Jump to ContentJump to Main Navigation
Ancient Greek History and Contemporary Social Science$

Mirko Canevaro, Andrew Erskine, Benjamin Gray, and Josiah Ober

Print publication date: 2018

Print ISBN-13: 9781474421775

Published to Edinburgh Scholarship Online: January 2019

DOI: 10.3366/edinburgh/9781474421775.001.0001

Show Summary Details
Page of

PRINTED FROM EDINBURGH SCHOLARSHIP ONLINE (www.edinburgh.universitypressscholarship.com). (c) Copyright Edinburgh University Press, 2021. All Rights Reserved. An individual user may print out a PDF of a single chapter of a monograph in ESO for personal use. Subscriber: null; date: 29 July 2021

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

(p.157) 5 Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts
Ancient Greek History and Contemporary Social Science

Federica Carugati

Barry R. Weingast

Edinburgh University Press

Abstract and Keywords

This chapter puts pressure on the ‘mass and elite model’ of Athenian litigation introduced by Ober. According to this framework, litigation is a game played by elite litigants and mass audiences; the ‘masses’ constitute a monolithic body with identical preferences; the ‘elites’ are thoroughly aware of, and willingly play by, the rules set by the masses. Moving from a different interpretation of Athenian political sociology, this chapter builds a new model of Athenian litigation that modifies Ober’s in three important respects: first, the jurors’ preferences are not the product of a monolithic and static ‘mass’ ideology; second, litigants (not only elites) can reasonably predict the location of the median juror; and third, litigants’ arguments are the product of a cost‐benefit analysis that depends a) on the relative expected position of their opponent; b) on the expected position of the median juror; and c) on the policy/legal agenda they are pursuing. The model proposed here suggests that repeated interactions in the law‐courts allowed diverse interests to be advanced and negotiated, which helped the Athenians collectively define the boundaries of their social relations while responding to the new challenges that a post‐imperial, highly fragmented Greek ecology posed to Athens’ stability and prosperity.

Keywords:   Athenian litigation, mass and elite, ideology

1 Introduction

In the Athenian law-courts, wealthy, educated and well-born elites fought one another to prevail as leaders and advisors of the masses. Regulated by the masses’ ideals of a good society, elite competition pushed Athens towards stability, prosperity and cultural immortality. Or did it?

Classical scholarship has long analysed interactions in the law-courts as a contest among elite litigants and mass jurors. The ‘mass and elite model’ (henceforth, M&E) owes its original elaboration to the work of Josiah Ober (1989). For Ober, Athenian political stability in the fourth century was the product of ideological negotiations between masses and elites carried out through the medium of rhetoric. Rhetorical discourse allowed the Athenians to manage the tensions arising from the existence of economic and social inequality in the context of a democratic system based on political equality. In the decision-making institutions of the polis – including, but not limited to, the law-courts – the masses regulated elite ambition by pitting elite advisors against each other and by bestowing honours on good advisors.1 In Ober’s own words (1989: 333), ‘the ideological hegemony of the masses effectively channeled the fierce competitiveness of elites … into patterns of behaviour that were in the public interest’. In this way, the Athenians were able to reap the benefits of having elite advisors – which were necessary for policy in a direct democracy – while preventing the elite from becoming a ruling elite – because their power was always provisional and subject to the dêmos’ delegation.

(p.158) For analytical purposes, we make explicit three implicit assumptions in the M&E model:2

  • Assumption 1 (sociological assumption): Litigation in Athens is a game played by elite litigants and mass jurors.

  • Assumption 2 (operative assumption): Elite litigants seek to win over their opponents as a means to gain honour.

  • Assumption 3 (operative assumption): Mass jurors have monolithic preferences.

In this chapter, we draw on the findings of two separate bodies of literature, one in classics, and one in political science, to put pressure on these assumptions. Our goal is to provide a more nuanced account of the dynamics regulating decision-making processes in the Athenian law-courts.

Before we continue, however, a note of clarification is in order. A complete account of decision-making in ancient Athens would track the process in its entirety: from the deliberations of the Council (boulê), which put agenda items on the Assembly’s calendar, to the negotiations that took place in the Assembly itself, and finally into the law-courts, where new laws and controversial pieces of legislation (both laws and policies) were either rejected or approved. Modelling this entire process is beyond the scope of this chapter. Our knowledge of procedural rules regulating Council and Assembly meetings is limited (at least compared to the law-courts, but see Canevaro in this volume). Moreover, modelling a sequential game involving three institutions of which we know relatively little would magnify the margin of error, compromising the soundness of the entire endeavour. As more work continues to illuminate the rules regulating the decision-making process in the boulê and the Assembly, others will take up the task. But the law-courts offer a promising starting point.

In the fourth century, the law-courts did not deal with all legislation passed in Athens: however, they dealt with laws (through the procedure of nomothesia and the graphê nomon mê epitêdeion theinai), (p.159) while the Assembly dealt with decrees. But if a decree presented in the Assembly was challenged as unconstitutional, it ended up in court though the graphê paranomon procedure.3 So the courts dealt with both laws and policy. Most importantly, the courts remained the locus where controversial measures (both decrees and laws) were debated and ultimately either rejected or approved. We assume that legislation that promoted major departures from the status quo would be more likely to be challenged in court. Therefore, modelling the law-courts enables us to cast a wider net over the legislative process, as well as to capture the mechanisms that regulated the high-end domain of law- and policy-making, where rule change was more likely to occur.

How, then, did the Athenians make law and policy in the law-courts? Recent work on Athenian society, politics and economics suggests that (1) proposers of laws and decrees do not belong solely to the Athenian elite; (2) jurors were representative of the Athenian population as a whole; and (3) extensive social networks linked adult, male citizens with other citizens and a wealth of other actors (including slave, metics, foreigners and women). In light of this evidence, we suggest that litigants and jurors interacting in the law-courts brought to the fore meaningfully pluralistic preferences that included, but were not limited to, those produced by their individual identities as adult, male, mass or elite citizens. Having put pressure on the M&E model’s sociological assumption, we seek to understand how diverse actors with pluralistic preferences interacted under the procedural constraints imposed by the law-courts. To do so, we compare the M&E model with the median voter theorem (MVT) – a model used extensively in political science to study the dynamics of institutions using majority rule. Based on these findings, we build a new model of Athenian litigation that modifies the M&E model in two fundamental respects. First, since jurors’ preferences are meaningfully pluralistic, litigants (who are not only elites) are uncertain about the exact location of the median juror on a given policy/legal issue. Second, we assume that litigants want to win as a means to gain honour, but they also have preferences over policy outcomes. Our model shows that, under these assumptions, and given the courts’ procedural rules (large and representative jury panels, majority rule, secret ballot and lack of deliberation), litigants had incentives to move towards, but (p.160) not necessarily to converge on, the position of the median juror when formulating new legislative proposals. We conclude that interactions in the courts enabled diverse preferences to be advanced and negotiated in ways that fostered both stability and innovation in Athenian law- and policy-making.4

The chapter proceeds as follows: section 2 challenges the sociological assumption of the M&E model – that Athenian litigation was a game played by elite litigants and mass jurors. In section 3, we put pressure on the model’s operative assumptions: namely, that mass ideology is monolithic, and that the elites engage in litigation solely to win over their opponents as a means to gain honour. In section 4, we present our model, and in section 5, we discuss its main implications. Our conclusions follow.

2 Beyond Mass and Elite

Who are the players of Athenian litigation? In this section, we rely on a recent and fast-growing body of literature on Athenian society, politics and economics whose findings put pressure on the notion that litigation was a game played by elite litigants and mass jurors.5 We draw on these findings to present a more nuanced picture of the actors’ identity and preferences. Before we delve into this analysis, however, we make one preliminary point.

