The Tyrant’s Progress: The Meaning of ΤΥΡΑΝΝΟΣ in Plato and Aristotle

In: Polis: The Journal for Ancient Greek and Roman Political Thought
Edmund Stewart Assistant Professor in Ancient Greek History, The University of Nottingham Nottingham UK

Search for other papers by Edmund Stewart in
Current site
Google Scholar
Open Access


This article considers a longstanding problem: what does the word τύραννος mean? And if it means ‘bad / tyrannical ruler’, why are good rulers called tyrants? The solution proposed here is that tyranny is not a fixed state of being, or not being, but instead a gradual process of development. To be called a tyrant, a ruler need not embody all the stereotypical traits of tyranny. If tyranny is, by definition, unconstitutional and illegitimate rule, then there may be no clear moment at which one ceases to be a general or king and becomes a tyrant, only a process by which the tyrant gradually strengthens his rule and develops the negative attributes associated with absolute and illegitimate power. A strong trend in not only the political theory of Plato and Aristotle, but also Greek poetry and historiography in general, is to trace the decline of tyrannies from initially good and popular governments to ones that are despotic and unpopular: the tyrant’s progress.


This article considers a longstanding problem: what does the word τύραννος mean? And if it means ‘bad / tyrannical ruler’, why are good rulers called tyrants? The solution proposed here is that tyranny is not a fixed state of being, or not being, but instead a gradual process of development. To be called a tyrant, a ruler need not embody all the stereotypical traits of tyranny. If tyranny is, by definition, unconstitutional and illegitimate rule, then there may be no clear moment at which one ceases to be a general or king and becomes a tyrant, only a process by which the tyrant gradually strengthens his rule and develops the negative attributes associated with absolute and illegitimate power. A strong trend in not only the political theory of Plato and Aristotle, but also Greek poetry and historiography in general, is to trace the decline of tyrannies from initially good and popular governments to ones that are despotic and unpopular: the tyrant’s progress.

The safest road to Hell is the gradual one – the gentle slope, soft underfoot, without sudden turnings, without milestones, without signposts.

C.S. Lewis, The Screwtape Letters

1 The Problem

The problem discussed here will be familiar to most students of Greek history: the meaning of the Greek word τύραννος in the classical period. While this question is hardly original, I can at least plead that, since it remains a topic of live dispute among scholars, an alternative approach – ‘the tyrant’s progress’ – may not be unwelcome. I argue here that, for the Greeks, the word tyrannos always potentially entailed negative connotations of ‘tyrannical’ or ‘despotic’ rule, not dissimilar to those associated with the English cognate ‘tyrant’. However, not all ‘tyrants’ necessarily exhibit all the characteristics associated with tyrannical rule at any one time. Some are more obviously ‘tyrants’ than others. This is because ancient authors conceived of tyranny and the career of the tyrant as a gradual decline into despotism and infamy.

Three factors determine the Greek understanding of this term. First, tyranny was not a constitutional or legal office, but a position of de facto, rather than de jure, power and authority.1 A tyrant may be distinguished from a king (βασιλεύς): an alternative and supposedly early form of autocrat whose privileges were hereditary and whose powers were restricted by custom (e.g. Thuc. 1.13.1). Tyrannos can thus be applied as a label to rulers unconstrained by law (e.g. Xen. Mem. 4.6.12) – especially, but not exclusively, usurpers. Tyrants are, in sum, simply those who possess the means to monopolise constitutional offices and the good life of the polis for their own enjoyment. (For this reason, extreme oligarchies or radical democracies may be said to possess tyrannical power, even though they are not strictly tyrannies).

Second, because tyranny is not a constitutional office, the Greeks considered tyranny not as a fixed state of being, or not being, but instead a gradual process of development. A tyrant will thus not generally begin as a tyrant and one cannot simply become a tyrant in the same way that one becomes a king or magistrate. A magistrate is a magistrate from the moment of his appointment or election, a king from the moment of his accession or coronation. But one becomes a tyrant through a gradual monopolization of power, wealth and honour within the polis. A king or other magistrate may, therefore, become a tyrant if he exceeds his prerogatives and so ceases to be constrained by law. This gradual process is termed here ‘the tyrant’s progress’.

