care with which its argument is built. As R. says, this is only the beginning, but he has mapped the terrain and the rest should be relatively easy. This book is not only for Delian specialists or ancient economists but for any one interested in studying the ancient world. It is a fine addition to a remarkably substantial series.