The rule of law as the measure of political legitimacy in the Greek city states

This paper explores how a conception of the rule of law (embodied in a variety of legal and political institutions) came to affirm itself in the world of the ancient Greek city states. It argues that such a conception, formulated in opposition to the arbitrary rule of man, was to a large extent consistent with modern ideas of the rule of law as a constraint to political power, and to their Fullerian requirements of formal legality, as well as to requirements of due process. The article then analyses how this ideal was formulated in the Archaic period, and how it became a key feature of Greek identity. Finally, it argues that in the fifth and fourth centuries BCE it came to be used as the measure of the legitimacy of Greek political systems: democracy and oligarchy, as they engaged in an ideological battle, were judged as legitimate (and desirable) or illegitimate (and undesirable) on the basis of their conformity with a shared ideal of the rule of law. Then as now, to quote Tamanaha, ‘the rule of law’ was ‘an accepted measure worldwide of government legitimacy’.


Introduction
The starting point of Tamanaha's (2004, p. 3) much cited book-length study on the rule of law is the recognition that 'the rule of law is an accepted measure worldwide of government legitimacy'. He remarks on the 'global endorsement' of the rule of law, far beyond any 'other single political ideal'. This global endorsement of the concept goes hand in hand with the belief that the rule of law produces a variety of social and economic goods, from human welfare to political stability and economic growth, and is therefore the key precondition for the development of a just and prosperous society. 1 And, because of this, scholars and governmental agencies alike use a variety of aggregate measures to test whether this or that state conforms to (some of) the conditions of the rule of law. Møller and Skaaning (2014) provide an extensive discussion of the many different (and occasionally incompatible) concepts of the rule of law presupposed by such aggregate measures, and attempt to bring some conceptual clarity to the field. But, regardless of the conceptual sophistication (or lack thereof) of these measures and debates, the bottom line is that, to return to Tamanaha, '[t]he rule of law is a major source of legitimation for governments in the modern world. A government that abides by the rule of law is seen as good and worthy of respect'-the rule of law (however it is measured and conceptualized) is held as the measure by which the legitimacy of a state, government or constitution should be judged. 2 Tamanaha, in the same context, also observes that '[u]nanimity in support of the rule of law is a feat unparalleled in history'. The emergence of such a notion (and of the related institutions) has been explained by investigating a distinctive and (for some) historically unique historical process arching back to Western and Central Europe in the Middle Ages. 3 The argument is often made that the rule of law as it manifests itself in the modern political traditions of the West has emerged out of a distinctive historical process the roots of which are to be found in the Middle Ages. Yet claims that the end results of such a development-ideas (and institutions) of the rule of law as opposed to the rule of man that entail, at the very least, Fullerian notions of formal legality and the requirement that government officials are limited in their actions by the law (whatever the definitional disagreements)-are in any way unique, immediately sound problematic to the historian of ancient Greek law, politics and political thought. One needs only to turn to Aristotle, who in the fourth century BCE, in the Politics, reports the following as an opinion widely held (and that he himself holds with some qualification): [S]ome people think that it is not according to nature for one person to have authority over all the citizens, where the city-state is established out of similar persons. For persons who are similar by nature necessarily have the same right and the same merit according to nature. […] Consequently, it is just to rule no more than to be ruled, and it is just [to rule and be ruled] by turns. But this is already law; for law is the order [taxis] [by which offices are shared]. Hence the rule of law is preferable to that of a single citizen. (Arist. Pol. 1287a10-14, 16-20;trans. Miller) 4 1 Similar considerations are the starting point also of the Møller's article in this same issue. Cf. Belton (2005: 5), Carothers (2006), Haggard et al. (2008), Møller and Skaaning (2014). 2 Tamanaha (2012: 232): 'the notion of the rule of law is perhaps the most powerful and often repeated political ideal in contemporary global discourse. Everyone, it seems, is for the rule of law. The rule of law is a major source of legitimation for governments in the modern world.' 3 E.g. Tamanaha (2004: 15-31), Jones (2008), Blaydes and Chaney (2013), Fukuyama (2011Fukuyama ( , 2014. See Møller in this issue for a sophisticated reconstruction of the historical process that, from the Middle Ages, has led to the modern notion (and institutions) of the rule of law. 4 For a comprehensive and accessible discussions of Aristotle's ideas on law and the rule of law see Miller (2007) and Bertelli (2018). All the abbreviations used for the works of ancient Greek authors are those of the Oxford Classical Dictionary.
That this is indeed a widely-held notion is confirmed even by a summary look at the speeches of the Attic orators, where we find a huge number of passages with variations on the same theme. One example among hundreds-in 322 BCE Hyperides declares: There cannot be complete happiness without independence. For men to be happy they must be ruled by the voice of law, not the threats of a man; free men must not be frightened by accusation, only by proof of guilt; and the safety of our citizens must not depend on men who flatter their masters and slander our citizens but on our confidence in the law. (Hyp. Epit. 25; tr. Cooper).
If we move outside Athens, to the very Panhellenic contingent of the Greek expedition to Asia narrated by Xenophon in his Anabasis, we find the same notions at play. In book 5 (5.7-8) the mercenary army of the Ten Thousand, composed by Greeks from all parts of Greece, sees discipline starting to break down, to the extent that one of the agoranomoi (an official) is attacked by the mob of the soldiers. The agoranomos flees, and Xenophon reproaches the soldiers for their behaviour, and for threatening to kill a man without trial. As a result, the assembly (agora) of the soldiers votes that anyone who instigates such behaviour in the future will be put to death after a trial; that the generals will bring the accused to trial; that the company commanders will be the judges; that the generals themselves will be put to trial for their conduct (and some of them are fined for various offences). Xenophon himself is accused of hybris and defends himself in a trial. The incident is evidence that there was widespread agreement among the Greeks that officials should behave in accordance with the laws, and be tried if they do not, but at the same time that everyone has a right to due process, and no one should be put to death without trial. The soldiers in the expedition came from many different Greek communities-Athens, Boeotia, Stymphalus Sparta, Thessaly, Achaea, Olynthus and Arcadia-yet they all recognized these basic principles. 5 Such passages clearly show that a notion of the rule of law as opposed to the rule of man (and to anarchy) was widely understood and endorsed in the world of the Greek poleis, and that it entailed curtailing the arbitrary power of government and officials and securing due process for the accused whose guilt had to be proven in accordance with the law. 6 These notions are not only found in the reflections of philosophers and orators, and in the narratives of the historians, but are enshrined in the official oaths of the city. The Athenian Ephebic oath, an oath that young Athenians citizens eighteen years of age pronounced on entering Ephebic service (something in-between military training and civic education), had them swear: I shall be obedient to whoever exercise power reasonably on any occasion and to the laws currently in force and any reasonably put into force in future. If anyone destroys these I shall not give them allegiance both as far as is in my own power and in union with all (tr. Rhodes and Osborne). 