Bryn Mawr Classical Review 03.04.10

Paul Millett, Lending and Borrowing in Ancient Athens. Cambridge: Cambridge University Press, 1992. Pp. xiii + 368. $59.50. ISBN 0-521-37333-6.

Reviewed by Edward E. Cohen, Philadelphia, PA.

This eagerly-awaited book, fifteen years in preparation, will seriously disappoint those who have expected the definitive "catalogue of almost nine hundred loan transactions of all types, drawn from the whole of classical Greece" which M. promised almost a decade ago [Trade in the Ancient Economy (1983) p. 43]. In Finley's words in 1984, the volume which his pupil M. "is now preparing" was to provide "the evidence" to confirm Finley's conceptualization of the Greek economy: "the impact of the arithmetic" would permit Finley's critics to be "rightly and briefly dismissed" (Ancient Economy2 197-98, and n. 83).

Lending and Borrowing, however, contains no arithmetic -- and no Greek. Rather than a systematic analysis of all relevant Athenian evidence, M. prefers "the deliberate suppression of detail which appears to be less significant in favour of material that is judged to be crucial" (p. 4). Specific passages, often highly ambiguous and open to varied interpretation, are offered only in English translation, or more often in paraphrase. By M.'s choice, "no Greek script appears in the text or notes" (p. xi).

The entire volume is devoted to a single, purportedly definitive thesis ("model"): the pervasiveness of credit in fourth-century Athens is essentially a manifestation of reciprocal social relations embedded in Athenian society -- it does not reflect a monied economy responsive to economic considerations. Early chapters consider a number of ideological issues (the possible origin of Athenian credit in reciprocal gift-giving, the purposes for which loans were sought, the political implications of borrowing, the role of interest). But the bulk of the volume contains an analysis of the sources of financing at Athens, what M. calls the "structure of credit."

This is history in the "Annales" tradition which, seeking to replicate common practice in the social sciences, generates conclusions by positing a theoretical "model" to be validated or rejected by factual analysis. For historians trying this approach, the principal difficulty has been the tendency to reconstruct the past to coincide with the theory, a danger which M. does acknowledge [the validity of an "ideal type" or model "can be established only by testing against other evidence" (p. 4)], but does not avoid. To the contrary, the poor implementation of M.'s methodology is the major shortcoming of this otherwise well-written, well-researched and well-organized volume: material compatible with the author's thesis is stressed, even where compatibility is obtained through interpretations not mandated by the evidence; material in conflict with the thesis is suppressed -- belittled as "exceptional" or irrelevant, or negated by brief, sometimes superficial argument; the findings of individual scholars almost invariably are treated with respect when compatible with the thesis, derisively dismissed when embarrassing.

In fact, the "deliberate suppression of detail" is so complete that the volume deals only with a portion of "the structure of credit": consumer finance among Athenian male citizens. M. judges maritime finance an "exception" to his model and largely avoids it -- although he concedes its importance (pp. 188 ff.). He ignores completely the clandestine (aphanês) economy (which is well-documented in the sources). "Women" are absent even from his index, and almost entirely from the book -- despite his acknowledgement in a footnote that "women could participate in credit operations involving valuable objects and substantial amounts of cash" (p. 290, n. 14). Widespread lending and borrowing by slaves, who are known to have operated autonomous businesses (including banks leased from their masters)1 and to have given and obtained credit, is entirely ignored, except for a half-sentence which fleetingly refers to "friendly loans by slaves" (p. 77). Metics and transients -- again of prime credit importance -- are little noticed. Excluding all this evidence, while possibly sparing the model some embarrassing refutation, vastly reduces the usefulness, and plausibility, of Lending and Borrowing.