The picture of Athenian litigation as a game between masses and elites emerges clearly and forcefully from the literary evidence – first and foremost the Attic orators. However, the Attic orators provide us with an abysmally incomplete and biased record. If we assume that the Athenians litigated c. 2,000 cases a year, then over the 80 years covered by the corpus, the c. 150 extant speeches amount to 0.09 per cent of the total.6 Moreover, as scholars consistently acknowledge, (p.161) accidents of publications and transmission have skewed the extant evidence towards over-representing elite participation in litigation. Still, scholars assume that the courts remained a primarily, if not exclusively, elite forum.7

In what follows, we move beyond the evidence provided by the orators in order to formulate three arguments: first, litigants did not solely belong to the elite; second, jury panels were representative of the Athenian population as a whole; and third, litigants and jurors participated in a thick web of social networks which demanded that individuals negotiate their preferences based on membership in various social groups.

2.1 Was litigation an exclusively elite domain?

Two claims on which most accounts of Athenian litigation are grounded are mutually exclusive: if we believe that the Athenians were so litigious that they litigated at least 2,000 cases a year, then we cannot also believe that the elite did all the litigating.8 The numbers, simply, don’t add up.

We are not the first to advance such an argument. In 1998, P. J. Rhodes moved from Hansen’s calculations of the volume of litigation in Athens to suggest that ‘in the law-courts as in the Assembly, the man who was a mere voter on one day may well be an active participant on another’ (Rhodes 1998: 145). Following Rhodes, Victor Bers maintains that evidence of sub-elite participation in litigation emerges from various, though perhaps less than reliable, sources: in courtroom speeches, the litigants themselves sometimes describe their socioeconomic condition as sub-elite, but the obvious advantages of doing so in terms of (mass and elite) rhetorical self-presentation suggest that we should be wary of any such statements (Bers 2009: 8–10); similarly, comedy’s frequent references to Athenian litigiousness provide plentiful evidence of sub-elite participation, but such evidence remains largely inferential (Bers 2009: 13–20). Like Rhodes, Bers seems to rely on Hansen’s calculations to drive the point home: if the courts were convened between 175 and 225 days a year, and if the average number (p.162) of empanelled jurors ranged between 1,500 and 2,000 per day (the exact number depending on the day’s mix of public and private cases), then ‘it is reasonable to assume that, other things being equal, the greater the proportion of small cases, the greater the number of idiôtai who appeared in court’ (Bers 2009: 21; cf. Lanni 2009: 710–3; 2016). For Bers, then, idiôtai (i.e. the poor or average Athenian) litigated in court with some frequency, particularly in private, small cases.

Bers’ conclusion prompts the question: were elite litigants exclusively or primarily involved in prominent public cases revolving around major political issues? Although the evidence from oratory may suggest that this was indeed the case, M. H. Hansen’s study of the number of decree-proposers in the Assembly in the period between 355 and 322 shows that legislation was a common form of political activity not dominated by a small elite. Hansen suggests that

in addition to the political leaders there was an important group of politically-minded citizens who were active, even as proposers, but only occasionally and not professionally; [that] the number of citizens involved in politics as proposers (and not only as voters) was much larger than usually believed, and there was no sharp distinction between the professional, the semi-professional and the ordinary citizen; [and that] hundreds of minor and probably also some major political figures are completely unknown to us.

(Hansen 1984: 155)9

Similarly, Claire Taylor has shown that decree-proposers

were not concentrated in the city area but came from demes throughout Attica … If decree proposers came from all corners of Attica and included a large number of citizens, then we might assume that the composition of the Assembly was quite varied. Thus, a wide range of citizens both attended and participated actively in the Assembly, and the Athenian democracy was not merely the concern of a privileged few.

(Taylor 2007b: 340)10

(p.163) Moreover, Taylor’s analysis of prosopographical evidence from a sample of 2,183 citizens reveals that non-urban participation was not limited to decree-proposers, but encompassed all sorts of office holders, including ‘stratêgoi, tamiai, bouleutai … archons, phylarchs, hipparchs, dikasts and other (unknown) officials recorded on pinakia’ (Taylor 2007a: 73). For many of these magistracies, the use of the lot contributed, over time, to reduce the over-representation of the city elite in Athenian politics (Taylor 2007a: 88; 2007b: 336–40). As a result, in the fourth century, ‘non-wealthy citizens played an increasingly active role in democratic politics’ (Taylor 2007a: 89).

Although both Hansen’s and Taylor’s findings focus primarily on the Assembly and magistrates, their conclusions have obvious implications for the law-courts as well: as decree-proposers, sub-elite idiôtai would have been at risk of graphê paranomon, and thus would have appeared as litigants in the law-courts in major public cases and not only, as Bers suggests, in small private disputes. Similarly, as magistrates of the polis, non-wealthy citizens would have had to undergo accountability procedures (dokimasia, euthyna) and risked accusations of misconduct (notably, eisangelia) that would have landed them in court as litigants in public cases.

The evidence discussed above does not allow us to conclude that participation was essentially uniform across the mass and elite divide. However, it does suggest that we should temper the picture provided by the orators with the image that emerges from other sources. Accordingly, we conclude that non-elite actors participated sufficiently frequently in litigation that we should not consider litigation as an exclusively or perhaps even primarily elite domain.

2.2 Who were the jurors?

At the beginning of every year, 6,000 citizens over the age of thirty were selected by lot to serve as jurors in the popular courts. On each day when the courts were in session, jurors would be assigned to jury panels through the allotment mechanism known as klêrôtêrion. Panels ranged in size from 201 jurors for small private cases to 6,000 (p.164) jurors for sensational public cases.11 The use of the lot and allotment mechanisms sought to guarantee that jury panels were broadly representative of the Athenian population: according to Hansen, ‘the 6000 were divided into ten sections each of 600 men and each section comprised sixty men from each tribe’ (Hansen 1999: 183). But tribe representation tells us little about the sociological composition of Athenian juries. First, the yearly selection occurred among those who had volunteered their name, with no barriers to entry. Second, jurors could iterate their mandate in subsequent years. As a result, representation in the courts was not subject to the strict rules that, according to Taylor, randomised the selection process in other magistracies and prevented the monopolisation of power by any one group, especially over time. Did, then, any one group dominate the courts? Scholars discuss three potential biases that may have skewed the composition of juries: provenance, age and wealth.

2.2.1 Provenance

In a world of walkers, it seems reasonable to suppose that city folk would enjoy greater access, and thus greater representation, due to their proximity to the courts. Prosopographical studies of jurors’ plaques (pinakia), however, rule out the city bias. As Hansen remarked, ‘astonishingly, it turns out that people from the coastal demes (paralia) and the inland (mesogeios) actually prevail over people from the city demes (asty)’. Even controlling for migration, ‘the geographical spread revealed by the jury plaques is so striking that the growth of the city population cannot be the whole story’ (Hansen 1999: 184; Taylor 2007b: 336–7; on pinakia, Kroll 1972). City folk, Hansen suggests, did not dominate the courts.

2.2.2 Age

To an extent, the Athenians built an age bias into the composition of the jury. The privilege of sitting as jurors in court was reserved for citizens over the age of thirty, a restriction that did not apply, for example, to the Assembly. According to Hansen (1999: 89), two thirds of all adult citizens were over the age of thirty.

Assuming that the fourth-century citizen population numbered 30,000, then 20,000 citizens were, every year, eligible for jury service. Of these, (p.165) 6,000 were selected as jurors. The numbers thus suggest that, if rotation had regulated the selection of jurors, each adult male citizen (regardless of his socioeconomic status) would have served as a juror once every three years. This, however, was not a rule.