7 Even more significant in this respect is the Judicial Oath, sworn in Athens at the beginning of each year by the 6000 Athenians citizens selected by lottery to serve as judges in the popular lawcourts for the year-a document for the expectations of the Athenians about the role of their lawcourts, and for the values that underpinned these institutions. We find 137 references to this oath in the speeches of the orators pronounced during trials in the lawcourts, 8 and from these references we can reconstruct four pledges: 1. to vote in accordance with the laws and the decrees of the Athenians (e.g. Aeschin. 3.6; Antiph. These four pledges, together, give a rather precise idea of what the Athenians expected from the judges. The first pledge, by far the most widely quoted in the orators, binds the judges to decide cases on the basis of the laws-their job was to assess whether the facts of the case matched the operative facts illustrated by the relevant law, and only in that case to inflict the normative consequences prescribed by the law. 9 The second pledge expresses a commitment to providing both parties with an equal hearing and equal opportunities. The third pledge binds the judges not to take into account enmity and extra-legal matters, and provides guidance in those cases in which the letter of a statute is not conclusive in guiding their decision: the judges must in these cases vote honestly and side with the party that offers the best legal argument. 10 The fourth pledge is connected to the second and is key in that it 7 The oath is preserved in a fourth-century inscription from the deme of Acharnae (RO 88), and in slightly modernised versions in . For good discussions of the oath see Siewert (1972Siewert ( , 1977, Rhodes and Osborne (2003: 440-449), Sommerstein and Bayliss (2012: 13-22). 8 See Harris (2013: 351-356) for a list of the occurrences. A text purporting to be the actual oath is preserved at Dem. 24.149-51, but it is a pastiche composed by a later forger on the basis of quotes in the orators and of some guesswork, see Canevaro (2013: 173-180) (Sommerstein and Bayliss 2012: 69-80 are marginally more optimistic on the reliability of some of its clauses, but ultimately agree that it is a later pastiche). 9 The terminology here is that of MacCormick (2004: 12-77). For laws often (but not always) formulated in casuistic form in the ancient Greek world see Carey (1998) and Harris (2013: 12-77). 10 The interpretation of the gnome dikaiotate (found only in four passages of the orators: Dem. 20.118,57.63; see also Arist. Pol. 1287a26) has been the focus of much disagreement among scholars. Some scholars (e.g. Todd 1993: 54;Christ 1998: 201-203;Too 2008: 107;Lanni 2006: 72;Forsdyke (2018) read this reference as providing a separate, alternative standard of judgement (justice as an alternative to the laws). But gnome dikaiotate comes into play only when 'there are no laws ' (cf. Harris 2013: 104-105 and Sommerstein and Bayliss 2012: 76-77, pace Mirhady 2007. It is not an alternative standard of judgement that excludes the laws-the authority of the laws is paramount, and the gnome shows that legal procedures were designed to guarantee a fair trial and equal opportunities to both parties. It effectively bound the judges to vote exclusively on the charge, and not on separate matters and on the basis of other considerations. 11 Athenian conceptions were not limited to notions of the rule of law as opposed to that of man, of the need to curtail the power of officials, and of securing due process. They created further requirements that are consistent with Fullerian notions of formal legality. 12 Fuller identified as key requirements of the rule of law: (1) generality of laws; (2) promulgation; (3) no retroactive laws; (4) clarity of laws; (5) no contradictions in the laws; (6) laws do not require the impossible; (7) relative constancy of laws through time; (8) congruence between official action and declared rule. We have discussed the requirements of the Judicial Oath, which bind the judges to pass judgements in accordance with the existing laws, as well as notions of the accountability of officials under the law (see also below). These features match Fuller's requirement of congruence between official action and declared rule (8). The Athenians at the end of the fifth-century, following oligarchic coups and revolutions, enacted rules that defined clearly what the requirements were for enacting a law, 13 and these rules match further requirements of the Fullerian notion of formal legality. These laws are discussed by Andocides (1.85-9). One of them states that 'it is not allowed for magistrates to use an unwritten (agraphos) nomos not even about a single matter'. This is a basic rule to be respected in the city: magistrates should perform their task adhering strictly to the instructions of the written laws, and not according to customs, or to any principle that is not enshrined in the laws of the city. This law matches Fuller's requirement for congruence between official action and declared rule (8). Another law preserved in Andocides states that 'no decree, neither of the Council nor of the Assembly, is to have more authority than a law'. This law for the first time introduces a clear-cut distinction between nomoi and psephismata, and a hierarchy between the two, with nomoi (laws) being rules of a higher level, which can overrule decrees but cannot be overruled by them. 14 Because of this law, decisions of the Assembly or of the Council cannot contradict the existing laws, and a certain level of clarity and coherence of the laws is guaranteed. This matches Fuller's requirements (4) and (5).
Footnote 10 continued dikaiotate of the judges comes into play only when the guidance offered by the laws is less than unequivocal (cf. Harris 2013: 104-114;Sommerstein and Bayliss 2012: 76-77;see Pelloso (2018) for the meaning 'the best legal argument'). Note also that references to the gnome dikaiotate appear only four times in almost one hundred speeches of the Attic orators, and in all cases they do not introduce arguments that contradict the written law, see MacDowell (2009: 76 n. 48) and Harris (2013: 104-105). 11 See Harris (2013: 114-136) and Thür (2008: 66-69). For accessible accounts of Athenian legal procedure and legal argument along the lines sketched here, see Canevaro (2018a, b). 12 Fuller (1969. Cf. the very similar accounts of Raz (1979: 14-18) andFinnis (1980: 170-171), and the discussions of various minimalist (as well as thicker and substantive) definitions provided e.g. in Tamanaha (2004: 91-113) and Møller-Skaaning (2014: 13-27). 13 For an account of this process, and of its roots in the earlier nomothetic tradition as well as in democratic developments, see Canevaro (2015), with previous bibliography. The content of these rules matches quite nicely the idea of 'rules of recognition' discussed in Hart (1994: 94). 14 Two pioneering essays by Hansen (1978Hansen ( , 1979 have demonstrated that the Athenians throughout the fourth century respected this subdivision and this hierarchy carefully. Cf. Canevaro (2015). The next law quoted by Andocides states (although in negative terms) that laws (nomoi) must be general in their formulation (Fuller's requirement [1]): 'It is not permitted to enact a law directed against an individual unless the same law applies to all Athenians'. 