Aristotle on justice in exchange: commensurability by fiat

Abstract

This essay offers an interpretation of Aristotle's remarks on the commensurability of goods in Book V of the Nicomachean Ethics It explores the term Tby hypothesis' (Ef UnoOéoemç) which Aristotle uses to describe the institution of currency through which commensurability is established. The term implies that Aristotle conceives the origins of currency to lie in a conscious act of stipulation rather than through a spontaneous process in which currency is established via the unintended consequences of individual action. In conclusion, contemporary theories of money are considered and it is asked with which Aristotle's conception of money aligns most closely.

Headnote

Abstract. This essay offers an interpretation of Aristotle's remarks on the commensurability of goods in Book V of the Nicomachean Ethics It explores the term Tby hypothesis' (Ef UnoOéoemç) which Aristotle uses to describe the institution of currency through which commensurability is established. The term implies that Aristotle conceives the origins of currency to lie in a conscious act of stipulation rather than through a spontaneous process in which currency is established via the unintended consequences of individual action. In conclusion, contemporary theories of money are considered and it is asked with which Aristotle's conception of money aligns most closely.

Keywords: Aristotle, commensurability, money, exchange, Nicomachean Ethics

Introduction

This essay offers an interpretation of an aspect of Aristotle's remarks on commensurability in the Nicomachean Ethics (Nie. Eth). It attends to a term that has received much scholarly attention in the context of Aristotle's logical works but not in the context of Nie. Eth. The term, Tby hypothesis' (Ef UnoQéosmç), is used by Aristotle to describe the way in which currency, qua unit for measuring the value of goods, is established (1133b21). After considering Aristotle's analysis of commensurability, the term by hypothesis is examined along with its cognate noun, hypothesis(UnôOemç), as they are used in Aristotle's Prior and Posterior Analytics. Aristotle's usage of the term in the Analytics sheds light on his discussion of commensurability in Nie. Eth. Many interpreters use by convention' or a close synonym to translate Ef UnoOéoemç in Nie. Eth. The many meanings of 'convention', however, engender ambiguity which can mislead. The ambiguity is captured in definition 9a of the Oxford English Dictionaryin which 'convention' is defined as tg]eneral agreement or consent, deliberate or implicit, as constituting the origin and foundation of any custom, institution, opinion, etc.'.[1] The 'deliberate/implicit' opposition makes for ambiguity, for a convention, custom, institution, etc. can be established deliberately - according to a plan or as a result of explicit agreement - but conventions can also arise implicitly - without design and therefore not as a result of explicit agreement. We think of natural languages and many of their rules when we think of conventions arising in this second, 'implicit', sense. The 'invisible hand' processes through which such conventions come into being are named after Adam Smith's adage about the promotion of the general good via individuals' intentions to promote only their own gain (Smith, 1976/1776, IV.ii.9). This understanding of convention is popular amongst economists in their attempts to explain the emergence of social institutions. A canonical example brings us close to the topic of this essay, for it concerns the origin of money (Menger, 1871). The 'implicit' conception of convention is not an appropriate way to approach to Aristotle's account of the origins of currency. For Aristotle, if something is established byhypothesis, it follows from a formal agreement, explicitly laid down. The subject is the term kata sunthëkën (Kar O ovvQrfKtfv), with which Aristotle describes the way in which currency comes to serve as a representative of need (Nie Eth. 1133a29). The term kata sunthëkën, like E UnoOéoernç, is often rendered by Aristotle's translators as 'by convention'. The term should, however, like Ef UnoOéoernç, be interpreted as a formal agreement rather than as an implicitly arising convention. The essay concludes with reflections on Aristotle and modern theories of money. In this section, what Aristotle means by the terms commonly translated as 'exchange' are subject to scrutiny, whereby it is argued that his terms for exchange are far more encompassing than the modern understanding of market exchange suggests. It is argued that Aristotle's discussion of 'justice in exchange' in Nic Eth addresses types of exchange which go beyond the scope of market exchange.

Commensurability and justice in exchange

For Aristotle, commensurability is the foundation of the community: a community or association cannot exist without exchange, for its individuals are not self-sufficient. If exchange is to be just, the items exchanged must be equalized, for which there must be a measure according to which goods are valued. Of this measure Aristotle (Nic. Eth, 1133a19-20) writes:

All items for exchange must be comparable in some way. Currency has come along to do exactly this, and in this way it becomes an intermediate, since it measures everything, and so measures excess and deficiency - how many shoes are equal to a house.[2]

He adds (Nic. Eth., 1133a26-31):

In reality, this measure is need, which holds everything together.... And currency has become a sort of pledge of need (chreia), kata sunthekeq in fact it has its name (nomisma) because it is not by nature, but by the current law (nomos), and it is within our power to alter it and to make it useless

Though things so different cannot become commensurate in reality, they can become commensurate enough in relation to our needs. Hence there must be some single unit fixed Ef UnoOéoeoç. This is why it is called currency; for this makes everything commensurate, since everything is measured by currency (Nic. Eth., 1133b18-23).