2.2.3 Wealth

The issue of the socioeconomic composition of Athenian juries has been hotly debated. Scholars have variously identified the typical juror as either a poor or a well-off citizen, based on the (skewed) evidence of the orators, the (dubious) references in Aristophanes’ comedy or the (subjective) assessments of the consequences of jury pay.12 We take the absence of a consensus on the issue to suggest, albeit indirectly, the absence of explicit or obvious socioeconomic biases. Therefore, we reconstruct the likely composition of juries based on the wealth distribution of the Athenian population as a whole. For this purpose, we rely on the composite picture recently offered by Geoffrey Kron (2011).

Kron’s study has slightly modified the picture of Athens’ fourth-century wealth distribution that J. K. Davies offered some thirty-five years ago (Davies 1981). For Kron, out of a total male population of c. 31,000, 300 people (or c. 1 per cent) belonged to the super-rich liturgy-paying class; 1,200 people (or c. 4 per cent) belonged to the rich eisphora-paying class; and 5,000 people (or c. 16 per cent) possessed no landed wealth and owned movable property amounting to c. 100 drachmas. As a result, assuming that Kron also accepts Davies’ assessment that c. 9,000 people (or c. 29 per cent) owned property amounting to at least 2,000 drachmas, this leaves about 15,500 people (or c. 50 per cent) owning some amount of wealth between 100 and 2,000 drachmas.13 Kron’s distribution of wealth – like other measures of inequality such as income (Ober 2017), nutrition (Lagia 2015) and house size (Morris 2004) – suggest that the Athenian ‘masses’ were a rather heterogeneous middle class.

In sum, we take the absence of explicit biases in terms of provenance and wealth as indirect evidence that Athenian juries were probably representative of the Athenian population as a whole in these respects. Moreover, even if we acknowledge the likely over-representation of older Athenians among jurors, this was partly by design, due to the (p.166) Greeks’ association of age with wisdom. As a result, age should be considered more as a requirement than a bias. But are provenance, age and wealth the only relevant determinants of individual preferences?

2.3 Social networks and plural preferences

But all associations are parts as it were of the association of the state. Travellers for instance associate together for some advantage, namely to procure some of their necessary supplies. But the political association too, it is believed, was originally formed, and continues to be maintained, for the advantage of its members … But the other associations aim at some particular advantage; for example sailors combine to seek the profits of seafaring in the way of trade or the like, comrades in arms the gains of warfare, their aim being either plunder, or victory over the enemy or the capture of a city; and similarly the members of a phylê or deme. And some associations appear to be formed for the sake of pleasure, for example thiasotai and eranistai, which are unions for sacrifice and social intercourse.

Aristotle, Nicomachean Ethics 1160a14

Athenian litigants and jurors were, for the most part, adult, male citizens.15 But as the quotation above suggests, adult male citizens participated in a thick web of social, economic and religious associations that thrived at the sub-polis level.16

In his monograph La cité des réseaux (2010), Paulin Ismard argues that Athens was a multi-layered society featuring many associations that tended for the most part to be open and non-hierarchical (Ismard 2010 (p.167) : 71). Although Athenian associations did not enjoy juridical status as persons (like, for example, today’s American corporations do), they were recognised in Athenian law: specific legal procedures may have been designed to solve disputes concerning commercial partnerships (dikai koinônikai) and loans (dikai eranikai) among members; associations may have also provided supporting speakers (synêgoroi) when members were summoned to court; finally, associations could own land (Ismard 2010: 146–9, 149–52, 163–79). Most of all, however, Athenian associations shaped citizens’ religious and economic life (Ismard 2010: chs 3 and 4).17 Ismard’s study thus depicts a world of tensions between centralised and decentralised institutions and concludes that ‘l’identité politique serait moins le résultat d’une pratique communautaire spécifique que le produit de ces espaces médiateurs où s’échelonnent et se configurent respectivement, dans un va-et-vient constant, les différentes identités dans la cité’ (‘Political identity was less the result of a specific communal practice than the product of mediating spaces where different identities in the city stretch and reconfigure themselves in a constant to-and-fro movement’; Ismard 2010: 251).

In a recent collection of essays, Claire Taylor and Kostas Vlassopoulos suggest that the focus on the polis has led scholars to over-emphasise the distinction between the citizen club and the outsiders – namely, women, metics and slaves – in shaping social identities (Taylor and Vlassopoulos 2015; Vlassopulos 2007). For the authors, although distinctions of status, of course, did matter in some important respect, there is a growing awareness that in many other contexts they were irrelevant or secondary. A number of works have already shown that in some arenas of social interaction, for example finance and banking, status distinctions were mostly irrelevant (i.e. E. Cohen 1992); similarly, in the case of religious associations that brought together citizens, metics and even slaves, status distinctions seem to be largely secondary. (Taylor and Vlassopoulos 2015: 3) (p.168) (But note Kierstead 2013: 51–2, suggesting that there is no positive evidence for mixed citizen–non-citizen associations before 325.) In her contribution to the same volume, Taylor examines a series of dedications illuminating the life of associations that brought together citizens and non-citizens. In light of this evidence, Taylor persuasively argues for the existence of communities and identities based less on legal status and more on networks of work, residence and cult that profoundly affected people’s identities, as well as their well-being (Taylor 2015, and in this volume).

Adult, male citizens did not exist in a social vacuum. The evidence discussed above suggests that in many forums of Athenian civic life, identities and preferences depended only in part on rigid distinctions in terms of gender and legal status. Accordingly, we suggest that participation in sub-polis associations contributed to shape individuals’ identities and preferences in ways that were meaningfully pluralistic.

3 The M&E Model and the MVT

How did the plurality of preferences discussed in the previous section affect litigants’ and jurors’ behaviour in the law-courts? In the next section (section 4), we present a formal model of Athenian litigation that is a variation on the basic MVT (Hotelling 1929, Black 1948 and Downs 1957). The MVT has been used extensively to study the dynamics of American institutions using majority rule, such as national elections.18 But the basic structure of the MVT fits the dynamics of Athenian litigation better than it does contemporary electoral practice for two reasons. First, the MVT is based on the assumption that voters are fully informed when casting their ballots (Ansolabehere 2006: 31). But whereas this assumption is problematic in modern, large-scale, representative democracies, the costs of becoming informed were much lower in Athens’ smaller-scale, participatory democracy.19 Second, the MVT assumes that voters face a binary choice (Ansolabehere 2006: 31). But whereas democratic politics in contemporary nation-states rarely involves a binary choice, Athenian litigation did: in fact, in the law-courts, litigants presented opposing arguments to the jury (such as convict/do not convict, approve/reject legislation) and the jury voted for one or the other outcome.

(p.169) In this section, we introduce the basic MVT and formulate two arguments. First, the assumptions of the basic MVT closely resemble those of the M&E model. For this reason, the two models can be fruitfully compared. Second, the long history of criticisms waged against the basic MVT offers a series of insights into how we can modify the M&E model’s operative assumptions to better capture the dynamics that regulated law- and policy-making in the Athenian courts.

The basic MVT can be articulated as follows:

  • Assumption a: Politicians seek to win office over their opponent.

  • Assumption b: Each voter has an ideal preference over a given policy issue that he or she prefers over all the other alternatives (technically called a ‘single-peaked preference’).

Based on these assumptions, the basic MVT predicts that competition for office will lead candidates to propose the median voter’s ideal policy.