As for the material which is included: a good illustration of M.'s method is his treatment of borrowing by Aeschines (a pupil of Sokrates, not the orator), in the author's opinion seminal evidence. In a fragment from Lysias (Gernet & Bizos Fr. 38, Athenaios 13.611d-12f), M. claims to find his "model of Athenian credit relations." The passage is discussed more than a dozen times in a text proper of only 221 pages, and revisited in Appendix II (p. 227). Indeed, the book opens with a five-page discussion of the excerpt and its implications, and closes with yet another treatment of this fragment (pp. 212-13), and other texts allegedly compatible, supposedly demonstrating the relative unimportance of "professional money-lending." M. justifies his extensive consideration and frequent citation of the Lysias passage as "provid(ing) a convenient introduction to my methodology, and to the chief characteristics of the structure of Athenian credit relations as presented in this study" (p. 1). It seems appropriate therefore to evaluate in detail M's use of this testimonium, which is indeed indicative of his methodology.

The reader's only direct knowledge of the passage is supplied in a translation "adapted from the Loeb edition, excluding for the sake of brevity less important sections of the fragment and all irrelevant comments" (p. 242). According to M. (p. 3), the fragment demonstrates that "when Aeschines wanted credit, he turned to a banker only in desperation," thus demonstrating M.'s thesis of "bankers as 'lenders of last resort.'" M. claims that

the fragments [sic] conveniently list (with humorous exaggeration) the various sources of credit that Aeschines had already exhausted before turning to the banker Sosinomus. He was no longer able to borrow from local shopkeepers, neighbours (implying friends), fellow-citizens in general through eranos-contributions (interest-free 'friendly' loans), or the men in the Piraeus who presumably specialized in high-risk maritime loans. The general impression is of a range or hierarchy of types of credit open to Athenian citizens. When a citizen had, like Aeschines, exhausted one or more of these sources, he moved on to the next available supply. Bankers were only the last in a long line of different species of credit.
The Lysias fragment, however, contradicts M.'s "model" on virtually every point. While M. places the banker "last in a long line," Lysias actually places him first -- it is the bank loan which is repaid by the proceeds of the financing with which the Lysian speech deals. Where M. conjectures a lineal progression, a "hierarchy of types of credit" to be exhausted in turn, the banker's loan was actually made in conjunction with a non-bank lender. Indeed, the bank borrowing, and the speaker's loan to repay the bank borrowing are the only transactions presented by Lysias as having occurred prior to the time of litigation: all the other allusions to Aeschines' dealings are presented in the Greek in the present tense, necessarily subsequent to the bank loan and seemingly contemporaneous with one another.

Quotation of the original text would have alerted readers of ancient Greek to the highly commercial context of the fragment, an environment of business borrowing largely muted in the author's "adapted" translation and completely inconsistent with M.'s model (which insists on Athenian credit as overwhelmingly consumption-oriented). Aeschines is a businessman conducting a perfume operation,2 obtaining bank, trade and other financing. Lysias' client's loan to Aeschines is actually for "working capital" for this perfume business (A)FORMH=S DE\ DE/OMAI). But M. alters the Loeb translation "I require capital" to "I need funds" (p. 1). Yet in another context M. describes Aeschines as having "acquired possession of a perfumery, but needed cash to run it." Since "working capital" (the cash needed to run a business) is the conventional, and correct, translation of aphormê,3 M. would have no reason to alter the Loeb translation -- other than its inconsistency with his model. Another example. To describe those who had dealings with Aeschines, the speaker uses the word KEXRHME/NOUS, a term replete with business implication: various forms of XRA/OMAI are routinely used to describe the congeries of dealings in banking relationships.4 But no indication of the actual Greek word, or its commercial implications, emerges from the author's translation, or from his commentary. In a further reference to business credit, the fragment tells of retail merchants (kapêloi) not receiving repayment for the advances which they have made, resulting in the closing of their shops. Although maritime lending is the one area which M. recognizes as contradicting his model (p. 189), he does not here note the implications, damaging to his thesis, of Aeschines' involvement with sea finance. And, of course, neither the original bank rate of 36% per annum, nor the still high 18% per annum yield on Aeschines' new loan suggest a financial culture embedded in mutual assistance and removed from considerations of monetary profit.