For the time being, two phrases - kata sunthëkën and Ef UnoOéoeoç - will be left untranslated. Both phrases have been translated as 'by convention', though, other translations exist. In the second of the three passages quoted above, Aristotle tells us that need is the measure, whereas, in the third, it is currency.[3] Neither 'solution' to the problem of commensurability is free of philosophical difficulty (Gallagher, 2012; Meikle, 1995), though many scholars, Meikle (1995) being a notable exception, see the need solution as Aristotle's final word on the matter of establishing commensurability (Finley, 1970; Judson, 1997; Will, 1954). Aristotle does not explain how currency becomes a 'representative' or 'pledge' (hupallagma) of need, but he does make clear that it is currency which does the measuring, even if currency is subordinate to need (Nic. Eth, 1119b26-27):

We call wealth (choematà) anything whose worth is measured by money.

Money is supplied as a common measure; everything is related to this and measured by it (Nic. Eth, 1164a1-2).

Currency, for Aristotle, is a human invention, its purpose to measure the value of different goods in exchange. While pronouncing on the function of currency as a measure (and as a store of value which allows for deferred payment (Nic. Eth, 1133b10-13)), Aristotle does not examine its historical origin. The verbs he uses with currency (nomisma) draw attention to currency's existence: nomisma, he writes, 'came along' (elëluthe) (Nie. Eth, 1133a20) in order to make all goods comparable (sumbiëta); currency must 'exist' (einai) (1133b19), lest exchangeable goods be incommensurable (summetra) and exchange thus unjust (Nic. Eth, 1133b16); currency is 'provided' (peporistai) (Nic. Eth, 1164a2) as a measure. These verbs remain vague about a historical act of establishing currency.

Aristotle states that it is impossible 'in reality' (alëtheia) (Nie. Eth, 1133b19) for diverse goods to become commensurable. If commensurability exists, its source lies outside the nature of the goods exchanged; goods, that is, can only be commensurable Ef UnoOéoernç - 'by hypothesis'. My purpose, in the two sections which follow, is to explore the terms left untranslated when citing Aristotle above. The first, Ef UnoOéoernç, will be the focus of the section following two subsections, whilst the second, KarO ouvOqKqv, which Aristotle uses in his discussion of need, is the subject of a subsequent section.

E Ùnodéosrnç (by hypothesis')

Let us start with the noun, hypothesis (ùnôOemç), the literal meaning of which is 'something laid down (under)', Its many meanings in Aristotle will be reviewed later, [4] but here, we examine the technical approach to hypothesis as an element of logical argument in the Posterior Analytics (Post. An.). It should be noted that, when the term is transliterated into a Romanised form - hypothesis - the word will be italicized to alert the reader to the danger of misconceiving the Greek term if one associates with the English word 'hypothesis'. As we will see, that there is, for Aristotle, nothing necessarily 'hypothetical' about a hypothesis

Aristotle defines a hypothesis as a type of 'posit' (thesis) that assumes one or other part of a statement (apophansis). That is, we have a hypothesis if we assume either that X is (exists)' or X is not' (Post. An., 72a9-24). [5] Aristotle alludes to a learning situation in which a statement is a hypothesis for the pupil if she assumes the statement to be true and provable without either proving it herself or having it proven to her. Such a hypothesis has validity relative to the learner (Post. An., 76b23-34). If the learner holds no opinion about the truth of the statement concerned, or if she opposes it, the statement is not a hypothesis but a 'postulate' (alterna). Because it is provable, a hypothesis is not 'merely hypothetical', in the modern English sense, that is, something conjectural or of uncertain validity (Wallace, 1981, p. 52; Wolfsdorf, 2008, p. 44). And because it is provable, a hypothesis differs from a 'fundamental principle' (arche), for a fundamental principle is 'immediate' and therefore not amenable to proof, that is, not derivable from something else (Post. An., 72a6-8; Metaphysics, 1005b14) (see Upton, 1985, pp. 287-288).

In the Prior Analytics (Pr An, I. 23, 29, 44), the term by hypothesis appears in Aristotle's discussion of 'syllogisms by hypothesis (onXkoyiopdi Ef ùnoOéoernç). These syllogisms proceed when one's interlocutor, I, agrees to accept a proposition, q, on the condition that another proposition, p, be proven to her. 'If p then q is the hypothesis here, and once p is deduced, /must accept q (Lear, 1980, p. 34; Strohach, 2001, p. 251). Aristotle offers an example (Pr. An., I. 44, 50a20-28) which involves two negative propositions, the first of which is deduced syllogistically, the second laid down hyhypothesis [6] The first proposition, ~p, is: 'not every capacity (dunamis) is of contraries'. This proposition is amenable to logical demonstration. [7] The second proposition, ~q, is: not every science (epistêmë) is of contraries'. The hypothesis upon which the protagonists agree is: ~p ^ ~q, that is, if one accepts 'not p' (or ~p), then one must accept 'not q' (or ~q). Once the hypothesis, ~p ^ ~q, is accepted, one s interlocutor is honour-hound, as it were, to accept that 'not every science is of contraries' once it has been demonstrated that not every capacity is of contraries. Whilst ~q may, like ~p, he amenable to logical demonstration, in the case at hand, it is not demonstrated hut assumed hy the hypothesis ~p ^ ~q.

Other syllogisms hyhypothesis are reductiones ad impossihile which involve an agreement to the hypothesis in advance of the demonstration to he given, as in the following example from the commentary on Pr. An. hy Alexander of Aphrodisias (259, 20-29). If one wishes to deduce that 'no human can fly', one hypothesises the opposite, viz., 'some humans can fly'. One adds to this the generally accepted premise: 'all flying things have wings', and one deduces the conclusion (if one has postulated that 'some humans can fly'): 'therefore some humans have wings'. The obvious falsity of the conclusion leads one to reject the hypothesis ('some humans can fly ), and so the opposite claim - 'no human can fly' - stands accepted.