The problem with the MVT, as Stephen Ansolabehere remarks in his overview of the literature, is that its ‘central prediction … does not hold. Candidates and parties generally do not converge on the median voter’s ideal policy’ (Ansolabehere 2006: 31). In an attempt to explain why the central prediction of the model does not hold in practice, political scientists have subjected its assumptions to intense scrutiny. We begin by discussing the assumption concerning the preferences of politicians (Assumption a), and then we move on to discuss the assumption concerning voters’ preferences (Assumption b).

3.1 Litigants’ preferences: Honour or policy?

In the basic MVT, candidates do not care about policy per se, but advocate policies in a purely instrumental way – namely, to win office and defeat their opponents. As noted, the MVT predicts that, in order to win a majority of votes, candidates will converge on the position of the median voter. But in many settings, candidates do not converge. Many explanations have been offered to account for this divergence, including factors such as ‘primary elections, incumbency advantages, political parties, […] interest group contributions’ (Ansolabehere 2006: 35; see more generally pp. 35–8). These analyses have led political scientists to modify the assumption concerning politicians’ preferences in order to produce a more nuanced understanding of the forces shaping candidates’ positions. In particular, Randall Calvert (1985) elaborated a version of the model in which (p.170) candidates are motivated by policy concerns as well as by the desire to win office.20

In the M&E model, litigants also want to win and defeat their opponents. For Ober, ‘Rich and well-born Athenians competed vigorously, sometimes savagely, with each other for political influence … This pattern of behaviour is not surprising in light of the degree to which the aristocratic ethos emphasized competition’ (Ober 1989: 84). Ober stresses competition for influence and honour as the fundamental mode of interaction among speakers in the law-courts, as well as in other public fora, especially the theatre (1989: 152–5).21 At the same time, Ober postulated the existence of different roles for elite speakers – namely, ‘mouthpiece’, ‘protector’, ‘advisor’ and ‘leader’ – suggesting that, at least in some of these roles, policy concerns also mattered (Ober 1989: 314–24; cf. Christ 1998: 34–6).

Our model assumes that litigants want to win (as a means of pursuing influence, honour and other social goods), but they also have specific preferences over policy/legal outcomes.22 This assumption stresses the fact that pursuing both social and policy goods might, and probably often did, require a trade-off.

3.2 Jurors’ plural preferences

The basic MVT assumes that voters have single-peaked preferences over policy issues. In other words, when presented with a binary choice (e.g. guns or butter), each voter knows which alternative he or she prefers, and casts his or her vote accordingly.

In this respect, the M&E model differs from the MVT. Because the M&E model largely neglects the role of individual preferences, mass ideology appears monolithic.23 In the previous section, we argued (p.171) that jurors’ and litigants’ preferences were meaningfully pluralistic. Our model assumes such plurality. In court, Athenian jurors were presented with a binary choice of outcomes (convict or not convict; approve or reject legislation). Based on their individual preferences over the issue – which were partially, but not uniquely, informed by their allegiances as members of the masses or of the elites – jurors voted for one or the other outcome.

4 Decision-Making in the Athenian Law-Courts

In this section, we build a new model of Athenian litigation that modifies the M&E model in two respects, based on the evidence discussed above: first, because jurors’ preferences are meaningfully pluralistic, litigants face uncertainty as to the precise position of the median juror; and second, litigants want to win as a means to gain honour, but they also have preferences over policy/legal outcomes.24 In what follows, we introduce the model through a thought experiment (a). We then present our modelling assumptions (b) and the analysis of legal process (c). We discuss the model’s implications in section 5.

4.1 A hypothetical case: Leochares v. Eratosthenes

Leochares is a shoemaker. He hails from the deme of Sphettos, southeast of Athens, but has resided in the city for a number of years. Nothing but a country boy, Leochares now owns a successful workshop, which employs a handful of slaves. He does not consider himself a politician, but he knows the ropes around the agora. Moreover, at the venerable age of forty-five, Leochares is quite versed in Athenian politics, due to his experience as a juror, as a magistrate and as an Assembly-goer.

It is the year 354 BCE. Athens has just lost the Social War and the polis’ resources are scarce. Among the topics to be discussed at the next Assembly is how to raise revenues to meet Athens’ post-war fiscal needs. The probouleuma of the Council is open, allowing citizens to formulate proposals on the floor of the Assembly.25 Leochares has an idea, which he has discussed for some time with his fellow artisans in (p.172) the agora (some of them metics, i.e. resident aliens), with citizen friends and even with his slave employees. Leochares’ idea is as follows:

suppose that, in the first place, we studied the interests of the resident aliens. For in them we have one of the very best sources of revenue, in my opinion, inasmuch as they are self-supporting and, so far from receiving payment for the many services they render to states, they contribute by paying a special tax. I think that we should study their interests sufficiently, if we relieved them of the duties that seem to impose a certain measure of disability on the resident alien without conferring any benefit on the state …26

Leochares knows that proposing any extension of privileges to metics might generate opposition. He must therefore consider his options carefully because of the risk that someone might bring a graphê paranomon against him. When the issue comes up in the Assembly, Leochares formulates his proposal:

since there are many vacant sites for houses within the walls, if the state allowed approved applicants to erect houses on these and granted them the freehold of the land, I think that we should find a larger and better class of persons desiring to live at Athens.

As soon as Leochares has finished speaking, Eratosthenes, an old, cranky mine-owner from Thorikos, indicts him by graphê paranomon. Eratosthenes has on his side a law restricting ownership of land to citizens.27 He further argues that it is utterly against the interests of the Athenian dêmos to extend privileges associated with citizenship to metics and other foreigners.28 Indeed, the proposed decree is so (p.173) iniquitous in Eratosthenes’ mind that he brings the indictment even before the Assembly has taken a vote on the proposal.29 Consideration of Leochares’ decree in the Assembly is therefore suspended and the case is transferred to the people’s courts, where a representative panel of 501 jurors will vote for Leochares or Eratosthenes by secret ballot, with majority rule and without deliberation. What kind of considerations weighed in on Leochares’ choice of proposal? And under what conditions will his proposal succeed?

4.2 Modelling assumptions

The model is based on the following assumptions:

  • (R1) The range of policy choices is represented as a single continuum (x). Because the issue presented here revolves around metic privileges, we might conceive the continuum as featuring, on one end, deportation and, on the other end, extension of full citizenship to metics.

  • (R2) Eratosthenes, Leochares and all members of the jury have symmetric, single-peaked preferences over the set of outcomes, meaning that each individual has an ideal outcome, and each prefers outcomes closer to their ideal to outcomes further away. In particular,

    • Leochares’ ideal outcome is L*, and

    • (p.174) Eratosthenes must argue for the status quo, q, so his precise preferences are irrelevant here.30

  • (R3) Leochares faces uncertainty about the exact location of the median juror’s ideal outcome, m. For simplicity, we assume that Leochares’ subjective beliefs about the location of the median are described by a symmetric, unimodal density function (e.g. a bell-shaped curve).31 Let m* be the expected value of m, according to that distribution.

4.3 Implications of the legal process for equilibrium outcomes

Based on R2, Leochares prefers his ideal policy over all other policies, but he also prefers policies closer to his ideal to those further away. Leochares’ preferences can thus be represented through a utility function, U(x). As indicated in Figure 5.1, U(x) is highest at L* (Leochares’ ideal point).

Based on R3, Leochares is uncertain about the median juror’s ideal policy. This uncertainty can be represented through a probability density function that depicts the likelihood that the median juror is located near any given point along the continuum. In Figure 5.2, the expected median of the distribution is located around m*.