Of all the transactions alluded to in the passage, only one possibly involves a non-commercial loan, a reference to eranoi, translated by M. as "eranos-contributions," but described by him as "interest-free 'friendly' loans." The text of the Lysias fragment, however, says nothing about the terms, nature or circumstance of these eranoi, and M. concedes that nothing definitive is known from antiquity about the characteristics of eranoi. We do know, however, that eranoi loans were clearly not intended as supportive gifts, to be repaid from socially-sanctioned mutual assistance: to secure repayment, they were often collateralized by valuable property,5 and law suits were pursued against defaulting borrowers.6 Far from eranoi being part of an autonomous credit system which excluded professional lenders, we actually know of a banker who provided the funds for a loan intended as an eranos (see Dem. 53.9: financing by the trapezitês Theokles).

If the author had been willing with an "open mind" to evaluate his thesis against the ancient evidence, he likely would have seen that the Lysias fragment actually contradicts the major themes of his model. But M. is not reticent in rejecting an "open mind" as a desideratum or even an epistemological possibility; to the contrary, for M. an absence of prejudgment implies acceptance of traditional philological methods of "collecting evidence and interrogating it" without purposeful bias, examples in M.'s opinion of properly-despised, anecdotal antiquarianism (p. 4).

I have considered this passage at length, because M. has emphasized it as seminal and illustrative of his methods, and it is. The book is replete with similar questionable treatment of ancient material, exacerbated by the failure to cite original texts. Under the model, for example, "friends" are supposed to provide loans on "friendly" terms. But after concluding that "the common term for 'usurer' is obolostatês," citing approvingly a definition in the Etymologicum Magnum, M. can only lamely label the same source as "erratic" (p. 182) when it defines this "usurer" (obolostatês) as "a person who lends small sums to his philoi." Or consider the harsh terms exacted by the "xenoi" in Demosthenes 53 who lent money to ransom Nikostratos. In his recent work on xenoi, Herman, another doctoral student of Finley, places this loan fully "within the context of ritualised friendship."7 But the terms of the advance, a 100% premium if the loan were not repaid within thirty days, starkly violate the concept of reciprocal social relations proposed in M.'s model. The author, who correctly considers Nikostratos' borrowing activity to be prime evidence for "the main characteristics of credit in fourth-century Athens" (p. 53), refuses even to consider the impact of the terms on his model. Instead he rejects Herman's characterization of "these money-lending xenoi as a species of philoi": for M., "all the signs here indicate an impersonal relationship between strangers" (pp. 265-66). But there are no "signs" present other than what M. calls the "hostile" terms of the loan: no other information survives concerning the xenoi and this borrowing. Instead of accepting the clear meaning of the text, as even Herman does, M. reconstructs the past by altering the meaning of the passage to coincide with the "ideal type." Similarly the high interest rate that Apollodoros agreed to pay on behalf of Nikostratos "cannot be satisfactorily explained" (p. 57) and so is relegated to a footnote, rather than use d to test the model.

And what of the Athenian evidence of whose "deliberate suppression" M. boasts? Consider Demosthenes 36.11, a passage mentioned only once by M. (p. 204) and there disposed of in half a sentence. M.'s thesis necessarily posits the relative insignificance of the Athenian banks (trapezai) in providing credit: significant lending activities by the banks, impersonal providers of funds, would have involved numerous discrete fleeting transactions -- unrelated to traditional familial, social or political relationships. To explain why the bankers could not have been a primary source of credit, M. asserts that for lending they could utilize only their own money, not that of bank depositors. But Demosthenes says just the opposite. He defines banking as a business which produces from other people's money risk-laden revenues. This is a crucial piece of evidence, possibly the single most important surviving statement concerning the provision of credit at Athens. But it directly negatives M.'s theory. In his brief reference to the passage, M. dismisses Demosthenes' claim as untrue, although he is unable to cite any ancient evidence to the contrary. Demosthenes is corroborated by Bion: again the usual single citation reserved for unhelpful evidence is allotted to the philosopher's paradox which confirms bankers' access to other people's money ("How do bankers have money and yet not have it? Because that which they have is not their own"). Here too -- unsupported by ancient authority -- M. argues with contorted circularity against the actual surviving evidence, ultimately concluding that Athenian bankers probably could accept, but not freely use bank deposits (p. 205).