The foregoing exposition is hut a schematic presentation of what Aristotle means hy hypothesis, and it does not do justice to the complexity of Aristotle's discussion. It nevertheless serves the present purpose of showing whence the term hyhypothesis comes. We may summarise hy listing three facets of the term as follows:

i) In syllogisms hy hypothesis, the hypothesis takes the form of a 'concession' (homología) (Pr. An., I. 23, 41a40) on the part of the disputants who agree to the hypothesis One party stipulates the hypothesis to which the other must agree if the demonstration is to proceed.

ii) The agreement described is made in advance of the demonstration and is a presupposition of the latter. Taking this and the foregoing point together, we may say there is an explicit, conscious act of laying down what one is assuming in advance of the demonstration. [8]

iii) A hypothesis is not derived from anything prior, and although it may be susceptible to derivation, its function in syllogisms by hypothesis is to provide the ground for what follows.

With this in mind, we may pursue the meaning of by hypothesis in Aristotle's ethical and political works.

By hypothesis in the Nicomachean Ethics

The following analysis is based on the hypothesis (!) that by hypothesis in Nic. Eth is used analogously to its use in the Analytics Aristotle's use of the same term in different works is thus not coincidental, and the sense of byhypothesis in the Analytics gives us a clue to its sense in Nie. Eth. Aristotle does not give a detailed explanation concerning arguments by hypothesis in the Analytics, something usually attributed to his audience's familiarity with hypotheses, as expounded in Plato's Meno (86e-89c) (cf. Striker, 1979, p. 34), in which Socrates proves that excellence is a variety of knowledge with the aid of the hypothesis that, if excellence is knowledge, it must be teachable (89c). Socrates likens his use of hypothetical argument with regard to excellence to its use in geometry (86e). To approach Aristotle's use of by hypothesis in Nie. Eth. the following paragraphs examine his use of the term in other, related, works by Aristotle.

In the Eudemian Ethics (Eud Eth.), Aristotle states that deliberation does not extend to the end (telos) of action in the productive or fabricating arts (poië tikais) because in these arts, one deliberates about means with an end already in sight; the end, here, is described as the foundation (arche) and hypothesis of action (Eud. Eth, 1227a9-10, 1227b29-30). Aristotle does not, as he does elsewhere, distinguish here between hypothesis and archë, according to their derivability or immediateness (Post. An., 72a6-24; cf. Metaphysics 1005b); his point is that the end is a given presupposition of action. Aristotle gives the example of a doctor, for whom it is given, like a hypothesis, that a patient should be made healthy; this is not a matter on which doctors deliberate, for only the means to make a patient healthy are subject to deliberation (End. Eth, 1227b25-6). Aristotle repeats this thought in Nie. Eth (1151a15-19), when he describes the incontinent person:

For virtue preserves the origin (arche), while vice corrupts it; and in action the end we act for is the origin, as the assumptions (hypotheseis) are the origins in mathematics. Reason does not teach the origins in either mathematics or in actions; [with actions], it is virtue, either natural or habituated, that teaches correct belief about the origin.

It is, as it were, virtue which provides the first principles of action, and hence the virtuous person does not have to deliberate about the end of action.

In the Politics (Pol), Aristotle avails himself numerous times of the term hypothesis. Sometimes, hypothesis refers to an assumption or premise of an argument (Pol, 1261a16-17, 1263b30, 1329a21). It is also used in the sense of a principle which stands behind a concept or institution, e.g. the principle of aristocracy, democracy or of a polity (Pol, 1269a33, 1273a4, 1317a36, 40, 1328b39); here, as in Eud. Eth. cited in the previous paragraph, Aristotle uses hypothesis as synonymous with arche (origin). Hypothesis can refer, too, to the principle behind a taxonomy or the headings of a classificatory scheme (Pol., 1300b14, 1314a26). In three places in the Politics, Aristotle uses the term by hypothesis. At Pol 1332a8-12, he refers the reader to Nie. Eth (1098a16) and its view that happiness (eudaimonid) is the complete realization and exercise of virtue. He is at pains to let the reader know that this fact is not 'conditional' (by hypothesis) but rather true without qualification (haplös) or necessarily (tanagkaia). Aristotle also notes that children are not citizens in the same way as adults; children are only citizens by hypothesis (Pol., 1278a5). The two latter uses of by hypothesis signify that something is conditional or subject to qualification. They resemble his use of by hypothesis in the discussion of commensurability in Nie. Eth There, as we noted, goods are not commensurable in reality but are so under a certain condition, that condition being when the values of those goods are considered according to the measure of currency.