15 The Athenians defined what a nomos was by forbidding laws for single individuals: laws must have a general content and apply to all Athenians alike. Another law that is likely to belong to the same context is the law of Diocles, which states that any new law is to be valid from the day of its enactment (unless the law itself provides for a later starting point). 16 Thus this law provides that laws cannot be retroactive, which matches Fuller's requirement that there should be no retroactive laws (3). These rules were complemented by the creation, in the same context, of comprehensive legislative procedures-rules of change-whose purpose was to secure the coherence (Fuller's requirement [5]) and the clarity (Fuller's requirement [4]) of the laws of the city, while at the same time guaranteeing their stability (Fuller's requirement [7]). 17 These rules also required that law proposals should be posted and publicised extensively, and all Athenian laws were freely accessible in the public archives as well as, often, as inscriptions in public places (Fuller's requirement [2]: promulgation). Formal legality was explicitly understood as functional to legal certainty-in Demosthenes' words: Opposing laws are repealed so that there is one law for each subject. This avoids confusion for private individuals, who would be at a disadvantage in comparison to people who are familiar with all the laws. The aim is to make points of law the same for all to read as well as simple and clear to understand. (Dem. 20.93;tr. Harris) The evidence I have presented so far should suffice to show that the Athenians (and, according to Aristotle, the Greeks more generally) were aware of a concept of the rule of law that is recognizable to us, and consistent with many of the features that we normally attribute to the rule of law. It also shows that they enshrined it in their legal institutions, in the relevant rules as well as in oaths and in the rhetorical commentary to the workings of these institutions. It would be misleading however to suggest that Greek historians unanimously accept that the Athenians consistently achieved and practiced the rule of law-this is one area in which the last thirty years have seen a heated debate. While some scholars, Edward Harris in particular, as well as (with different approaches and nuances) Martin Ostwald, Raphael Sealey, Lene Rubinstein, Gabriel Herman, P.J. Rhodes, James Sickinger, Carlo Pelloso, Paul Gowder and myself, have argued that by and large the Athenian legal system conformed both to emic and to modern notions of the rule of law (and that the institutions and rules described above, by-and-large, worked!), others, such as Robin Osborne, Josiah Ober, David Cohen, Stephen Todd, Mathew Christ, and Adriaan Lanni have faulted it for (allegedly) privileging notions of personal standing, vengeance and extra-legal considerations of justice as opposed to lawfulness. 18 Many of these treatments rely (much as many of the aggregated measures applied to modern states do) on varying and incompatible definitions of the rule of law, and too often, in my opinion, the criteria used to deny that Athens achieved the rule of law would see most modern Western liberal societies fall short as well. Many of the skeptics require of the Athenians standards of legal certainty that are not only unrealistic for any society, but also theoretically problematic and potentially undesirable, as they would conflict with other staples of the rule of law such as due process and the procedural requirements that parties be allowed to argue anything (concerning matters of fact as well as matters of law) that can be possibly argued. 19 Apart from these brief remarks, it is not my aim in this article to argue that Athenian practice fulfilled modern requirements of the rule of law-I have fought this battle elsewhere, and I plan to fight it again in the future. For the purpose of this article, it is enough for me to have shown that the Athenians (and, as we shall see, the Greeks more widely) entertained notions of the rule of law, as found in their political ideas, in their laws and institutions, that are recognizable to us, regardless of whether the relevant institutions succeeded in achieving this ideal to an extent that would satisfy us (and our aggregated measures). I believe that very few (if any) ancient historians would deny this. 20 My aim is rather to argue that this notion of the rule of law was not just an ideal among many. In the world of the ancient Greek city states, these notions became, from the Archaic period onwards, a normative ideal that not only permeated legal and political thought and shaped the laws and institutions of individual cities across a variety of constitutional arrangements, but also by reference to which the different constitutional arrangements were judged. This amounted to a veritable consensus, across an ecosystem of over 1500 city states, 21 that a well-governed city is one in which laws, not men, are sovereign. The all-pervasiveness of this consensus is enough to make one question the accuracy of Tamanaha's statement that '[u]nanimity in support of the rule of law is a feat unparalleled in history'. In the rest of this article I shall explore how a very similar unanimity served as the baseline-the measure-of political and ideological debates 18 Harris (2006Harris ( , 2013, Ostwald (1986), Sealey (1987), Rubinstein (2000Rubinstein ( , 2007, Rhodes (2004), Herman (2006), Sickinger (2008), Gowder (2014), Canevaro (2013Canevaro ( , 2015Canevaro ( , 2018a, Pelloso (2018), Osborne (1985), Ober (1989), Todd (1993), Cohen (1995, Christ (1998), Lanni (2006. For studies that approach related problems from the point of view of New Institutional Economics, rational choice theory and game theory, see Carugati (2014), Carugati, Hadfield and Weingast (2015), Carugati, Calvert and Weingast (2016). Unlike Hansen and Ostwald, Harris does not believe that the concern with the rule of law was an innovation of the fourth century BCE, to temper the 'extreme' democracy of the fifth, see in particular Harris (2016). 19 See, on the problems of excessive requirements of legal certainty, MacCormick (2004) and Waldron (2011). Forsdyke (2018) is the first study of the rule of law in Athens, to my knowledge, that deals more systematically with issues of definition. She questions how much legal certainty can be found even in modern systems, but still underestimates Athenian conformity, and overestimates the incompatibility between the rule of law and legal uncertainty as determined by the existence of argument about the law. 20 For the ideal of the rule of law in Greek identity see e.g. Hall (1989: 198-200) and particularly Harris (2006: 3-29). 21 See Hansen and Nielsen (2004) and Hansen (2006) for an invaluable inventory of the Archaic and Classical Greek city states. After the conquest of Alexander city states spread even further around the Mediterranean Sea and beyond, so the number of Greek city states must have been considerably higher. across constitutional divides in the world of the Greek city states. Section 2 focuses on the Archaic Greek world, showing that recognisable ideas about the rule of law informed the very earliest written statutes preserved from ancient Greece, and discussing some of the poems of Solon to demonstrate that these took shape in the context of explicit reflection on the rule of law. It will also show that the rule of law was deployed in opposition to tyranny-rule of one man-and that its ideological purchase was so strong that monocratic power itself, to be considered legitimate, had to characterise itself as law-abiding. After establishing the place of the rule of law as the universal foundation of political legitimacy in the Greek city states, Section 3 discusses some important texts of the late fifth and fourth centuries BCE in which the political legitimacy of different constitutional forms-chiefly democracy and oligarchy-is debated in terms of whether these constitutions conform to the shared ideal of the rule of law. It shows that a recognisable notion of the rule of law was the measure according to which, in the fifth and fourth centuries BCE, the legitimacy of particular constitutional arrangements was measured.