If the median is located between Leochares’ ideal point and the status quo, then Leochares can win with any proposal to the left of q. However, where his proposal would fall on the continuum depends on the relative weight that Leochares attributes to honour and policy. We define ‘honour’ as a form of social capital that enables a litigant to solidify his position of leadership to the detriment of his opponent. In the model, we operationalise honour as a function of the number of votes a litigant receives. In other words, the larger the proportion of the jury that votes in favour of a litigant, the greater the honour.

Let q′ be the reflection of q around m. This means that q′ and q are equidistant from m*, but on opposite sides of m*: hence, a voter whose ideal is m* is indifferent between q′ and q; further, m* prefers all points between q′ and q to either q′ or q (because these points are closer to m* than either q′ or q).32

If Leochares cares only about policy, then he will try to push his (p.175)

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

Figure 5.1 Leochares’ utility functions

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

Figure 5.2 Probability density function

proposal as far away from q (and as close to his ideal) as possible. If Leochares cares only about policy, then, his proposal will be slightly to the right of q′, as represented in Figure 5.3.

If Leochares cares only about honour, then his proposal will be somewhere between m* and q, and probably very close to q, as in Figure 5.4.

If Leochares cares equally about honour and policy, then his proposal will be somewhere between q′ and m*, as in Figure 5.5. (p.176)

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

Figure 5.3 Leochares’ equilibrium proposal (r*): policy

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

Figure 5.4 Leochares’ L* equilibrium q′ proposal m* (r**): honour r**r**q

Rethinking Mass and Elite: Decision-Making in the Athenian Law-Courts

Figure 5.5 Leochares’ equilibrium proposal (r***): honour and policy

(p.177) 5 Implications

The model shows that the institutional design of the law-courts, paired with the procedural constraints of the graphê paranomon, encouraged the proposer of new legislation to take into account both the expected position of the median juror and the location of the status quo.33

The model yields two important results. First, the graphê paranomon performed a stabilising function by pushing proposers of new legislation close to the median. Second, the graphê paranomon enabled a proposer to move away from the status quo, thus yielding innovation in policy-making. When the median is located between Leochares’ ideal point and the status quo, Leochares has incentives to move closer to the median, and his proposal is likely to fall within a boundary created by the status quo and its reflection around the median. If the status quo and the median are very close, the boundary is small, enabling proposers to depart only marginally from both the status quo and the median’s preferences. If, instead, the status quo and the median are very far apart, the number of proposals capable of commanding a majority proportionally increases, enabling a proposer to push his proposal closer to his ideal, but also closer to the median’s preferences.34

The model yields another result. Innovation in policy-making was more likely to arise from proposers who cared more about policy than about honour. It is worth remembering that we define ‘honour’ here as a form of social capital that enables a litigant to solidify his position of leadership to the detriment of his opponent. Assuming that major politicians cared about honour more than other citizens who participated in politics only occasionally (such as Leochares), then the model shows that those who cared little about honour had the greater incentives to push new proposals as far away as possible from both the status quo and the median (compare Figures 5.3 and 5.4). The model does not suggest that policy innovation always came (p.178) from sub-elite actors. However, it does suggest that the design of the law-courts made this forum extremely receptive to diverse sources of expertise and dispersed knowledge.

The case presented above is a synchronic representation of policymaking in the law-courts. But the model also suggests a series of insights about policy change over time. In particular, the model captures two important features of the law-court environment: because juries were large and representative of the population as a whole, the median juror embodied a relatively stable ordering of preferences across cases, as well as normative expectations that were consistent with those of the average or typical member of the Athenian dêmos.35

These features have important implications for the debate on whether the Athenian legal system produced the kind of consistency and predictability that we associate with modern rule-of-law institutions (Harris 2007; 2013; contra Lanni 2004; 2009; 2016). For Harris (2013: 249–50), Athenian courts achieved consistency and predictability in the application of the law through the use of non-binding precedent, appeals to the intent of the lawgiver, and the use of analogous statutes. In contrast, for Lanni (2016: 84), ‘[i]ndividual Athenian court verdicts were the untraceable result of many individual jurors’ complicated weighing of a variety of factors, both statutory and extra-statutory’. Because individual verdicts were ad hoc, Athens’ courts did not achieve predictability and consistency in the application of the law, thus undermining the deterrence effect of statutes.

The model suggests a slightly different conclusion: because the median juror embodied a stable ordering of preferences across cases, and because verdicts were reasonably consistent with the normative expectations of the median, the Athenians did achieve some form of predictability and consistency, even if individual verdicts were not predictable. Athenian citizens did not model their behaviour on statutes. Instead, they modelled their behaviour based on what they believed the average member of their community would have considered ‘wrongdoing’.

Our model does not suggest that the Athenians always got it right. Surely, there remained plenty of room for able speakers to convince a jury to follow a risky policy, for larger-than-life personalities to carry the day or for the jurors to simply get it wrong. However, institutional design, paired with the use of penalties, enabled the Athenians to reduce the likelihood that proposals would move too far from the expectation (p.179) of the median. In fact, there were costs involved in entering the graphê paranomon game: if a prosecutor failed to gather one-fifth of the vote, he was punished. The same applied to a defendant whose proposal was successfully challenged, but only within a year from the proposal. Punishment usually took the form of fines.36 The fear of punishment thus acted as a deterrent for those who sought to advance radically novel proposals that conflicted with community preferences, or who sought to abuse the legal system to pursue frivolous litigation.37

6 Conclusion

In this chapter, we put pressure on the assumptions of the M&E model, as it applies to the process of decision-making in the law-courts. Drawing from epigraphic and prosopographical studies of political participation, and from analyses of sub-polis associations and social networks, we suggested that the identity of litigants and jurors, as well as these actors’ preferences, were meaningfully pluralistic and could not be reduced to a conflict between masses and elites. Comparing the M&E model and the MVT, we modified the former in two important respects: first, because jurors’ preferences are meaningfully pluralistic, litigants (who are not only elite) are uncertain about the exact location of the median on a given policy/legal issue; second, litigants want to win as a means to gain honour, but they also have preferences over policy/legal outcomes. Based on these assumptions, we built a new model of Athenian litigation. Our model suggests that the institutional design of the law-courts incentivised litigants to move towards the median juror’s ideal policy. More specifically, new measures were likely to fall within a boundary created by the median and the status quo. Within this boundary, the extent to which proposals diverged from the status quo depended on the litigant’s preferences. Our model reveals, perhaps unexpectedly, that those actors who cared (p.180) less about honour, who were more likely to be sub-elite Athenians, had the greatest incentives to move as far away as possible from the median. In other words, sub-elite Athenians may have been one, if not the sole, engine of legislative innovation.

Our model supports the view that political stability was a product of negotiations among community members to lower the stakes of class and other interpersonal conflicts, as Ober critically argued. However, our model emphasises that, at least when it came to law- and policymaking, those stakes had to be constantly renegotiated. In particular, our model suggests that, to understand both Athens’ stability and the polis’ success over time, we must come to terms with a much more varied and dynamic set of compromises, whose solution was not so much a winner-takes-all model (i.e. the hegemony of a monolithic mass over competing elites) as the ability to enable genuine negotiation and cooperation among citizens, while lowering the probability that these negotiations would lead to conflict.

A final note: by operationalising the findings of recent scholarship, this chapter hopes to contribute to the study of Athenian culture and society below the level of the polis and the citizen. But most importantly, by merging languages and methods from both classics and positive political science, we hope to encourage further explorations of the benefits that lie across disciplinary boundaries.