Drawing Aristotle's uses of by hypothesis together, then, and applying them to his discussion of commensurability in Nie. Eth., we may say that Aristotle understands the coming into being of currency by hypothesis as follows: currency arises through explicit agreement, it is stipulated and agreed to as a measure of value, and its existence is a presupposition of the commensurability of goods which only become commensurable conditionally, that is, relative to the hypothesis through which currency comes into existence. The italicised words in the foregoing give us three aspects of by hypothesis- agreement, stipulation and presupposition. All are to be found in translations of by hypothesis in Nie. Eth. Although some translators translate by hypothesis as by convention', the term is most often translated into English as 'agreement', though 'stipulation' is also common, and 'presupposition' less so. The fourth aspect of by hypothesis- the conditionality aspect - is captured in translations of the term not only as 'by stipulation', but also as 'by arbitrary usage'. The 'arbitrariness' of the hypothesis captures the conditionality of commensurability, for goods are not by nature commensurable but have to be made thus by some stipulation which is arbitrary relative to the nature of the goods themselves. [9]

Only if agreement is understood to be an explicit and conscious act of assent does it concur with the assent given to a hypothesis in syllogisms by hypothesis of Aristotle's Analytics, where the interlocutor or pupil agrees to accept a hypothesis proposed by the instructor. Let us now turn to the other term - KarO ovvQrfKtfv - which is mentioned by Aristotle in his discussion of commensurability and which, too, is often translated as 'by convention'.

Need andkata sunthëkën

Aristotle's need solution to commensurability, which was referred to, proposes that exchangeable items are made commensurable - sufficiently, at least, for the purpose of assuring justice in exchange - in relation to people's need. Currency, according to this view, arises as a 'representative' or 'pledge' (hupallagma) of need. Currency comes to represent need kata sunthëkën (Nic Eth., 1133a29). Like the term by hypothesis, kata sunthëkën has the sense of an agreement, though 'convention' is also offered as a translation. [10] Let us explore the connotations of kata sunthëkën by looking at its use in other parts of the Aristotelian opus.

In the Politics, the noun sunthëkë signifies a formal compact between allies (Pol., 1280a38, 1284a40). It is synonymous with summachia (alliance, usually military) and sumbolon (treaty) (Pol, 1280a39-40).[11] Aristotle also avails himself of the term sunthëkë in the passage immediately after his discussion of commensurability in Nie. Eth In discussing political justice, he observes that what is 'legal and conventional' (nomikon kai sunthëkë') (Nic. Eth, 1134b32) changes according to the constitution of the polis in which law is made. 'Conventional' is a reasonable translation of sunthëkë here if it carries the sense of formal agreement (cf. de Ste. Croix, 2004, pp. 328-329). With regard to justice, Aristotle tells us, those things which are 'just by agreement and expediency' (ta de kata sunthëkën kaito sumpheron tön dikaiön) (Nie. Eth., 1134b35), are analogous to weights and measures, for they vary from place to place. Indeed, this is the case of political constitutions generally, for they differ from one another. The important point for our discussion is the sense of kata sunthëkën as 'by agreement', for weights and measures in Greek poleis were 'prescribed by law (de Ste. Croix, 2004, p. 329). The formally stipulated and enforced weights and measures make it obvious that any institution which is 'just by agreement and expediency' must, like the norms of Greek metrology, be formally laid down and agreed.

As in the case of by hypothesis, then, sunthëkë and kata sunthëkën are best understood in the sense of a formal agreement or pact (Liddel and Scott, 1996). This concurs with Aristotle's use of sunthëkë in the Prior Analytics, when he states that the premises of hypothetical syllogisms are not amenable to analysis but are 'conceded by agreement' (alla da sunthëkë s hômologë menos) (Pr. An. I. 44, 50a17-18). Sunthëkë here signifies the prior agreement that parties give to a particular hypothesis, usually in an explicit way. As Walter Leszl (1981, p. 316) writes, in some cases 'the term hypothesis is treated as synonymous with homología or with sunthëkë, in evident allusion to the agreement that is presupposed by or reached in a dialectical debate'. We may therefore conclude that Aristotle describes the representation of need by currency kata sunthëkën to indicate that this representation is based on formal agreement. The way in which the representation ensues may be deemed 'conventional' in that it might differ from polis to polis, but the explicitness of the agreement is unmistakeable. We have therefore reached the conclusion that Aristotle's term byhypothesis implies a formal act of deliberate establishment. But Aristotle does not offer us further details about how the unit which denominates value is established. The following summary provides a characterization of the way in which the measure that currency provides (Nie. Eth., 1133b20-1) comes into being:

Just exchange within a community presupposes a measure of the value of goods. This measure is currency which exists for the purpose of providing a common unit and thus of making goods commensurable. The commensurability effected by currency is neither a consequence of the nature of goods, nor is it the product of a convention if the latter is understood in its 'implicit' sense. Currency and the ensuing commensurability are arbitrary stipulations Once the members of the community have agreed to this stipulation, exchangers can ascertain whether the goods they proffer in exchange are of equal value to those they accept in return. There will thus be justice in exchange.

Conclusion: Aristotle, 'exchange' and modern theories of money

Prima facie, Aristotle's discussion of currency in the Nicomachean Ethics has, as its focus, market exchange within the political community. This, at least, is how most commentators understand the fifth book of the Nicomachean Ethics If this interpretation is correct, currency thus manifests itself as a medium of exchange in Aristotle's work, yet anterior to this function of currency is its function of unit of account. Without a unit which ensures commensurability of the things exchanged, there can be no justice in exchange, and hence the unit is a presupposition of just exchange. Hence Aristotle's analysis in the Nicomachean Ethics aligns itself with those modern analyses of money for which the unit of account function is primary. J. M. Keynes' (1930, pp. 3-5) is one such analysis, for it states that MONEY-OF-ACCOUNT ... is the primary concept of a Theory of Money'. Keynes was influenced by the 'Chartalist' (or state) theory of money, developed to its highest form in the twentieth century by Georg Friedrich Knapp (1923), for whom the unit account function of money is also of prime importance to money's use as a means of payment. The Chartalist theory has recently been revived in modern economics (see, e.g., Wray, 1998; 2012). Aristotle, as argued above, holds that the unit of measure - currency - is stipulated by agreement (by hypothesis).