The Rule of Law and the Rule of Man in Archaic Greece
The first written law preserved on stone from the Greek world comes from the Cretan city of Dreros, and is dated to the mid-seventh century BCE (Koerner 1993: no. 90 = Nomima I no. 81). 22 Its text is fragmentary, but the key provision can be confidently read: The polis has decided: when someone has been kosmos, within ten years the same person is not to be kosmos again. But if he does become kosmos, whenever he gives judgment, he himself is to owe a fine of twice the amount, and he is to be without rights (to office?) as long as he lives, and whatever he does as kosmos shall be void. And oath-swearers (are) the kosmos and the damioi and the twenty of the polis. (tr. Gagarin-Perlman) It is striking that this very early text of law is concerned with limiting the power and the term of office of the highest official in the city, and implicitly states that political power-even the highest official-is subject to the laws, which are sovereign and must be respected by everyone, even by the rulers. The law also creates penalties for the kosmos (the highest magistrate in Dreros) that breaks the law. The ideas underpinning such provisions are already compatible with recognizable conception of the primacy of the law over the decisions of man, and point to ideas of the rule of law. This text is not an isolated case: another Cretan law, of the sixth century, from Gortyn, provides that the same man cannot be kosmos again within three years, gnomon within ten, and kosmos of the foreigners within five (Koerner 1993: no. 121 = ICret IV 14). The concern with limiting the power of political officials is an overarching one in Archaic laws, which show a commitment to preventing the arbitrary rule of man and the concentration of power in few hands. The Greeks created laws that distributed powers and prerogatives among various boards of officials, political bodies, and sections of the population: Solon distributed powers and prerogatives among four property classes ([Arist.] Ath. Pol. 7.3-4); a law from Chios (Koerner 1993: no. 61) grants different duties and responsibilities to various bodies and officials (cf. also Koerner 1993: no. 39, 74, 87). Archaic laws also establish term-limits for magistrates (Koerner 1993: no. 77, 90, 121), impose penalties for magistrates that do not uphold the law (Koerner 1993: no. 31, 41), and often assign the enforcement of a rule not to an individual but to a board of officials to avoid the concentration of power. 23 These checks over the power of public officials are matched in Archaic laws by a parallel concern with securing equality before the law for all citizens. A key example of this concern is the Gortyn Code, the most extensive collection of laws from any Greek polis, which starts with the provision: 'If anyone wishes to contest the status of a free man or a slave, he is not to seize him before a trial'. The following provisions set heavy fines for whoever violates this rule, and create procedures for bringing relevant disputes to trial. The law is immediately opposed to the arbitrary will of men, and secures access to judicial proceedings to all. No one's status can be denied or contested without trial, and if it is contested, the defendant has the right of access to the courts. The same principle is enshrined in the reforms that, according to several sources, 'founded' the Athenian legal system: those of the lawgiver Solon, enacted in the early sixth century BCE. 24 The most important of these reforms created the possibility for all members of the Athenian demos (the plethos, in the words of [Arist.] Ath. Pol. 9.1) to appeal (ephesis) to a dikasterion (a lawcourt), and introduced the generalized standing to sue-that is they 'entitle the volunteer to exact a penalty in the interest, in the name, and on behalf of the offended party'. 25 [Arist.] Ath. Pol. 9.1 mentions these reforms together with a ban on loans on the security of the body, which outlawed enslavement of free Athenians for debts. The possibility of appealing against a judicial decision (in most cases taken by chieftains and magistrates) and have a case heard by a dikasterion (the author of the Ath. Pol. tells us that the demos was in charge of that dikasterion) is clearly aimed at restricting the arbitrary powers of individuals and securing equal access to the law for all. And, from this point of view, it is also easy to see the reason for the introduction of the possibility for a volunteer prosecutor to sue on behalf of someone else: how else could an orphan's rights be protected? How could an old man sue his own son, who was failing to support him?
It is easy to identify in all these laws a common ideological stance, which affirms the rule of law over the rule of man, creates general and reliable laws for all, as well 23 For these anti-tyrannical concerns in Archaic laws and the methods to avoid concentration of power see Harris (2006: 3-39), with plenty of examples. For a general discussion of procedures to hold officials accountable across the Greek world, see Fröhlich (2004). 24 See Rhodes (2006: 255-256) for these sources and their reliability. Even Mossé (1979: 433-434) and Hansen (1989) agree that these reforms are probably Solonian, although Mossé mistrusts the possibility of a dikasterion and supposes that the ephesis may have been to the Areopagus, while Hansen leaves open the possibility that these reforms may also be fourth-century inventions. 25 Cf. Plut. Sol. 18.2-3, 6-7 and also Arist. Pol. 1273b 35-1274a5, 1274a15-18. For a recent discussion of the institution of ephesis, with abundant discussion of previous scholarship, see Pelloso (2016). as rules and institutions to secure the neutrality of judicial proceedings and equality before the law. Harris, in an important essay, identifies in these features the 'spirit of Greek laws'. 26 And we are not reduced to extrapolate this ideological stance exclusively from the extant fragments of Archaic Greek laws. Solon, the lawgiver, was also a poet, some of whose fragments are extant. Although repetition, expansion, re-orientation and re-performance surely affected the tradition of these fragments, 27 it is risky and ultimately unwarranted by the textual evidence to consider them altogether later constructs. 28 And, as argued forcefully by Martin on the basis of a sensible use of ethnographical comparisons, 'Solon's poems are more than "prime sources" for politics. They are politics, and politics as performed by the most adept practitioners even today, whether in the first world or the third'. 29 Solon is very clear about the role he envisions for himself as a lawgiver. 30 He is not a monarch, not a tyrant, not a ruler. He refuses such a position explicitly: If I spared my homeland and did not grasp tyranny/and brute force, bringing stain and disgrace on my reputation,/I am not ashamed. For I think that in this way I shall be more able/to outstrip everyone. (fr. 32 West; tr. Gerber) His position as lawgiver gave him the chance, instead of setting up laws for everyone, to grasp power for himself, but he did not take it. 31 The reasons for his refusal to become an arbitrary ruler are clear-he refuses to accept that the arbitrary power of one man can be reconciled with the good order of the city (eunomia). 32 He 26 Harris (2006: 3-39). 27 On the effects of these processes in the Theognidean corpus see the essays in Figueira and Nagy (1985). 28 This is the position of Lardinois (2006) and Stehle (2006), but see the careful discussion of these aspects in Noussia-Fantuzzi (2010: 45-66). 29 Martin (2006), and in particular p. 158. 30 For a fuller discussion of these poems and of the ideology of the rule of law they delineate, see Harris (2006: 10-14). 31 As Solon is in fact giving laws to the Athenians, he describes Athens before his intervention as characterised by dysnomia, arbitrary power and abuse. It is he, with his laws, that establishes eunomia, and therefore the rule of law. The chance he had to grasp power for himself was due to the dysnomia that characterised Athens before his laws. 32 There has been much debate about the meaning of eunomia in an archaic context, and scholars usually recognize that the reference is not specifically to written laws: in fact, if Arist. Pol. 1307a 1 and Strab. 