Third, the Greeks believed that this process was inevitably accompanied by a progressive moral decay, sometimes over several generations within one dynasty. A strong trend in Greek political theory and historiography is to trace the decline of tyrannies from initially good and popular governments to ones that are despotic and unpopular. This pattern is explained by the corrupting effects of excessive good fortune and by the tyrant’s own jealous fear for his position. Slowly, but inevitably, he or his successors will degenerate into full and complete tyrants. A tyrant thus does not need to be a bad person or a bad ruler, at least at first – in actual fact, the successful pursuit of power requires singular ability and daring – but there is always the strong possibility that he will become one. As a result, even if some tyrants are gifted rulers, tyranny is nonetheless to be understood categorically as an evil. Thus Herodotus has the Corinthian Socles say that besides tyranny ‘there is nothing more unjust among mortals nor more stained with blood’ (5.92α1). Aristotle (Eth. Nic. 1160a31–b9; Pol. 1289a26–b3) likewise saw tyranny as the worst of the three perverse constitutions (the others being oligarchy and democracy). For Plato, the soul of the tyrant is similarly the most unjust and decadent of all (Resp. 576b–c; Phdr. 248e3): he lives a life that is as wretched for himself as it is calamitous for his victims.

The tyrant’s progress explains, I argue, a major problem: why do some rulers who are called tyrants not fit the stereotype of the illegitimate and tyrannical despot? Some illegitimate usurpers were nevertheless good and successful rulers, such as Pisistratus of Athens or Gelon of Syracuse. Some legitimate kings were barbaric and cruel, such as Pheidon of Argos (Hdt. 6.127.3; Arist. Pol. 1310b.26–31). There were even kings from established royal houses who were usurpers, such as Archelaus of Macedon (Pl. Gorg. 470c9–471d2; Alc. II 141d5–e3), and tyrants who successfully bequeathed their rule to their sons and grandsons, such as the Orthagorids of Sicyon or the Cypselids of Corinth (Arist. Pol. 1315b13–26). The ‘tyrant’s progress’, I suggest, provides a way of understanding these contradictions. A tyrant is not simply a bad ruler but one rather who is liable to be corrupted by the power and possessions he holds.

This argument assumes that attitudes towards tyranny were largely consistent in the classical period and that apparent differences in our literary sources can be explained by the context, genre and purpose of these works. Such a claim inevitably calls into question many of the prevailing orthodoxies concerning the Greek view of tyranny. First, it has been assumed that the word tyrannos was in the archaic period a neutral term signifying ‘ruler’.2 The later contrast between bad tyrants and good kings was ‘a product of a construct of ruling developed in the late fifth and fourth centuries, especially at Athens’.3 Plato and Aristotle, from this perspective, are principally responsible for adopting and popularising tyrannos as the term for a stereotypically despotic ruler. Sian Lewis has even claimed that ‘Aristotle could be said to have invented tyranny’.4

One difficulty with this view is that the pejorative connotations of tyrannos are potentially evident even in texts from the archaic period.5 A second related line of interpretation, therefore, is to suppose that contrasting attitudes to tyranny were also, to some extent, contemporaneous. As Lynette Mitchell has put it: ‘although some Greeks may have rejected rule by one man, not all did, and we need to ensure that we do not consider the views of some to be the views of all’.6 The negative view of tyranny may equally be thought of as something peculiar to specific groups, rather than the Greeks in general. And where else should we look for the enemies of tyrants but in democratic Athens? Thus, Victor Parker has supposed that the negative meaning of tyrannos is something peculiar to the Attic dialect.7

From this second perspective, Greek thought appears to have evolved over the classical period: Plato and Aristotle are the ultimate critics of tyranny, while Herodotus, together with Sophocles (that ‘last great exponent of the archaic world view’),8 represent a mid-point where these two opposing views are most in evidence.9 For Andrewes, the earlier neutral sense of tyrannos was thus retained by the Athenian tragedians, just as it became less common in the prose of Athenian democrats.10 This interpretation also complements well the longstanding theory of the archaic tyrant as a popular champion of the demos.11 Our literary texts, it is supposed, thus testify to a debate on the nature of tyranny, spanning two centuries, that was eventually won by Plato and Aristotle.