In contrast to Aristotle's account on currency in the Nicomanchean Ethics, the account he offers in Book I of the Politics focuses on the origins of money in the context of long-distance trade. The explanation of money in the Politics makes no reference to justice but instead to convenience: currency overcomes the obstacles to exchange when the latter is conducted as barter. This account seems to align itself more closely to the orthodox economic position that money arises in the context of market exchange, whereby a particularly 'saleable' commodity becomes the dominant medium of exchange (cf. Menger, 1871). Aristotle might then be held to have two theories on the origin of money. One might hold these views to be mutually incompatible, but before one ascribes inconsistency to Aristotle, one should note two things. First, his analysis of currency follows a tradition in Greek thought which divides exchange within the political community (the subject of the Nicomachean Ethics) from external exchange or long-distance trade between different communities (the subject of the Politics).

In Homer's Iliad and Odyssey, long-distance trade (emporia) is well attested, with the Phoenicians being its main purveyors. Trade amongst Greeks who inhabit different political communities, or amongst Greeks living in the same political community, is, on the other hand, scarcely attested in Homer (Peacock, 2011). Hesiod's Works and Days follows a similar pattern, whereby trade within the community of Ascra is not attested, though Hesiod has much advice for the long-distance trader. The distinction between the internal and external movement of goods stretches its tentacles into the classical period in which Plato and Aristotle wrote (fourth century BCE). It is reflected in the distinction in terminology for 'retail traders' (kapeloi), who hawk their goods within the political community, and long-distance traders (emporoi) (Peacock, 2016). Aristotle can be placed into this tradition of ancient Greek thought. In the Ethics, Aristotle is concerned with justice in the context of intra-community exchange, for the cohesion of the community depends thereon. But justice is not a concern in the context of long-distance trade.

There is thus a decisive difference between the two types of exchange - intracommunity and long-distance. One may even say that Aristotle's two discussion of money do not pertain to the same thing. To support this argument, the types of exchange to which Aristotle refers will he examined; for whereas longdistance trade (as discussed in the Politics) answers to the name of commerce, the exchange which forms the context for the discussion of currency in the Nicomachean Ethics is not so easily subsumed under the heading of commercial activity. One can provide support for this statement hy considering the terminology with which Aristotle describes 'exchange'.

There are numerous terms, all used in Book V of Nie Eth, which can he (and are) translated as 'exchange': allaktikais (Nie Eth., 1132b32), metadosis (Nie Eth., 1133a2), antidosin (Nia Eth., 1133a6), metadidonai (Nic. Eth., 1133a10), allage (Nie. Eth, 1132b13, 1133a19, 24, 28, 1133b11, 15, 17, 26). Let us investigate the connotation of these terms to descry the extension of what we ascribe to Aristotle with the term 'exchange'.

The first, allaktikais, refers not only to commercial exchange but also to the exchange of gifts (cf. Plato's Sophist, 223c). Gift exchange was both a Mycenaean and an archaic (Homeric) practice amongst the Greek élite, and there are good reasons for holding that the aim of the practice was not the making of profit at the expense of one's gift-exchanging partners. Rather, gift exchange was a method of conducting 'foreign policy' (Donlan, 1982, p. 149), whereby the leader of a wealthy estate (oikos) cemented friendly and peaceful relations with others who might live at some distance. This is not market exchange as understood in contemporary or ancient societies.

The second term with which Aristotle describes exchange is metadosis The term is used three times in the Politics In one instance it describes pre-monetary exchange (barter) between independent families or peoples (1257a24). A second instance comes in the context of a comparison between a true political community and an 'alliance' (summachia): people who live in proximity to one another and regulate their doings in a way that prevents wrongdoing in their exchanges (metadoseis) would not, Aristotle insists, constitute a political community (Pol., 1280b18-24). Aristotle continues:

Let us suppose that one man is a carpenter, another a farmer, another a shoemaker, and so on, and that their number is ten thousand: nevertheless if they have nothing in common hut exchange, alliance, and the like, that would not constitute a state (polis).

A polis, Aristotle writes, does not consist merely in people sharing a common place with the purpose of avoiding mutual harm and for the sake of exchanging goods; rather a polis is a community of families which exists for the sake of complete and self-sufficient life (Pol., 1280b30-31). The passage applies to the exchangers Aristotle describes in Nic. Eth. who are not mere exchangers of goods looking out only for their own interest in their transactions, for if they were, they would constitute only an alliance of people. To constitute a polis presupposes friendship between them and the aim of a good and happy life (Pol., 1280b38-1281a1). This conception of the polis will occupy us again presently, hut first we must attend to Aristotle's terms for exchange. Metadosis, in the sections of the Politics just cited, clearly refers to mercantile affairs. Does this imply that his use of metadosis at Nie. Eth. (1133a2) likewise refers to commercial exchange? The answer is 'no', and that for the following reasons:

i) Metadosis may include mercantile exchange but it is not limited to its sense of commerce. Elsewhere (Pol. 1321a26), Aristotle uses the term in the sense of allowing a share in something (specifically, allowing the people a share in the running of government). [12]

ii) In Nie. Eth. (Book V), in which metadosis is used, Aristotle refers to the grouping in which exchangers are held together. The group is not an 'alliance', in the sense just described, but a community (koinonia) or a polis. This implies that the sense of exchange to which he refers with the word metadosis stretches beyond commercial exchange, for Nie. Eth. is not concerned with mere alliances but with communities based on friendship which exist for the purpose of pursuing a good and happy life.