8.4.10 are right when they claim that eunomia was the title of the poems of Tyartaeus in which he praised the diarchy and the restoration of the order of Sparta as a guarantee of its stability, it is then clear that there is no necessary connection between eunomia and written law. Solon after all gave the Athenians thesmoi. And yet a connection clearly exists between eunomia and order, and even the etymological sense of nemein, which refers to '(due) sharing' implies that eunomia has to do with a good organization of roles, behaviours, functions. And nomos clearly means in archaic texts 'custom', which has to do with substantive rules of behaviour, with proper dealings with others (including the gods), and with respect of one's own and other people's rights and prerogatives. This is why at Hom. Od. 17.487 and Hes. Op. 249-255 the gods test the mortals' hospitality and they find either their eunomie or their hybris. Eunomie has to do with knowing one's place, one's duties, other people's rights and prerogatives, and behaving accordingly. Unsurprisingly its opposite is hybris, which has to do instead with overstepping one's own rights and therefore dishonouring other people. For the meaning of eunomia in Archaic texts, see e.g. Ostwald (1969: 20-54), Mülke (2002: 150-152), Noussia-Fantuzzi (2010: 258-261). is convinced, on the contrary, that the arbitrary power of one man can only reduce the city into slavery (doulosyne). This is evident from another fragment: From a cloud comes the force of snow and hail,/thunder from a flash of lightning,/from powerful men a city's destruction, and through ignorance the masses fall/enslaved to a tyrant. If they raise a man too high, it's not easy to restrain him afterwards;/it is now that one should consider everything. (fr. 9 West; tr. Gerber) Not only does Solon refuse to exploit his position as lawgiver to become a tyrant. According to later legends about his legislative action, he also refused to combine the roles of lawgiver, ruler and judge. According to both [Arist.] Ath. Pol. 7.2 and Plut. Sol. 25, after Solon gave the Athenians their laws, he had them swear by the gods not to change them for 100 years, and, as a statement that the function of the lawgiver and that of administering the laws are not the same and to make sure that the Athenians could not force him to change the laws himself, he left Athens for ten years. 33 It is far from sure that this story is reliable, and nevertheless it is an important document for how the Athenians understood the action of their lawgiver, and the kind of regime it instituted: not the rule of one man, but the rule of law, independent from any individual who may have created the laws, and administered by judges that are not the same as the legislators and the rulers. Very similar stories are preserved for the lawgivers of other city states, which is evidence that this understanding of the foundational role of the lawgiver in instituting the rule of law, as opposed to the arbitrary rule of man, was a widespread feature of the Greeks' understanding of their political systems. 34 Another Solonian poem (fr. 36 West) is even more explicit in defining what his legislative activity implied-what were its purposes and presuppositions. This poem has given rise to plenty of scholarly debate, and it is not necessary for our purposes here to discuss all the interpretative problems, let alone to solve them. 35 In this poem, Solon defends his legislative action against criticism, and the argument starts from the defence of the freedom of the Athenians against enslavement (douleia), and turns then to the chief instrument to secure their freedom: enacting laws for everyone. It is a powerful statement of the centrality of equality before the law, and of the rule of law, as the main barrier to arbitrary power and political subjection. The poem opens with a rhetorical question to Solon's critics: 'Before achieving what of the goals for which I brought the demos together did I stop?' 36 A vindication of his achievements follows. His first claim is to have freed Dark Earth, the mighty mother of the Olympian gods, 33 See Harris (2006: 11-13) and Westbrook (2000: 42): 'legislation became detached from the legislator'. Cf. also Naiden (2013). 34 For a thorough discussion of the traditions of the lawgivers see Hölkeskamp (1999: 28-59 36 At least, the portion we have opens with this question, because the particle de shows that this can be the beginning of the poem only if it was part of a sympotic chain, cf. Noussia-Fantuzzi (2010: 460). Some scholars take this question to be an admission of partial failure. But see Jaeger (1945: 452 n. 59), Blaise (1995 and Noussia-Fantuzzi (2010: 461-462). previously enslaved, removing the horoi (boundary markers) that burdened it. 37 After this, Solon enters into detail. Many Athenians were reduced to slavery, sold abroad or working in Attica, and many were forced to flee abroad because of necessity. Solon restored them to their rightful status, and made it secure. At ll. 18-20 Solon explains how he achieved this: he wrote laws (thesmous) that defined the rights, prerogatives, duties, timai of all citizens, the agathoi and the kakoi alike. Solon claims to have legislated for everyone, and this is a powerful statement of the importance of equality before the law. The argument is that an ordered society (eunomia) is one in which one is secure in his position and will be treated, and will receive justice, appropriately and in accordance with his rights and his status. The last few lines of the poem, ll. 20-27, elaborate on the way in which Solon has managed not to side with anyone, despite the pressures, and therefore to set an order in which everyone is provided for. He nicely summarizes the difficulty of his task with a metaphor: he was like a wolf among a pack of dogs.
For Solon, then, the arbitrary rule of man brings disorder (dysnomia) and slavery (douleia), and the alternative is the rule of law. This order, as we have seen from the evidence of Archaic statutes, is guaranteed by equality before the law for officials and citizens alike, and by institutional devices to avoid the concentration of power. And, accordingly, the lawgiver himself needs to step aside, because the function of giving laws to the city cannot be performed by the same person who then administers the laws, and runs the city more widely. While the rule of law is explicitly conceived and predicated in opposition to the arbitrary rule of man, the relation between an understanding of the rule of man as problematic and the commitment to the rule of law is not straightforward: the first does not precede the second, but they rather reinforce each other. In fact, the ideology of the rule of law affirmed itself across the world of the Greek poleis to the extent that perfectly traditional forms of monarchical rule, that are known very well from the Homeric poems, and are there portrayed as good and legitimate, came to be understood as arbitrary, illegitimate, and to be associated with lawlessness. 38 The Greeks even found a new word to indicate those traditional forms of monarchical rule (and to dissociate their contemporary manifestations from Homeric forms of basileia): tyrannos. 39 This was a foreign word, which was used to characterize monarchical rule as quintessentially 37 On the meaning of this expression see Harris (1997).  Cairns (2015). Taylor (2017) makes the important case that the forms of monarchical power that are later understood as tyrannical are not qualitatively different from previous legitimate forms of basileia. What changes is the ideological consensus about what is legitimate: with the rise of the ideology of the rule of law (and its expression in written statutes), these monarchical forms become unacceptable, and are cast as lawless and contrary to the rule of law. 39 The traditional word basileus was on the other hand normalised and used to indicate public officials within the new legal order of the polis. For the meaning of the term basileus see Lévy (1985) and Carlier (2006). un-Greek. 40 Monarchical rule (tyranny) came to be associated with arbitrary rule-with absolute authority exercised out of whim, without restraint, and outside the control of the laws. The sources provide countless stories about the misdeeds of the tyrants: we read of Phalaris, the tyrant of Acragas in Sicily, who had the habit of roasting alive his opponents in a hollow bronze bull (the first source is Pind. Pyth. 1.95-6); we read of Cypselus of Corinth, who tried to have 300 Corcyrean aristocrats castrated as eunuchs in Sardi, who murdered his wife and had sexual intercourse with her dead body, who gathered the women of Corinth together and stripped them naked, and exterminated his rivals (Hdt. 3.50; 5.92). This understanding of tyranny as a form of personal rule that steps out of the rule of law and becomes fully arbitrary and uncontrolled-the kind of power that Archaic statutes tried to prevent-is encapsulated in Aristotle's much later account of the rise of tyranny: 'earlier tyrannies were due to basileis exceeding the traditional limitations and aspiring to a more despotic form of rule… Thus Pheidon in the case of Argos and others became tyrants when they were already basileis.' (Arist. Pol. 1410b14-28).