This final articulation of the meaning of the word tyrant in the works of Plato and Aristotle is often characterized as very largely theoretical and overly schematic. By contrast, as P.J. Rhodes has recently put it, ‘real life and the ways in which words can be used in real life … are of course complex’.12 Plato in Republic VIII and IX provides a detailed portrait of the tyrannical city and the tyrant’s soul, yet he adduces no historical examples or empirical evidence to support this schema. His aim is to describe the most unjust city and most unjust man, so as to compare this type of polity and being with the just city of the philosopher kings. His tyranny thus represents an ideal of corruption and an apogee of moral degradation achieved, in actuality, by few regimes. Indeed, it is impossible to identify any one individual tyrant as a model for Plato.13 For some scholars there is, moreover, something improbable and entirely unreal about Plato’s tyrant. Andrewes thus dismissed Plato’s account as a caricature of a ‘pantomime ogre’.14 In the most recent study on the tyranny in the Republic, Cinzia Arruzza suggests that it is not tyranny that primarily concerns Plato, but rather democracy: ‘the tyrant in the dialogue is in fact a theoretical figure corresponding to a specific form of democratic leadership developed in Athens in the last third of the fifth century.’15 From this perspective, Plato’s description of tyranny bears only a passing resemblance to political reality.

Aristotle’s discussion of tyranny in Politics V has met with similar criticism. This is surprising, given that Aristotle had access to a vast wealth of empirical information (contained in the constitutional handbooks prepared at the Lyceum). Nevertheless, Aristotle’s clear distinction between kingship and tyranny has seemed as problematic to some scholars as Plato’s discussion in the Republic. Lewis has claimed that Aristotle’s schema lacks nuance – ‘it was not in the interest of his work to blur distinctions’.16 Apparent discrepancies between Aristotelian theory and known examples of actual tyrannies have suggested to some that Aristotle and his school in fact made little use of the historical facts available or, worse, distorted them to fit his theories.17 Ivan Jordović has thus concluded that ‘the usefulness of “the chapter on tyranny” as a source for the historical phenomenon of ancient Greek tyranny is severely limited’.18

It is argued here that such a characterization of Plato and Aristotle’s understanding of tyranny – as either a product of the fourth century, a partisan view, or an overly theoretical conception – is misguided. Both philosophers held views that were less innovative than is sometimes supposed and which were based on the Greek collective experience of real tyrannical regimes. Throughout the classical period, I argue, the Greeks held a relatively coherent and unified conception of what a tyrant was and what the word tyrannos meant. The notion of opposing concepts and ideologies of tyranny has come about, I suggest, through a misreading of the views of Plato and Aristotle.

In what follows, I first outline the idea of ‘the tyrant’s progress’ as it appears in Plato and Aristotle. In the second part, this understanding of tyranny is compared with the views of a selection of predecessors and contemporaries who are sometimes thought to espouse contrasting notions of tyranny. These are Pindar, Herodotus, Xenophon, and Isocrates. This comparison will demonstrate, first, that Plato and Aristotle’s understanding of tyranny differs only in points of emphasis and not on fundamental principles and, second, that the common view of tyranny that emerges is founded on these authors’ understanding of historical facts. It does not follow from this, of course, that their view of tyranny was correct (though I believe this to be the case also). Rather, the argument is merely that the Greeks of the classical period believed such a pattern of behaviour to have been followed regularly by actual tyrants. For reasons of space, the discussion is restricted to these authors alone, though others could be added. Views of tyranny in tragedy will be considered in more detail in another publication.

2 The Tyrant’s Progress in Plato and Aristotle

Even though Plato believed that the soul of the tyrant was entirely corrupt, in the Republic he nonetheless assumes a gradual process of moral decay. He does not require all tyrants to be equally immoral to fit his schema, but rather suggests that they are embarked on, and have reached various stages along, a common road to perdition. Even if tyrants occasionally perform civilized or virtuous acts, it would be wrong on this basis to assume that tyranny can be a civilizing or virtuous institution.