Taking (i) and (ii) together, we may hypothesise that Aristotle's use of metadosis in Nie. Eth. takes on the broad meaning of sharing rather than the narrow meaning of commercial exchange. This can be appreciated better if we consider the context of Aristotle's use of metadosin.

Immediately after Aristotle uses the term metadosis, he adds the following comment, one of the least analysed in his remarks on exchange:

that is why they make a temple of the Graces prominent, so that there will he a return of benefits received. For this is what is special to grace; when someone has heen gracious to us, we must do a service for him in return (Nie. Eth. 1133a2-5).

The three Graces (Chantes) were goddesses whose name is derived from the ancient Greek word for grace (jàpiç), which designates a type of pleasure of benefit. Charis signifies a convention of reciprocity, whereby the receipt of a favour or benefaction is to he reciprocated (MacLachlan, 1993, chapter 1). Commentators often relate chans to the sort of reciprocity involved in gift exchange (MacLachlan, 1993; Wilkinson, 2013). One of few commentators to have treated this allusion to the Graces in detail and to have integrated it into the analysis of justice in exchange is Robert Gallagher (forthcoming), who offers an ingenious analysis to the equalisation involved in exchange, as Aristotle perceives it. The reader is referred to Gallagher's essay, but here it is merely noted that Gallagher's analysis coheres with that of the present essay, for Gallagher's solution to the riddle of proportionate reciprocity in exchange implies that 'Aristotle reaches outside the realm of the material to social goods to complete the transaction' (p. 13). That is, Aristotle's understanding of exchange (here, metadosin) is not to be understood as market exchange in the conventional modern sense. The reason for this is that exchange between members of the same community should involve reciprocity between the exchangers which is not accomplished if one considers only the value of the products they are to exchange.

Antidosin is the next term for exchange to be considered. It is used but once in Nie. Eth., Book V. The term has a technical sense in Athenian law (Christ 1990), the context for which concerns the public benefactions or liturgies which the wealthiest Athenians were obliged to make for the benefit of the polis. Aristotle is not adverting to this sense in Nic. Eth. Metadidonai (1133a10) is another term which refers to exchange. It is used not only in Nie. Eth, Book V, hut also during the discussion of friendship. Like metadosis, metadidonai connotes sharing. At Nie. Eth. 1171b, it refers to sharing of one s bad fortune with friends, whilst at 1177a8-9, it is used to convey the idea that a slave does not share (partake) in happiness. Again, a purely mercantile meaning cannot he ascribed to the term, and this counsels a look heyond what is conventionally considered to he market exchange when we interpret Aristotle.

The final term for exchange, allage(s), stems from the verb allasso (to change, alter) and is the word Aristotle most frequently uses for exchange in Nie. Eth. It refers to the exchange of goods (as in Plato's Republic 371h), and Aristotle uses it at Nie. Eth. 1132h13 to explain the concepts of loss (zemia) and profit (kerdos). Aristotle also uses the term to descrihe harter in the Politics (1257a13,19) as well as to descrihe the exchanges hetween allies who do not live in one and the same polis (as descrihed ahove in the elucidation of Aristotle's term metadosis). The verh ukXaoorn is the term which most closely approximates what we understand hy market exchange, though, in Nie. Eth., in contrast to Politics (Book I) and Plato s Repuhlic, Aristotle avoids standard words for trade/trader (emporiaJemporos), retail/retailer (kapelike/kapelos), huying (öneomai) and selling (poleo), and 'clings to the neutral word 'exchange ", thus avoiding the connotation of commercial or market exchange (Finley, 1970, p. 14). This choice of words would he that of one who aims to show that exchange is (or should he) emhedded in the rules of communal justice (Finley, 1970, p. 8).

The foregoing discussion of exchange highlights the capacious understanding of exchange which Aristotle treats. Part of the difficulty in interpreting Book V of the Ethics arises when one assumes that 'exchange', for Aristotle, corresponds to what one commonly understands hy market exchange. In particular, the analysis of his terminology reveals the non-commercial senses of the term as used in Nie. Eth.. Had Aristotle wished to focus on purely 'economic" exchange, a different terminology stood at Aristotle's disposal. Édouard Will (1954, p. 218) concurs with Gallagher (cited ahove) when he writes that exchange in Nie. Eth. 'is situated within a more comprehensive scheme of social ethics' than reference to commercial exchange would imply. Sitta von Reden (2003, p. 185) echoes the point when she states that Aristotle's pronouncements in Nic. Eth. are 'clearly not statements on justice in the market place, but on justice in the social interaction between citizens'. Included in this social interaction is exchange in the sense of redistribution between citizens, and redistribution occurs, inter alia, in the realm of benefactions (liturgies) by the wealthy to communal activities and goals. Currency, as observed above, is essential in making items exchanged commensurable and hence in ensuring justice in exchange. But in light of this, it becomes clear that if a function of currency is to be singled out to be of particular import to Aristotle, it is that of a means of payment, whereby payment is to be understood far more broadly than payment in market exchange.