Thus, the very understanding of sole-rulership as arbitrary and illegitimate was a development contemporary and connected with the affirmation of the rule of law ideology: traditional monarchical forms were recast within this shared framework and found wanting, and isolated as un-Greek-'[t]he rule of law [became the] accepted measure […] of government legitimacy'. And the strength of this ideological constraint was such that even tyrants, in order to legitimate themselves, came to portrait themselves as guarantors of justice and of the rule of law. This is clear in the positive traditions about particular tyrants, which stress that they were law-abiding and that they fought to enforce the rule of law against aristocrats that attempted to overthrow it. We learn from Herodotus, for instance, that an oracle stated that Cypselus would set Corinth on a path of justice (Hdt. 5.92). The family of the the Orthagorids of Sicyon were also said to have 'treated their subjects moderately and in many respects enslaved themselves to the laws' (Arist. Pol. 1315b 16). Peisistratus, the tyrant that in the sixth century BCE seized control of Athens, was known to have administered everything 'according to the laws', to have created legal institutions such as travelling judges to bring the rule of law also in the most remote villages of Attica, and to have subjected himself to trial when accused with a homicide charge ([Arist.] Ath. Pol. 16.8). His figure is the prototype of the law-abiding tyrant, whose legitimation comes from respecting and enforcing the rule of law. To such an extent was the ideology of the rule of law the chief source of legitimation for any government in a Greek city, that even sole rulers were forced to defend the legitimacy of their rule within this framework. 41 40 Andrewes (1956: 22), and Lewis (2009: 7) believe that the word tyrannos is of Lydian origin. Austin (1990: 289) argues that the word came from Asia Minor. Parker (1998: 145-149) finds Hittite and Old Testament terms for rulers and judges that are similar to tyrannos. The previous section has shown that the Greeks shared from the Archaic period onwards a constitutional identity based on the ideal of the rule of law as opposed to the rule of man (characterized as tyranny). This identity was enshrined in a variety of institutional arrangements that secured the accountability of officials that had to act within the limits of the law, fought concentration of power and its potential consequences for the primacy of the law, attempted to create legal certainty, and ensured (at least notionally) equality before the law and due process. These ideals and institutions were shared across the world of the Greek city states, and were conceptualized as distinctively Greek, as opposed to the monarchical and autocratic world of the barbarians. This opposition formed the ideological backbone of the Greek resistance to the Persians at the time of the Persian Wars. The Spartan exile Demaratus, as narrated by Herodotus, explains to Xerxes, the Persian king, the difference between Greeks and Persians precisely in these terms. The courage of the Greeks is founded on their intelligence and on the force of law (Hdt. 7.103), he explains, and then remarks on the Spartans in particular: The point is that although [the Spartans are] free, they're not entirely free: their master is the law, and they're far more afraid of this than your men are of you. At any rate, they do whatever the law commands, and its command never changes: it is that they should not turn tail in battle no matter how many men are ranged against them, but should maintain their positions and either win or die. (Hdt. 7.104; tr. Waterfield) Within the framework of this opposition between Greek city states characterized by the rule of law and autocratic barbarians, constitutional differences between Greek cities, as to how wide the citizenship franchise and how equally spread political power were, were underplayed. This progressively changed in the years following the Greek victory in the Persian Wars, when the Greek world slowly broke down into two areas of influence, that of Sparta and that of Athens, engaged repeatedly in warfare against each other. 42 In Sparta, the regime was unquestionably an oligarchy, characterized by political power confined in the hands of a small minority, whereas in Athens all freeborn Athenians were politically equal and had equal access to political decision making and political offices. 43 As this opposition intensified, it was accompanied by a heated ideological battle about the desirability of the different constitutional arrangements that, by the later fifth century, brought into sharp focus no longer the common features of the Greek political systems, but the differences between them. While the rule of law was a shared ideological (and 42 The classic account of the fifty years between the Persian Wars and the outbreak of the Peloponnesian War between Athens (and allies) and Sparta (and allies) is the Pentekontaetia in book 1 of Thucydides (1.89-117). For an accessible and rather comprehensive account of the development of this period see Rhodes (2010: 14-76  institutional) feature of the Greek city states, and therefore focusing on it was functional to stressing their similarities and creating a common Greek political identity, stressing the differences between oligarchy and democracy was functional to sharpening the ideological divide between Athens and Sparta. But, as we shall see in this section by analysing some important texts, ideological concerns with rule of law did not for this reason become less important. On the contrary, 'unanimity in support of the rule of law' became the measure of the ideological battle between oligarchy and democracy. 'The rule of law [was the] accepted measure […] of government legitimacy', and oligarchy and democracy were celebrated and criticized inasmuch as they were argued to be compatible or incompatible with the rule of law, to foster or hamper the rule of law as opposed to the rule of man. 44 The first text to be discussed is the Politeia of the Athenians attributed in antiquity to Xenophon, but recognized by scholars as written before 424 BCE, probably in the late 430s, and as the work of a different author. The political outlook of this author is encapsulated by the nickname with which scholars refer to him: the 'Old Oligarch'. 45 This Politeia of the Athenians, normally translated as 'Constitution', is not in fact a constitution in the modern sense-the term refers rather to the political organization and the ways of communal life of a particular community. 46 This text is emblematic of the context of the struggle (ideological, political and military) between Sparta and Athens, and provides a (very partial) formulation of the ideological choices before the Greeks (and the Athenians) at this time, conceptualized as an alternative between the government of the 'best men' and that of the 'worst men'. Within this framework, the Old Oligarch introduces the word demokratia as a negative term, interpreting the term demos not as referring to the people as a whole, but rather to 'the lower classes', 'the mob', as opposed to the leaders of the community, characterized as morally, physically and intellectually superior with terms such as agathoi, aristoi, chrestoi. 47 The members of the demos, 44 These formulations are (modified) from Tamanaha (2004: 3), as they apply equally well to the ancient Greek world. 45 For an accessible introduction to, and English translation of, the Old Oligarch see Marr and Rhodes (2008). For important recent studies, see e.g. Hornblower (2000), Faraguna (2011). 46 The term politeia (Antiph. 3.2.1; Thuc. 2.37; Pl. Resp. 8.562a; Aeschin. 1.5; Arist. Pol. 3.1279a) is defined by Plato as 'the institution of the magistracies and of the magistrates', and by Aristotle as 'a kind of organisation of those that live in the city' (Arist. Pol. 1274b34). It can be used with various meanings, and it indicates mainly the community of citizens and their rights and prerogatives (e.g. Hdt. 9.34.1; Thuc. 6.104.2, [Arist.] Ath. Pol. 54.3) or the constitutional and social arrangement of a city state (e.g. Thuc. 2.37.2,Dem. 18.87,19.184). See in general Bordes (1982), Schofield (2006, Harte and Lane (2013: 1-12 and passim). 47 Cf. Gomme (1956: 107-108): 'demokratia can mean either simply majority rule in a state where all citizens have the vote… or the consistent domination of the state by the masses'. Demos, within the word demokratia, could be used with both the meaning of the people in its entirety, and with that of the 'masses'. There is also much debate on the meaning of the term kratos, which for some refers to a violent and arrogant form of power, while, for instance, Ober (2008a, b) makes the argument that it simply meant in the context of the word demokratia, 'capacity to do things'. Raaflaub (1983: 517-544), Hansen (2006) and Ober (1989) e.g. believe that the meaning 'rule of the majority' or 'rule of the mob' were intentionally pejorative diminutions. Scholars such as Sealey (1974: 282, 292-295) and Canfora (2006: 8), on the other hand, believe that the pejorative meaning was the original one, and that the term demokratia was then 'rescued' by the Athenians to mark, with a positive sense, their regime. in this opposition, are characterized as morally, physically and intellectually inferior, with terms such as kakoi or poneroi. 48 [Xen.] Ath. Pol. 1.5-9 is emblematic of this argumentative scheme, and shows that this opposition is construed in terms of adherence (or otherwise) to the rule of law. The passage uses the word demokratia negatively, from the point of view of the oligarchic, pro-Sparta block, as an insult to the Athenian constitutional model. But the basic presuppositions of the discussion are familiar, and rely on a shared understanding of the rule of law. The 'best men' are naturally more suited to fostering justice and opposing injustice (adikia), whereas the people are characterised by disorder (ataxia) and lawlessness. Good government under good laws (eunomia, the same ideal enunciated by Solon, and typical of Athens, Sparta and any other Greek polis) is described by the Old Oligarch as leading naturally to the subjection (douleuein: lit. enslavement) of the people. 49 Where the people are free and strong, the regime is one of kakonomiabad government characterised by bad laws-the opposite of the rule of law. To create and preserve eunomia, it is necessary that only the more capable are allowed to make the laws (tous nomous tithentas), and that the 'worst' are controlled by the 'best', and are not allowed to be members of the Council or speak in the Assembly. According to the Old Oligarch, eunomia is necessarily characterized by the subjection (or, actually, enslavement, douleia) of the people. 50 Despite the novel ideological aim of the Old Oligarch's discussion-to denigrate demokratia-the basic presuppositions are still those familiar from Solon: the positive model is still eunomia (a concept that, as we have seen, implies notions of the rule of law), and its contrary is kakonomia, a variant of anomia or dysnomia, as lawlessness. The ideological move of the Old Oligarch is to exclude demokratia from the horizon of the constitutional forms compatible with eunomia, and to characterize it as a form of lawless arbitrary power (similar in fact to tyranny). In demokratia, according to the Old Oligarch, we do not find the rule of law, but that of men-of the 'worst'.