The discussion of tyranny forms part of a wider explanation of how the ideal city might decline, through the successive stages of timocracy, oligarchy and democracy. For Plato, tyranny was the final stage in the perversion of the state after democracy and develops out of strife between the demos and the rich. The tyrant’s progress starts from seemingly good intentions, as a representative of the people. Plato thus assumes that the tyrant is popular at first – or rather that the popular leader is a nascent tyrant.

Socrates: This is clear, then, that whenever a tyrant sprouts up, he is rooted in the role of popular representative and no other (ἐκ προστατικῆς ῥίζης καὶ οὐκ ἄλλοθεν ἐκβλαστάνει).

Adeimantus: It is clear indeed.

Socrates: What then is the beginning of the change from popular representative to tyrant?

Resp. 565c12–d4

Plato’s metaphor here of growth in nature, where the tyrant grows up like a plant from a root, emphasises that this is a gradual change. There is a distinction between the neophyte and complete tyrant. Plato indeed conceives of multiple ‘stages’ in the tyrant’s development that are reached in succession if the aspiring ruler is not killed in the attempt: the request for a bodyguard is one of these, which ‘all those who have progressed to this point discover’ (πάντες οἱ εἰς τοῦτο προβεβηκότες ἐξευρίσκουσιν, Resp. 566b5). The ‘finished tyrant’ (τύραννος ἀπειργασµένος, 566a10) only comes about once he has been exiled and has successfully returned. Then, when his power is complete, he appears transformed from popular representative into tyrant (τύραννος ἀντὶ προστάτου ἀποτετελεσµένος, 566d2–3). Yet even at this late stage, Plato still distinguishes between ‘the early days’ when there is still some pretence at popular rule and the undisguised despot of the final stage of development.

Then surely in the early days and for some time afterwards (ταῖς µὲν πρώταις ἡµέραις τε καὶ χρόν) he will smile at and greet all those he meets, and he will say that he is not a tyrant and promise many things, both in private and public, and he abolishes debt and distributes land to the people and those around him and pretends to be gracious and mild to all.

Resp. 566d8–e4

Yet all of this is a front and an act, since he has truly now become a tyrant.

Plato’s tyrant may not make any very serious attempt at virtue. On the other hand, evil is not a fundamental part of his nature: rather he is gradually corrupted by his position. The rise of tyranny follows partly from decisions taken under previous constitutions. It was a necessary (ἀνάγκη, 562d9–10) consequence of democracy that freedom was taken to such excessive lengths that the constitution was transformed into tyranny. Similarly, Plato adopts the myth of the cult of Lycaean Zeus (565d), where those who taste human blood are necessarily transformed into wolves (ἀνάγκη δὴ τούτῳ λύκῳ γενέσθαι, 565d10–e1), to emphasise that the tyrant’s savagery is an inevitable result of the initial killings that saw him rise to power.

Socrates: For such a man, surely, is not what follows an inevitability (νάγκη) and is it not fated that he will either die at the hands of his enemies or be a tyrant (τυραννεῖν) and become a wolf instead of a man?

Adeimantus: It is very much inevitable.

Resp. 566a2–5

The tyrant is unable to live in peace, even if he wishes to do so. The need to justify his rule and impoverish, and thereby distract, his people, leads him to carry on wars against other cities, which will ultimately be unpopular (566e6–567a3). Similarly, it is fear of retribution that will force him to rely on slaves and foreign mercenaries and to distrust and plot against good citizens ‘whether he wants to or not’ (εἴτε βούλεται εἴτε µή, 567c2). He ends therefore in enslaving the demos and is, in consequence, hated and feared by his people (568e8–569c3), but Plato does not state that this was his initial intention. In a similar way, the soul of the tyrant is, at least initially, the passive creation of ‘tyrant-makers’ (τυραννοποιο, 572e4).

The aspiring tyrant is not a fundamentally bad man in Plato’s view but someone who has not applied himself to understand what goodness is and how, therefore, to be good. This is most evident in the Myth of Er. Here the souls of the dead must choose a new life. The danger is that a soul may choose a bad form of life, and so each man must be able to discern what virtue truly is:

With the result that he is able, in examining the nature of the soul, from all of these things to come to a conclusion and make a choice as to the worse and the better life. And he will call that life worse that leads (ξει) the soul to become more unjust, and that one better that leads it towards being more just.