Footnote

Endnotes

[1] 'convention, n. 9a', OED Online. June 2013 (Oxford 2013) emphasis added, <http://www.oed.com.ezproxy.library.yorku.ca/view/Entry/40714?redirectedFrom= conventions

[2] I base translations of ancient Greek works on the translations given in the bibliography. Where I modify them, I have used the Greek versions available on the Perseus Digital Library: http://www.perseus.tufts.edu/hopper.

[3] Translating xP£ia as 'need' rather than 'demand' is argued for particularly well by Judson (1997, pp. 158-160), though see Danzig (2000, pp. 414-415).

[4] See Wolfsdorf (2008) for further usages.

[5] Many commentators identify hypotheses with existential statements (McKirahan, 1992, p. 43; Gómez-Lobo, 1976-77, p. 436). Aristotle is not consistent in applying the existential definition, as McKirahan (1992, p. 47) concedes; see also Barnes (1975, pp. 103-104) and Robinson (1953, pp. 100-103).

[6] I draw on Strobach's (2001, pp. 252-253) interpretation in what follows.

[7] Alexander of Aphrodisias supplies a proof of the first proposition in his commentary on the Prior Analytics (386, 31-6)

[8] See Leszl (1981, p. 293). Also Robinson (1953, pp. 94-95, 105) for whom 'positing' (nûn^i) is deliberate in that it is consciously doing something which we need not do'.

[9] Uses of 'agreement' to translate ££ Unoûéosoeç include the English translations of Nic. Eth by Chase (1911), Crisp (2000), Rackham (1934), Ross rev. Brown (2009), Rowe (2002), Thomson (1976) and Welldon (1927). German translations likewise prefer the 'agreement' ( Übereinkunft) translation, e.g. Rieckher (1856), Rolfes (1911), Stahr (1897), as do some Spanish translations which use the term 'acuerdo', e.g. Bonet (1985), Araujo and Marias (1970). 'Stipulation' or 'posit' is used by Apostle (1984) and Irwin (1985), and 'presupposition' by Bartlett and Collins (2011) as well as by Gigon (1967) who uses the term Voraussetzung. "By arbitrary usage' comes from Ostwalds translation.

Those who use 'convention' include the English translators Peters (1909) and Warrington (1963) and the French translators Gauthier and Jolif (1970), Tricot (1990) and Voilquin (1961). Less common are translations which use the term 'hypothesis', e.g. Bodéüs (1990) - 'fixée par hypothèse' - and Natali (1999) - 'per ipotesi'. Natali, acknowledges the awkwardness of using 'hypothesis' ('[Un genera si traduce ex hupotheseos con 'per convenzione") but defends the translation on etymological grounds.

[10] For 'convention', the reader is referred to translations by Apostle, Araujo and Marias, Bodéüs, Bonet, Crisp, Gauthier and Jolif, Irwin, Peters, Rackham, Ross, Rowe, Thomson, Voilquin and Warrington. For 'agreement', see Bartlett and Collins, Chase, Gigon, Natali, Ostwald, Rolfes and Stahr.

[11] See also Pol. (1275a10); [Const. Ah.] (LIX.6); de Ste. Croix (2004, pp. 328329). Plato uses the term ouvüfan in the sense of 'collusion' (Laws, 879a).

[12] Xenophon (Cyr. 8.2.2) uses metadosis in this sense when he describes one of Cyrus' methods of increasing his popularity, namely, giving a share (^STàôooiç) of food and drink to others. Robert Gallagher (forthcoming) translates metadosis, as it is used in the Nie. Eth. as 'giving-of-a-share' in order to denote the particular type of reciprocity which Aristotle has in mind in his discussion of justice in exchange.

References

References

Alexander of Aphrodisias (2006), On Aristotle's Prior Analytics 1.32-46", trans. Ian Mueller, London: Duckworth.

Aristotle (1956), Metaphysics, trans. John Warrington, London: Dent.

Aristotle (1964), Prior and Posterior Analytics, trans. John Warrington, London: Dent.

Aristotle (1975), Posterior Analytics, trans. Jonathan Barnes, Oxford: Clarendon.

Aristotle (1985), Nicomachean Ethics, trans. Terence Irwin, Indianapolis: Hackett.

Aristotle (1996), The Politics and The Constitution of Athens, trans. Benjamin Jowett (revised Stephen Everson), Cambridge: Cambridge University Press.

Barnes, Jonathan (1975), Aristotle's Posterior Analytics, Oxford: Clarendon.

Christ, Matthew (1990), 'Liturgy avoidance and antidosis in classical Athens', Transactions of the American Philological Association 120, 147-169.

Danzig, Gabriel (2000), 'The political character of Aristotelian Reciprocity', Classical Philology 95 (4), 399-424.

de Ste. Croix, Geoffrey (2004), Athenian Democratic Origins, Oxford: Oxford University Press.

Donlan, Walter (1982), 'Reciprocities in Homer', Classical World75 (3), 137-76.

Finley, Moses (1970), 'Aristotle and economic analysis', Past and Present 47, 325.