In the new fifth-century context of the Greek world divided along political, military and constitutional lines, the eunomia model is no longer used to unite the Greeks (as it did against the Persians), but it is no less central. It is 'the accepted measure […] of government legitimacy', and as such it is used as the shared baseline for criticism across the constitutional divide. It is the foundation of oligarchic criticism of Athens, whose regime is attacked as arbitrary because the people have more authority than the laws. This line of criticism becomes standard, and is still what we find, in less polemical but no less stark terms, in Aristotle a century later. In the Politics (1298b13-15) Aristotle states that 'the democracy which is most considered to be democracy nowadays is one in which the demos is master over the laws-namely an extreme democracy' (trans. Strauss). Aristotle had previously explained that the main feature of extreme democracy is that the plethos (the masses) and not the nomos is sovereign, and this translates into a situation in which the psephismata (decrees/orders passed by the people in the Assembly) and not the laws are sovereign (1292a1-7). 51 And the Aristotelian Atheniaon Politeia (41) characterizes the final phase of Athenian democracy as an extreme democracy in which the people, and not the law, are sovereign, in very similar terms: The eleventh [and final phase of the Athenian democracy, the fourth-century one] was the one which came into being after the return of the exiles from Phyle and the Piraeus, from which date it continued to exist until it reached its present form, all the time granting even more power to the people. For the people has placed itself in control of everything and administers everything through its decrees and its courts, in which the people holds the power.
On the one hand, the ideology of the rule of law becomes the foundation of the criticism of democracy. On the other, the defence of democracy is not founded on a refusal of this ideology, but rather is construed by stressing the compatibility of democracy with the rule of law, and denying that oligarchic constitutional forms are in fact compatible with the rule of law. Pericles' Funeral Speech for the first year of the Peloponnesian War, as reported by Thucydides, is a key witness of this ideological operation, from the same period as the Politeia of the Athenians by the Old Oligarch. 52 This text has often been read as an uncontroversial celebration of democracy, but it is in fact a very polemical text. 53 Its aim is to rescue demokratia from Spartan and oligarchic criticism that denies its compatibility with the rule of law, and to argue that demokratia is in fact eminently compatible with the rule of law, casting doubts at the same time on Sparta's own adherence to that shared model. 54 This ideological move, of which this text must have been only one example among many, eventually won the day (even as Athens ended up losing the war), and by the end of the fourth century BCE most Greek poleis represented themselves as demokratiai, in accordance with a constitutional model understood as the combination of the rule of law and the absence of citizenship franchises. 55 51 Note that the rule of law is not in Aristotle only the measure of the legitimacy of democracy, but of any regime: 'where the laws do not rule, there is no constitution' (1292a32). For a thorough discussion of these passages (and in general of book IV of the Politics) see Besso et al. (2014), with previous bibliography. 52 See in particular Harris (2006: 29-39) for the polarity built in this speech between the Athenians and the Spartans, and Hussey (1985: 123-129) for the view of Spartan values found in the speech. On the genre of the funeral speech see in general Loraux (1986), and, as a selection of fairly recent discussions, Ziolkowski (1981), Thomas (1989: 196-237), Parker (1996: 131-141), Mills (1997: 58-78), Herrman (2009, Low (2010), Shear (2013. The literature on Pericles' speech is immense, but see at least Hornblower (1987: 45-65) (with discussion of previous treatments), Harris (2006: 29-39), Bosworth (2000. Longo (2000) is a good commentary of the speech, with an extensive bibliography at pp. 101-11. For a good discussion of what we can reconstruct of Pericles the orator see Azoulay (2014: 40-51). 53 See Canfora (2006: 7-35) and Hansen (2008) for a recent example of the diatribes to which this speech can give rise, well beyond Classical scholarship. 54 See Harris (2006) for a reading of the speech in these terms. I shall concentrate here on one particular passage of Pericles' speech, which sets this ideological operation in stark relief, and highlights its nuances (Thuc. 2.37).
Our form of government (politeia) does not emulate the laws (nomoi) of our neighbours: we are an example to others, rather than an imitation of them. Its name is democracy (demokratia) because the administration is in the hands of the many, not of the few. In private disputes they guarantee equal treatment to all according to the laws (nomoi). In public matters, if one has acquired renown in some field, we do not select for office from a small number of citizens, but according to the value of each citizen; nor is poverty, for the obscurity of reputation that derives from it, a barrier to office for those that have some good to offer to the city. We are open and free in the conduct of our public affairs and in the uncensorious way we observe the habits of each other's daily lives: we are not angry with our neighbour if he indulges his own pleasure, nor do we put on the disapproving look which falls short of punishment but can still hurt. We are tolerant in our private dealings with one another, but in all public matters we abide by the law: it is fear above all which keeps us obedient to the authorities of the day and to the laws, especially those laws established for the protection of the injured and those unwritten laws whose contravention brings acknowledged disgrace. (Thuc. 2.37; trans. modified from Hammond) The first point to note is that the celebration of Athens' regime, even in a paragraph that is often read as the vindication of democracy, does not start from the demos, but rather from the laws. It is the excellence of the laws that is cause of the excellence of the city. It is instrumental, in the context of criticisms of democracy such as those of the Old Oligarch, that characterise democracy as incompatible with the rule of law and subject to the rule of man, to start the defence of Athens' constitution explicitly with its laws, to mark their primacy. Democracy comes into play only after this statement-Pericles states explicitly that this politeia, as defined by the laws, is a demokratia. It is a demokratia because the running of the polis is in the hands of the many, not of the few. 56 And Sparta is the implicit polemic target here, because in Sparta the running of the polis is in the hands of the few. Pericles therefore characterises demokratia not as a regime in which the demos has absolute sovereignty, even over the laws, but rather as regime founded on the laws, and that by virtue of these laws gives the many the power the run the city. This vindication of democracy is framed by the laws, which not only precede it, but also follow it: right after this statement, Pericles goes back to the laws and states that the laws guarantee equal treatment to all (and the expression kata tous nomous is a reference to the Judicial Oath, and found very often in Athenian decrees). Thus, Pericles affirms the compatibility of democracy with the rule of law by framing the defence of democracy in-between statements of the supremacy of the law over the power of man, and of the equality before the law granted in Athens to all citizens. In affirming that the democratic regime of Athens abides by the rule of law, Pericles sticks to the typical language of isonomia, and states that the sovereignty of the demos (of the many) in matters of administration does not equate to any form of rule of man-it is inscribed within the rule of law.