This chapter draws on our larger project on Athenian law, which includes Carugati et al. (2015; forthcoming-a; forthcoming-b). The authors gratefully acknowledge Randy Calvert and Josiah Ober for their comments on earlier versions of this chapter, as well as the participants in the 7th Leventis Conference in Greek. Among these, Paul Cartledge deserves special thanks for his helpful comments on the conference draft.


Bibliography references:

Ansolabehere, S. (2006), ‘Voters, candidates, and parties’, in B. R. Weingast and D. A. Wittman (eds), The Oxford Handbook of Political Economy, Oxford: 29–49.

Bers, V. (2009), Genos Dikanikon: Amateur and Professional Speech in the Courtrooms of Classical Athens, Washington, DC.

Black, D. (1948), ‘On the rationale of group decision-making’, Journal of Political Economy 56 (1): 23–34.

Bonner, R. J. and G. Smith (1938), The Administration of Justice from Homer to Aristotle, Chicago.

(p.181) Calvert, R. L. (1985), ‘Robustness of the multidimensional voting model: Candidate motivations, uncertainty, and convergence’, American Journal of Political Science 29: 69–95.

Canevaro, M. (2013), ‘Nomothesia in classical Athens: What sources should we believe?’, Classical Quarterly 63: 139–60.

Canevaro, M. (2015), ‘Making and changing laws in classical Athens’, in E. M. Harris and M. Canevaro (eds), The Oxford Handbook of Ancient Greek Law, Oxford <http://www.oxfordhandbooks.com/view/10.1093/oxfordhb/9780199599257.001.0001/oxfordhb-9780199599257-e-4>.

Carawan, E. (2007), ‘The trial of the Arginusai generals and the dawn of “judicial review”’, Dike 10: 19–56.

Carugati, F. (2015), In Law We Trust (Each Other): Legal Institutions, Democratic Stability and Economic Development in Classical Athens, PhD dissertation, Stanford University.

Carugati, F. (forthcoming), The Athenian Constitution: Law, Stability, and Growth in 4th Century Athens.

Carugati, F., G. Hadfield and B. R. Weingast (2015), ‘Building legal order in ancient Athens’, Journal of Legal Analysis 7: 291–324.

Carugati, F., J. Ober and B. Weingast (2016), ‘Development and political theory in classical Athens’, Polis: Journal for Ancient Greek Political Thought 33: 71–91.

Carugati, F., R. L. Calvert and B. R. Weingast (forthcoming-a), ‘Constitutional litigation in ancient Athens: Judicial review by the people themselves’.

Carugati, F., G. Hadfield and B. R. Weingast (forthcoming-b), ‘Achieving inclusion: Extending access to the rule of law in ancient Athens’.

Christ, M. R. (1998), The Litigious Athenian, Baltimore.

Cloché, P. (1960), ‘Les hommes politiques et la justice populaire dans l’Athènes du IVe siècle’, Historia: Zeitschrift für Alte Geschichte 9: 80–5.

Cohen, D. (1995), Law, Violence, and Community in Classical Athens, Cambridge.

Cohen, E. (1973), Ancient Athenian Maritime Courts, Princeton.

Cohen, E. (1992), Athenian Economy and Society: A Banking Perspective, Princeton.

Davies, J. K. (1971), Athenian Propertied Families, 600–300 B. C., Oxford.

Davies, J. K. (1981), Wealth and the Power of Wealth in Classical Athens, reprints edn, Salem.

Dillery J. (1993), ‘Xenophon’s Poroi and Athenian imperialism’, Historia: Zeitschrift fur Alte Geschichte 42: 1–11.

Downs, A. (1957), An Economic Theory of Democracy, New York.

Goodell, T. D. (1893–4), ‘An Athenian parallel to a function of our Supreme Court’, Yale Review 2: 64–73.

Goodwin, W. W. (1895), ‘The Athenian graphe paranomon and the American doctrine of constitutional law’, Transactions of the American Philological Association 26: lx–lxi.

Gowder, P. (2014), ‘Democracy, solidarity, and the rule of law: Lessons from Athens’, Buffalo Law Review 62: 1–67.

(p.182) Hansen, M. H. (1974), The Sovereignty of the People’s Courts in Athens in the Fourth Century BC and the Public Action against Unconstitutional Proposals, Odense.

Hansen, M. H. (1984), ‘The number of rhetores in the Athenian ecclesia, 355–322 bc’, Greek, Roman, and Byzantine Studies 25: 123–55.

Hansen, M. H. (1987), ‘Graphe paranomon against psephismata not yet passed by the ekklesia’, Classica et Mediaevalia 38: 63–73.

Hansen, M. H. (1990), ‘Review article: Ober, J. 1989 Mass and Elite in Democratic Athens’, The Classical Review 40 (2): 348–56.

Hansen, M. H. (1999), The Athenian Democracy in the Age of Demosthenes, Norman OK.

Harris, E. M. (2007), ‘Did the Athenian courts attempt to achieve consistency? Oral tradition and written records in the Athenian administration of justice’, in C. Cooper (ed.), Politics of Orality, Leiden: 343–70.

Harris, E. M. (2013), The Rule of Law in Action in Democratic Athens, Oxford.

Herman, G. (1995), ‘Honour, revenge and the state in fourth-century Athens’, in W. Eder and C. Auffarth (eds), Die athenische Demokratie im 4. Jahrhundert v. Chr.: Vollendung oder Verfall einer Verfassungsform? Akten eines Symposiums 3.-7. August 1992, Bellagio, Stuttgart: 43–60.

Herman, G. (1996), ‘Ancient Athens and the values of Mediterranean society’, Mediterranean Historical Review 11: 5–36.

Ismard, P. (2010), La cité des réseaux: Athènes et ses associations, VIe–Ier siècle av. J.-C., Paris.

Johnstone, S. (1999), Disputes and Democracy: The Consequences of Litigation in Ancient Athens, Austin.

Jones, A. H. M. (1957), Athenian Democracy, Oxford.

Jones, N. F. (1999), The Associations of Classical Athens: The Response to Democracy, New York.

Kamen, D. (2013), Status in Classical Athens, Princeton.

Kierstead, J. C. (2013), A Community of Communities: Associations and Democracy in Classical Athens, PhD dissertation, Stanford University.

Kroll, J. H. (1972), Athenian Bronze Allotment Plates, Cambridge.

Kron, G. (2011), ‘The distribution of wealth in Athens in comparative perspective’, Zeitschrift für Papyrologie und Epigraphic 179: 129–38.

Lagia, A. (2015), ‘Diet and the polis: An isotopic study of diet in Athens and Laurion during the classical, Hellenistic, and imperial Roman periods’, in A. Papathanasiou, M. P. Richards and S. C. Fox (eds), Archaeodiet in the Greek World: Dietary Reconstruction from Stable Isotope Analysis, Hesperia Supplement 49: 119–46.

Lanni, A. (2004), ‘Arguing from “precedent”: Modern perspectives on Athenian practice’, in E. Harris and L. Rubenstein (eds), The Law and the Courts in Ancient Greece, London: 159–72.

Lanni, A. (2006), Law and Justice in the Courts of Classical Athens, Cambridge.

Lanni, A. (2009), ‘Social norms in the courts of ancient Athens’, Journal of Legal Analysis 1 (2): 691–736.

Lanni, A. (2010), ‘Judicial review and the Athenian “constitution”’, Harvard Law School Public Law & Legal Theory Working Paper Series, No. 10–21.

(p.183) Lanni, A. (2016), Law and Order in Ancient Athens, New York.

Markle, M. M. (1985), ‘Jury pay and Assembly pay at Athens’, in P. Cartledge and F. Harvey (eds), Crux, London: 265–97.