In contrast to this intolerance for material from Athens negative to its thesis, Lending and Borrowing abounds with frequent reference to supposedly-supportive material from societies far removed from the Athenian. M. finds modern British lending strikingly similar to Athenian,8 and uses the fish and fruit dealers of nineteenth-century Glasgow and similar denizens to establish the limited resources of "professional money-lenders" in Attica, "although evidence for Athens is almost entirely lacking" (pp. 180-88). Lending and Borrowing frequently employs such material, cited independent of context and culled from the secondary literature of disciplines in which M. lacks expertise. Although M. claims to "being only too aware of the pitfalls of the comparative method" (p. 18), this assertion again amounts only to wariness of comparative material which negates his model. He largely avoids use of comparative material on Roman credit relations -- he finds the orientation of Andreau's "massive and pain staking study" La Vie financière dans le monde romain (Rome 1987) uncongenial (p. 248, n. 34) -- but he is quick to cite a second-hand allusion to an outdated study of Rossel Island in the 1920's (p. 18) which seems supportive of his thesis. Again, assuming Athens to be a "pre-capitalist" or "non-capitalist" society, he appeals indiscriminately to other "pre-capitalist" or "non-capitalist" experiences in molding his model.9 But definitions of capitalism, and hence of "precapitalism" or "non-capitalist" societies, are notoriously impressionistic. If capitalism be understood in Weberian terms as "an actual adaptation of economic action to a comparison of money income with money expenses," then "the capitalistic enterprise and the capitalistic entrepreneur, not only as occasional but as regular entrepreneurs, are very old and were very widespread ... as in Babylon, Hellas, India, China, Rome." (Weber, Spirit of Capitalism 19-20).10

M.'s insistence on analyzing Athenian practices through modern terminology11 renders ironic his frequent, castigation of other scholars for falling into allegedly Modernist anachronisms.12 Even the arch-critic of "Modernism," Hasebroek, is charged with being "distinctly modernist in outlook" -- but only when he strays from M.'s model (p. 13). Stock charges of "modernism" are levelled even at scholars working strictly from the ancient evidence (which -- if in conflict with M.'s thesis -- is then arbitrarily rejected as "exceptional" or "atypical").13 But an errant savant may be partially redeemed through a stray conclusion compatible with the author's model. Thompson's conclusions concerning Athenian investment contradict the model: M. devotes many pages to exposing Thompson's "manipulation of the evidence" (p. 165). M.'s model is served by Thompson's conclusion that banks did not generally pay interest on funds left with them: here Thompson is "surely right" (p. 205). Bogaert's study of Greek bankers and banking is "dominated by modern banking theory" (p. 197) and is therefore "inadequate" (p. 201) in its various conclusions incompatible with the model. But when in agreement with M.'s model, his "conclusions about the mechanism s of ancient banking" (p. 202) are "a valuable corrective to the view prevailing among his predecessors" -- and successors. Ste. Croix's well-reasoned and persuasive examples ("part of an attack on Finley") are dismissed as "pure speculation" (p. 278 , n. 52), but M. accepts without comment his comforting defense of Dem. 33.7 as a non-maritime loan (p. 307, n. 23).

Lending and Borrowing is, then, a polemical brief for a partisan view, rather than the promised empirical confirmation of reciprocal social relations as the basis of Athenian credit. Although the author suggests that his may be "possibly the last full-length study of this aspect of Greek civilization" (xii), the nature of Athenian credit remains sub iudice: the major contribution of this volume may be the stimulation of fresh analyses.