Gallagher, Robert (2012), 'Incommensurability in Aristotle's theory of reciprocal justice', British Journal for the History of Philosophy20 (4), 667- 701.

Gallagher, Robert (forthcoming), 'The role of grace in Aristotle's theory of exchange', Méthexis, page proofs online under: http://s3.amazonaws.com/academia.edu.documents/32872884/Grace_iMethexis_p roofs.pdf?AWSAccessKeyId=AKIAJ56TQJRTWSMTNPEA&Expires=14735513 57&Signature=O1X9HvfxxboI4iPJ7Gp7h0iaosI%3D&response-contentdisposition=inline%3B%20filename%3DThe_Role_of_ Grace_in_Aristotles_Th eory_o.pdf.

Gómez-Lobo, Alfonso (1976-77), 'Aristotle's hypotheses and the Euclidian postulates', Review of Metaphysics 30 (3), 430-439.

Hesiod (1996), Hesiods Works and Days: A Translation and Commentary for the Social Sciences, translated and edited by David Tandy and Walter Neale, Berkeley: University of California Press.

Homer (1951), The Iliad of Homer, trans. Richmond Lattimore, Chicago: University of Chicago Press.

Homer (1991), The Odyssey of Homer, trans. Richmond Lattimore, New York: HarperPerennial.

Judson, Leslie (1997), 'Aristotle on fair exchange', Oxford Studies in Ancient Philosophy 15, 147-175.

Keynes, John Maynard (1930), A Treatise on Money, Volume I: The Pure Theory of Money, London: MacMillan.

Knapp, Georg Friedrich (1923), Staatliche Theorie des Geldes, 4th edition, München and Leipzig: Duncker & Humblot.

Lear, Jonathan (1980), Aristotle and Logical Theory, Cambridge: Cambridge University Press.

Leszl, Walter (1981), "Mathematics, automatization and the hypotheses', in Enrico Berti (ed.), Aristotle on Science: The "Posterior Analytics ", Padova: Editrice Antenore, pp. 271-328.

Liddel, Henry and Scott, Robert (1996), An Intermediate Greek-English Lexicon, 9th edition, Oxford: Clarendon.

MacLachlan, Bonnie (1993), The Age of Grace: Charis in Early Greek Poetry, New Jersey: Princeton University Press.

McKirahan, Richard (1992), Principles and Proofs: Aristotle 's Theory of Demonstrative Science, New Jersey: Princeton University Press.

Meikle, Scott (1995), Aristotle 's Economic Thought, Oxford: Clarendon.

Menger, Carl (1968), Grundsätze der Volkswirtschaft, Tübingen: J.C.B. Mohr, 1871.

Peacock, Mark (2011), "Rehabilitating Homers Phoenicians: on some ancient and modern prejudices against trade', Ancient Society 41, 1-29.

Peacock, Mark (2016), "Wealth and commerce in archaic Greece: Homer and Hesiod', in Eugene Heath and Byron Kaldis (eds.), Wealth, Commerce, and Philosophy, Chicago: Chicago University Press, forthcoming.

Plato (1975), The Laws, trans. Trevor Saunders, London: Penguin.

Plato (1987), The Republic, second edition, trans. Desmond Lee, London: Penguin.

Plato (1990), Sophist, trans. William Cobb, Boulder, Colorado: Bowman and Littlefield.

Plato (2005), Meno and Other Dialogues, trans. Robin Waterfield, Oxford: Oxford University Press.

Robinson, Richard (1953), Plato's Earlier Dialectic, second edition, Oxford: Clarendon.

Smith, Adam (1776), An Inquiry into the Nature and Causes of the Wealth of Nations, Oxford: Oxford University Press, 1976.

Striker, Gisela (1979), 'Über Syllogismen "aufgrund einer Hypothese"', Hermes 107 (1), 33-50.

Strobach,Niko (2001), 'Schlüsse aus Annahmen bei Aristoteles: eine argumentationstheoretische Deutung des syllogismus ex hypotheseôs', Zeitschrift für philosophische Forschung 55 (2), 246-257.

Upton, Thomas (1985), 'Aristotle on hypothesis and the unhypothesized first principle', Review of Metaphysics 39 (2), 283-301.

von Reden, Sitta (2003), Exchange in Ancient Greece, second edition, London: Duckworth.

Wallace, William (1981), 'Aristotle and Galileo: The uses of UnôOsoiç (Suppositio) in scientific reasoning', in Dominic O'Meara (ed.), Studies in Aristotle, Washington D.C.: Catholic University of America Press, pp. 47-77.

Wilkinson, Lisa (2013), Socratic Charis: Philosophy without the Agon, Lanham, Maryland: Lexington.

Will, Édouard (1954), 'De 'l'aspect éthique des origins grecques de la monnaie', Revue Historique 212 (2), 209-231.

Wolfsdorf, David (2008), 'The Method s£, unoOsosmç at Meno 86e1-87d8', Phronesis 53 (1), 35-64.

Wray, L. Randall (1998), Understanding Modern Money: The Key to Full Employment and Price Stability, Cheltenham, Gloucester: Edward Elgar.

Wray, L. Randall (2012), Modern Money Theory. A Primer on Macroeconomic Systems for Sovereign Monetary Systems, Basingstoke, Hampshire. Palgrave MacMillan.

Xenophon (1956), Cyropaediavol 2, trans. Walter Miller, London. Heinemann.