The openness of the Athenian regime is then vindicated also in the context of the selection of officials, which is open to all and performed on the basis of merit alone. The polemical target is once again the Spartan model, and Pericles' argument is that it is possible to retain the rule of law without reducing the poor to a condition of subjugation that resembles enslavements (the douleia mentioned by the Old Oligarch). This is democracy's contribution to (and qualification of) wide-spread notions of the rule of law. And it is no chance that the next few sentences of this passage are in fact concerned with freedom. The freedom of the demos is defined not as freedom to do whatever they wish, without rules or restraint (as argued by the Old Oligarch), but rather has to do with fear of the laws, 'which keeps us obedient to the authorities of the day and to the laws'.
This demokratia is therefore conceptually vindicated and justified on the basis of the same notion which were used to criticise it: the rule of law. It affirms its compatibility with, and adherence to, this ideal, against the criticism of its detractors. The 'debate' between the Old Olgarch and the Thucydidean Pericles is indeed evidence that '[t]he rule of law [was the] accepted measure […] of government legitimacy'. And, in the century following this ideological confrontation, the Athenians stuck to this understanding of their constitutions as founded on the rule of law, with the laws of the city enabling democracy. Little more than thirty years after Pericles' speech, Lysias, in another funeral speech (Lys. 2.19), argued that it is typical of wild beasts to attempt to gain power one over the other, through violence, whereas men define what is just through the laws, persuade one another through reason, and live their lives submitting to the rule of law and to reason. 57 The same notion of the primacy of the law is found in the Ephebic oath, discussed above. And this submission to the rule of law as opposed to that of man is understood as the only guarantee of democracy, and as its foundation. It also comes to be understood as what distinguishes democracy from other regimes, which are described in turn as incompatible with the rule of law (with a move that is the reversal of what the Old Oligarch, and the fifth-century oligarchs more generally, attempted). Aeschines' speech Against Ctesiphon (3.6) is a prime example of this line of argument: You are well aware, men of Athens, that there are three kinds of constitution in the whole world, dictatorship (tyrannis), oligarchy, and democracy, and dictatorships and oligarchies are governed by the temperament of those in power, whereas democratic cities are governed by the established laws. None of you should fail to note, in fact everyone should be clear in his mind, that when he enters the courtroom to judge an indictment for illegality, he is about 57 Cf. Harris (2013: 3). to give a verdict that day on his own right to free speech. This is why the legislator made this the first clause in the judges' oath: "I shall vote according to the laws". He was well aware that when the laws are protected for the city, the democracy, too, is preserved. (Aeschin. 3.6; trans. modified from Carey).
Oligarchy and tyranny are characterized as regimes founded on the rule of man, not on that of law. Democracy is on the other hand founded on the rule of law, and is made possible by the rule of law. A passage of another speech of Aeschines, the Against Timarchus (1.4-5), makes the same point, with very similar words, but making the implications of the rule of law even clearer: It is agreed that there are three kinds of constitution in the whole world, dictatorship (tyrannis), oligarchy, and democracy, and dictatorships and oligarchies are governed by the temperament of those in power, but democratic cities are governed by the established laws. You are aware, men of Athens, that in a democracy the persons of citizens and the constitution are protected by the laws, while dictators and oligarchs are protected by distrust and armed guards. Oligarchs and all who run a constitution based on inequality must be on guard against people who attempt to overthrow the constitution by force; but you, and all who have a constitution based on equality and law, must watch out for people whose words and way of life contravene the laws. For your real strength is when you are ruled by law and are not subverted by men who break them. (Aeschin. 1.4-5; trans. Carey) Democracy is here represented as the only political constitution compatible with the rule of law, and is legitimised by virtue of this compatibility. The rule of law is still the chief normative ideal-the one that gives legitimacy to political systems. Thus, as democracy is characterised as compatible with the rule of law, oligarchy and monarchical rule are painted as the epitome of the arbitrary rule of man, justified by distrust and force. At both ends of this debate on constitutional forms, the baseline that guarantees mutual understanding, and justifies all arguments, in whatever direction, is still the shared ideal of the rule of law. Much like in the modern world -in our debates across political divides and in the aggregate measures to assess the quality of different political systems, in the world of the Greek poleis '[t]he rule of law is [the] accepted measure […] of government legitimacy'.

Conclusions
In the Greek city states, the rule of law, in its earliest manifestations, was connected to the refusal of monarchical authority-of the arbitrary power of one man. This article shows that the alternative to the rule of one man was soon conceptualised in terms of limitations to the authority of those in power through fixed rules-written laws-that defined precisely their prerogatives, made them accountable for any overstepping of such prerogatives, and avoided concentration of power. Evidence for such measures is found across the world of the Greek city states. But the evidence shows us more than ad hoc devices to limit the concentration of power: it shows us that these laws came to be understood as the embodiment of a shared ideology that affirmed, against the rule of man, the rule of law, understood first and foremost as the superiority of law over any form of political power, which was to be subjected to the law if the citizens of a city wanted to remain free, and not be enslaved (douleia is used in our sources both metaphorically, to refer to extreme political subjection, and literally).
This basic understanding had deeper consequences in how the Greeks conceptualised their legal and political systems. From a legal point of view, the superiority of the law as the key instrument to avoid the enslavement of the citizens translated into a concern with equality before the law and with legal formality, which is particularly evident in Athens, but is found across the Greek world. From a political point of view, the shared commitment to the rule of law soon became a key trait of Greek identity, and therefore the ultimate measure of the legitimacy of any political regime. During the Persian Wars, the rule of law became the mark of Greek exceptionalism vis-à-vis the autocratic political forms of the Persian Empire, and bound together the Greeks in their struggle. In doing so, it served to obscure significant constitutional differences, which came to the fore in the two centuries after the Persian Wars. And yet-I have argued in this article-ideological positioning and debate between democratic and oligarchic constitutional forms did not replace the centrality of the rule of law. On the contrary, the rule of law affirmed itself even further as the highest (and widely-shared) political ideal, the baseline for the discussion, and ultimately the measure of the political legitimacy of any regime.
This extended process, which took place in, and affected, a world of over 1500 city states over several centuries and across a huge geographical stretch, is a significant precedent for the modern 'global endorsement' of the rule of law as the chief political ideal of our age. It is a significant case study-possibly the most significant case study-of a historical process fundamentally independent from that which has led to the affirmation of the modern ideology of the rule of law, and yet one which led to the formation of, and to wide consensus on, a concept of the rule of law which is recognisable to us, and was driven by concerns and aims to a large extent analogous to ours.