Morris, I. (2004), ‘Economic growth in ancient Greece’, Journal of Institutional and Theoretical Economics 160: 709–42.

Ober, J. (1989), Mass and Elite in Democratic Athens, Princeton.

Ober, J. (2008), Democracy and Knowledge, Princeton.

Ober, J. (2017), ‘Inequality in late-classical democratic Athens: Evidence and models’, in G. C. Bitros and N. C. Kyriazis (eds), Democracy and Open Economy World Order, New York: 125–46.

Osborne, R. (1985), Dêmos: The Discovery of Classical Attika, Cambridge.

Rhodes, P. J. (1998), ‘Enmity in fourth century Athens’, in P. Cartledge, P. Millett and S. von Reden (eds), Kosmos: Essays in Order, Conflict, and Community in Classical Athens, Cambridge: 162–9.

Schwartzberg, M. (2013), ‘Was the graphe paranomon a form of judicial review?’, Cardozo Law Review 34: 1049–62.

Sundahl, M. J. (2000), The Use of Statutes in the Seven Extant Graphe Paranomon and Graphe Nomon Me Epitedeion Theinai Speeches, PhD dissertation, Brown University.

Sundahl, M. J. (2003), ‘The rule of law and the nature of the fourth-century Athenian democracy’, Classica et Mediaevalia 54: 127–56.

Taylor, C. (2007a), ‘A new political world’, in R. Osborne (ed.), Debating the Athenian Cultural Revolution: Art, Literature, Philosophy, and Politics 430–380 BC, Cambridge: 72–90.

Taylor, C. (2007b), ‘From the whole citizen body: The sociology of election and lot in the Athenian democracy’, Hesperia 76: 323–45.

Taylor, C. (2015), ‘Social networks and social mobility’, in C. Taylor and K. Vlassopoulos (eds), Communities and Networks in the Ancient Greek World, Oxford: 35–53.

Taylor, C. and K. Vlassopoulos (2015), ‘Introduction: An agenda for the study of Greek history’, in C. Taylor and K. Vlassopoulos (eds), Communities and Networks in the Ancient Greek World, Oxford: 1–31.

Todd, S. C. (1990), ‘The purpose of evidence in Athenian courts’, in P. Cartledge, P. Millet and S. Todd (eds), Nomos: Essays in Athenian Law, Politics, and Society, Cambridge: 19–39.

Todd, S. C. (1993), The Shape of Athenian Law, Oxford.

Vlassopoulos, K. (2007), ‘Beyond and below the polis: Networks, associations, and the writing of Greek history’, Mediterranean Historical Review 22: 11–22.

Whitehead, D. (1986), The Demes of Attica, 508/7–ca. 250 B. C.: A Political and Social Study, Princeton.

Wittman, D. (1983), ‘Candidate motivation: A synthesis of alternatives’, American Political Science Review 77: 142–57.

Wolff, H. J. (1970), ‘Normenkontrolle’ und Gesetzesbegriff in der attischen Demokratie, Heidelberg.

Yunis, H. (1988), ‘Law, politics, and the graphe paranomon in fourth-century Athens’, Greek, Roman, and Byzantine Studies 29: 361–82.


(1) Ober 1989: 132–55 sees the dynamics in the law-courts similarly at play in the Assembly, the boulê, the Areopagos, the street and the theatre. Johnstone 1999: 12 criticised this methodology, suggesting that there are marked differences between the law-courts and other public fora.

(2) These assumptions similarly apply to other models of Athenian litigation that elaborated on Ober’s account – most notably, those of David Cohen, Gabriel Herman and Matthew Christ. David Cohen argued that the dynamic between elite litigants mimicked a type of feuding behaviour that revolved around the pursuit of honour and revenge typical of Mediterranean societies (Cohen 1995). Herman (1995; 1996) challenged Cohen’s pessimistic view of Athenian litigation as a zero-sum game among elites and stressed instead the role of cooperative attitudes, such as self-restraint and non-retaliation. Christ (1998) rejected either extreme and emphasised how Cohen’s and Herman’s opposing views were the product of the ambivalence the Athenians felt towards litigation.

(3) Nomothesia was a rather complicated procedure for approving new laws and rejecting old ones: Canevaro (2013: 150; 2015) is the most recent account. The graphê nomon mê epitêdeion theinai was a public action against an unsuitable law. The graphê paranomon was its sister procedure: a public action against an unconstitutional decree (both procedures are described in Hansen 1999: 205–12). We discuss the graphê paranomon in more detail in section 4.

(4) The model that we present below illustrates a case of graphê paranomon. However, the model captures the dynamics at play in all cases where one of the litigants advanced a new legislative proposal and the other litigant challenged it, thus de facto advocating for the status quo. These include the graphê nomon mê epitêdeion theinai and the repeal stage in the process of nomothesia. The repeal stage, where existing laws that conflicted with a new proposal had to be discussed and repealed, took place in a law-court and therefore fits the assumptions of the model (Dem 20.93; Canevaro 2015: 18). The inconsistencies about the nature of the last stage of nomothesia, where new legislation was approved by the nomothetai, make it difficult to determine a possible fit between the procedure and the model. We therefore prefer to suspend judgement.

(5) Early challenges to the political sociology of Ober’s M&E model include Hansen 1990 and Johnstone 1999. For an account of these positions cf. Carugati 2015: ch. 4.

(6) The 2,000 cases per year is a conservative number based on Hansen’s calculations of the volume of litigation (Hansen 1999: 186–8). Multiplying 2,000 cases by 80 years yields a total number of 160,000 cases. Because each case must have featured at least two speeches, the extant evidence actually amounts to less than 0.05 percent of the total.

(7) Cohen 1995: 193–4; Christ 1998: 32–3; Johnstone 1999: 19; Lanni 2006: 34 (acknowledging overrepresentation in the orators); Kamen 2013: 100; Gowder 2014.

(8) Of the three components of Ober’s tripartite definition of the elite (1989: 11–17), the evidence below tracks wealth and, to some extent, birth, but not education.

(9) The argument so far leaves out the possibility that whereas a large percentage of Athenians proposed legislation in the Assembly, the sub-elite unmistakably proposed legislation that was unlikely to be important enough to be challenged and end up in the courts. However, in Hansen’s list, of proposers whose names begin with the letter A (which we take to be not correlated with their wealth) only three (or maybe five) out of fifteen can be found in Davies 1971’s list of Athenian propertied families. Moreover, sub-elite proposers are recorded as proposing a number of critical decrees, ranging from domestic to foreign policy.

(10) Taylor discusses and ultimately dismisses the role of migration: if migration surely occurred from the coastal and inland areas to Athens, it is hard to see why it would only concern elite, as opposed to sub-elite, citizens. Conversely, it seems more likely that poorer (particularly landless) individuals would try their luck in the big city, most notably in search of jobs, including public service. Moreover, as Taylor suggests, ‘it is very difficult to assess whether this movement was widespread or occasional, [or] whether it was permanent or temporary’. Cf. Taylor 2007b: 335; 339 and n. 80.

(11) For most public cases, the normal jury size was 501. Private cases were usually heard by panels of 201 jurors if the sum at issue was less than 1,000 drachmas, and by 401 jurors if the sum at issue was larger. On juries’ size, see Hansen 1999: 186–7.

(12) For Jones 1957, jurors were middle class; for Markle 1985 and Hansen 1999 jurors were poor; for Todd 1990, jurors were neither poor nor rich, but farmers. For an overview of these positions see Hansen 1999: 184–5.

(13) On the distribution of wealthy Athenians in demes see Kierstead and Klapaukh in this volume.

(14) Translation adapted from H. Rackham; see also Kierstead 2013.