  • [1] See Dem. 36.13-14, 37; Harrison, Law of Athens (Oxford 1968) I, p. 176, n. 2.
  • [2] M. translates KATASKEUA/ZOMAI TE/XNHN MUREYIKH/N as "I am setting up in the trade of making perfumes," altering the Loeb translation of "I am setting up the business of distilling perfumes," presumably substituting "trade" for "business" to avoid commercial connotation. But Greek technê, while similar to modern trades or professions in requiring personal commitment, knowledge and skill (see Xen. Oik. 1.1; Pollux 4.7, 22; Dem. 45.71), encompassed "businesses" such as banks (trapezai) and workshops producing and selling items such as beds and shields (Pollux 7.170, 7.155 and 7.159; cf. Dem. 27.9).
  • [3] For the extension of the meaning of A)FORMH/ from "starting point" to "working capital," see the compendium of uses in Korver, De Terminologie van het Crediet-Wezen in het Grieksch (Amsterdam 1934) 22-5.
  • [4] See Dem. 33.7 (Herakles' bank); Isok. 17.4, 6 (Pasion the banker). Cf. LSJ s.v. XRA/OMAI. C.III.4. M. overlooks this commercial usage when he attempts to associate the term with "reciprocal relationships between neighbors" (p. 54).
  • [5] Finley acknowledges this with understatement: "eranos-loans were not infrequently backed up by hypothecation." Studies in Land and Credit2 (New Brunswick 1985) 85. Cf. his discussion of the numerous eranos loans secured by real-estate encumbrance [Land and Credit (100-6)].
  • [6] The DI/KAI E)RANIKAI/ were even included by Aristotle among the DI/KAI E)MMHNOI (Ath. Pol. 52.2). Cf. Pollux 8.37. Harpokration joins the DI/KAI E)RANIKAI/ with the explicitly commercial DI/KAI E)MPORIKAI/ (s.v. DI/KAI E)/MMHNOI).
  • [7] Ritualised Friendship & The Greek City (Cambridge 1987) 93.
  • [8] For example, "the gap between loans to market traders in the Athenian agora and to costermongers in late-Victorian London is narrow" (pp. 181-82). M. seeks to find in sixteenth-century England (p. 206) the relationship between brokerage and banking denied by Bogaert in his extensive study of the Athenian evidence.
  • [9] M. uses "pre-capitalist" interchangeably with "non-capitalist" (pp. 15-18, 35, 71, 282 n. 15). In contrast, Marxist historians consistently refer to ancient societies as "precapitalistic" since dependent on "the forms of slavery, serfdom etc.," which preceded the utilization of nominally-free wage labor (Marx, Capital III.8.19). See Garlan, Les esclaves en grèce ancienne (Paris 1984) 218.
  • [10] For the semantic, historic and economic complexity of identifying "capitalism," and the varied uses of the term by Weber, see J. R. Love, Antiquity and Capitalism (London 1991). Goldsmith [in a relevant work also published by Cambridge but not cited by M.: Premodern Financial Systems: A Historical Comparative Study (1987)] has suggested "premodern" as a preferable, although still arbitrary term for financial structures predating the eighteenth century.
  • [11] Two examples: the modern notion of "credit-worthyness" frames his superficial discussion of pistis (pp. 7-8) ; his admirable investigation of philia (pp. 109-26), thoughtful and inclusive, is still set in Sahlins' "spectrum of reciprocities" [although the latter's concentration on kinship is incompatible with M.'s model, forcing a "break with Sahlins" (p. 111)].
  • [12] Braudel's suggestion of sophisticated banking mechanisms in ancient Greece (anathema to M.) "need not be taken too seriously" (p. 245, n. 13); Isager and Hansen are "unconvincing" but "display great ingenuity" (p. 243, n. 3); Hopper's competent chapter on finance and banking [Trade and Industry in Classical Greece (London 1979)] is "a hotch-potch of material, full of errors of fact and interpretation" (p. 248, n. 25). Glotz offers "confused misinterpretation" (p. 243, n. 4); the great Böckh, "a garbled survey" (p. 10).
  • [13] Despite our almost complete ignorance about the curricula vitae of any other bankers,the author asserts that "the careers of Pasion and Phormion were utterly untypical" (p. 254). At pp. 272-73 three embarrassing passages are dismissed as "atypical." The justification: the word "philergia" "comparatively rare" in M.'s opinion, appears in each.