(15) The results presented in this chapter do not apply to those courts that featured the participation of non-citizen actors: for example, the dikai emporikai, or cases heard by the court of the polemarchos involving metics. We discuss non-citizen participation in the law-courts in Carugati et al. (forthcoming-b).

(16) For an analysis of social networks in the polis’ central institutions: Ober 2008. Much work on sub-polis networks focuses on the deme. The relationship between citizens and non-citizens is touched on, but the focus remains largely on citizens. For Osborne 1985, demes are microcosms of the polis, and their institutions are but a mirror of the central organisation. For Jones 1999, demes are a response to the existence of geographic and legal barriers to participation in the central institutions. For Whitehead 1986: 258, 97–108, demes and polis are ‘intricately and deliberately interlocked with each other’, as emerges, most clearly, from the workings of citizenship procedures. In Kierstead 2013: ch. 3’s recent formulation, the contribution of the demes to the polis relies less on the existence of formal institutional procedures and more on the ability of the demes to create social capital by fostering solidarity and information-gathering among its members.

(17) Ismard studies sanctuaries (2010: 209–11); regional associations (2010: 211–24) and the cultic systems of the Salaminian genos (2010: 224–39) and of the Tetrapolis of Marathon (2010: 239–49). The in-depth investigation of cultic and financial associations leads Ismard to question the notion that the city-state was the main agent of historical development: in fact, for Ismard, the civic aspects of cultic and economic life are but a few of the many aspects of the radically dynamic and fluid networks of associations.

(18) For a detailed overview of the origins and development of the MVT see Ansolabehere 2006’s survey.

(19) This does not mean that Athenian actors were fully informed – no actor in a complex institutional setting ever is. In fact, as we will see in the next section, our model assumes that if jurors have complete information about the position of the litigants, litigants are uncertain about the precise position of the median juror.

(20) Calvert 1985 modifies the conclusions reached by Wittman 1983.

(21) Cohen 1995: 9 takes the argument to an extreme, suggesting that litigants in court have no other goal than ‘to pursue and intensify their antagonism’.

(22) The parallel debate, among scholars of Athenian law, about whether political or legal arguments prevailed in litigation (especially in graphê paranomon cases), suggests that both types of argument played an important role: for the primary role of legal arguments see Goodell 1893–4; Goodwin 1895; Bonner and Smith 1938; Wolff 1970; for political arguments see Cloché 1960; for both arguments see Hansen 1987; 1999; Yunis 1988; Sundahl 2000; 2003; Carawan 2007; Lanni 2010; Schwartzberg 2013.

(23) We should be clear that we do not intend this point as a criticism of Ober’s model. His account (1989: 297–8) discusses the perils of assuming ‘same-mindedness’ (homonoia) among the demos. However, Ober’s model simply neglects the role of individual preferences, particularly preferences related to policy (as opposed to preferences concerning the characteristics of the good orator). These preferences are instead key to our model of decision-making in the courts.

(24) Our model of Athenian litigation draws on Calvert 1985’s modification of the MVT. More technical developments in the literature on the MVT, such as discussions of the role of political parties and interest groups, would be inapplicable to Athens.

(25) For the distinction between open and closed probouleumata see Hansen 1999: 138–9.

(26) Note that the passage (and its continuation below) is a direct citation from Xenophon’s pamphlet Poroi, dated to the middle of the fourth century (Xen. Por. 2.1–7; transl. Loeb). In the Poroi, Xenophon listed a series of measures aimed at raising state revenues by boosting indirect taxation in two main areas of the Athenian economy: maritime commerce and silver-mining. As such, the proposal we present here ought to be considered as representing ideas that may well have been under the Assembly’s consideration. On Xenophon’s Poroi see Dillery 1993; Carugati et al. 2016.

(27) We lack direct evidence for a law that restricted land ownership to citizens. However, the principle seems incontrovertible. As Todd 1993: 243 put it, ‘the fundamental rule at Athens was that ownership of land … was in principle restricted to citizens, and this right was only in exceptional circumstances extended to metics by the special privilege of enktêsis’.

(28) Note that the decree would not grant citizenship to metics, but it would extend to them benefits associated with citizenship, such as the right to own land. This scenario must be somewhat similar to the one that instituted the commercial cases (dikai emporikai) in the middle of the fourth century, which extended to some categories of non-citizens the privilege to participate as litigants in Athenian courts. On the extension of inclusion through the dikai emporikai see Carugati et al. (forthcoming-b). On the dikai emporikai more generally see Cohen 1973; Lanni 2006.

(29) A graphê paranomon could be brought (1) before the Assembly voted on a new proposal; (2) after the Assembly’s vote; or (3) after the proposal became a law (Hansen 1974; 1987; on the existence of a third stage, after the enactment of the decree, see Carawan 2007). For the purposes of this chapter, the distinction is irrelevant. Strategically, it is a risky move to expose oneself to the potential costs associated with the procedure before feeling the Assembly’s pulse. Nonetheless, we have direct evidence of such a practice (Hansen 1987). Moreover, other considerations may have offset the risk of indicting a proposal immediately after it was presented to the Assembly: an immediate indictment is a powerful signalling device, suggesting that someone feels strongly about the detrimental effect of said proposal – so strongly, indeed, that he is willing to forfeit the benefits of the vote. For simplicity, and in order to focus on the graphê paranomon and cognate cases, we assume that the median is the same in the Assembly and in the courts. Future work will relax this assumption, providing a more accurate sequential game of the entire process of Athenian policy-making.

(30) More specifically, if he brings a graphê paranomon, Eratosthenes must prefer q to L*, but his position is fixed at q, so his precise preferences are irrelevant.

(31) A unimodal density function has one peak (i.e. there is only one area in which the probability that the median is located there is high relative to the probability in other areas).

(32) By the same token, any policy to the left of q′ or to the right of q would have a low probability of winning.

(33) Note that the multi-step legislative process imposed a baseline constraint on proposals. For example, if Leochares were to propose to raise state revenues by eliminating all compensation for public service, such a proposal would be unlikely to end up in court in the first place, and more likely to be rejected (shouted down) in the Assembly. The range of proposals that could reach the courts was therefore limited.

(34) Similarly, when Leochares’ ideal point is located between the median and the status quo, he has strong incentives to enter the graphê paranomon game: he can win by proposing his ideal point, and establish a new status quo that is closer to the median’s preferences than the old status quo. Conversely, when the status quo is located between Leochares’ ideal and the median, he has no incentives to propose legislation in the first place.

(35) Simple comparative statistics results suggest that, as the size of the jury grows, the variance in the distribution of preferences falls, so that the median in the courts converges on the median of the whole citizenry. This result suggests that changing the size of the jury based on the relevance of the issue was a strategy to increase the likelihood that verdicts would align with community expectations.

(36) Fines ranged from relatively small sums, such as 25 drachmas, to very large ones (usually to the amount of 1,000 drachmas, but we also have evidence for higher sums). If unable to pay, the guilty individual was liable to incur temporary atimia (disenfranchisement) until the debt was paid. Permanent atimia, instead, awaited a proposer against whom three judgements were made (Hansen 1974: 28–43). Because of these costs, classical scholars have often assumed that only members of the Athenian elite could afford to enter the graphê paranomon game. But there were also reputational costs that affected elite actors. The pecuniary costs are greater for a sub-elite Athenian, but not as great as to overpower the reputational cost for an elite actor.

(37) Elsewhere, we discuss the use of coordination devices to police proposals that moved too far from the median, such as the deployment of the figure of Solon: see Carugati 2015: ch. 4; (forthcoming: ch. 3); Carugati et al. (forthcoming-a).