The life of the tyrant is exactly the sort of life that will lead to injustice. It is no ‘pantomime ogre’, but an innocent man who has achieved some measure of virtue in a well-ordered city, if by habit and not by philosophy, who foolishly opts to be reincarnated as a tyrant (619c6–d1). He only realises his mistake when he sees what future is in store for him in this role. It is the position of tyrant that turns him to evil, just as the well-ordered city had kept him nominally good. This conclusion is informed by the Socratic theory that no one does wrong except in ignorance. The tyrant is simultaneously the greatest dupe and the greatest criminal.19

The importance of education and philosophy is also evident when we examine the other constitutions discussed in Republic VIII. For Plato, the just city could only be created through the strict education of the guardian class. Other forms of government are inferior because their members are progressively less well educated. The members of the timocratic state pursue virtue (ἀρετή) only in part and neglect philosophy (547e1–548a3), while the timocratic man has abandoned reason (λόγος, 549b6), which will prove a fatal flaw. The oligarchic man neglects education entirely in favour of money-making, and while the timocratic man replaced reason as his guide with the spirited part of the soul (τὸ θυµοειδές), the oligarch reveres only the appetitive part (τὸ ἐπιθυµητικόν, 553b7–c7). His son, the democrat, thus lacks the education that would enable him to resist ‘unnecessary desires’ (559a5) and so the ‘acropolis of his soul’ is captured by false reason (560b6–9). The worst forms of desire then take control of the tyrannical soul and choose a single overwhelming Desire (Ἔρως) as their tyrant (571b2–573b7).

If a lack of education is the main cause of moral decline, one may wonder whether a tyrant, or a member of a tyrannical city, if he were removed from the noxious influences of his environment and provided with the education he lacks, could not be transformed by degrees into a philosopher-king. In other words, could the process by which the tyrannical man was corrupted be put into reverse, if, that is, the tyrant were not so perverted as to be immune to the improving influences of philosophy? To paraphrase Plato’s description of the cave in Republic VII, the tyrant would thus be led gradually from deep darkness towards the light of understanding and then, once trained, sent back to reform his oppressed and brutalized subjects. Such a view is not stated explicitly in the Republic, but it follows naturally from Plato’s general theory of education in this work. All human beings possess the ability to learn, which simply involves the ‘turning’ of the pupil (περιαγωγή, 518d4) from contemplation of worse objects to higher and better things. Socrates intends to compel (ἀναγκάσαι, 519c9) the guardians to view the Form of the Good. The creation of the tyrant is essentially the reverse of this process: his soul is turned (ἀγόµενον, 572d10) to crime by bad influences and the necessary consequences of his actions.

A passage from Laws IV shows that Plato did envision the possibility of educating a tyrant. Here the Athenian discusses a law-giver’s requirements for the successful foundation of a new state. One of his hopes is that the city for which he is designing a new constitution should be ruled by a tyrant (709e6–710a2). Arruzza has suggested that ‘tyrant’ cannot have the same meaning here as in the Republic and she attempts to draw a distinction between a theoretical conception of tyranny and Plato’s understanding of actual reality.20 Yet such speculations are in fact unnecessary. Cleinias, one of the Athenian’s interlocutors in the Laws, has at least no illusions regarding the evil of tyranny, since he says that he has no desire to see a tyrannical city (711a7). The value of tyranny from the Athenian’s perspective is that it is possible for a tyrant, once under the guidance of a wise advisor, to transform his regime quickly and effectively to the ideal city the lawmaker envisages (711a1–3, b4–c2). The Athenian specifies that the tyrant should be the sort who is open to such persuasion: ‘young and with a good memory and good at learning and brave and great-hearted by nature’ (νέος καὶ µνήµων καὶ εὐµαθὴς καὶ ἀνδρεῖος καὶ µεγαλοπρεπὴς φύσει, 709e7–8). This young man should not therefore have embarked far along the ‘tyrant’s progress’ and, like the guardians in the Republic, should have the natural qualities needed to be responsive to good education. Young men, it is assumed, are still at a stage where they can be moulded by good (or bad) influences. Similarly, in Republic VIII, the successive corruption of each soul takes place in youth (549e–550b8, 553a9–b5, 558a10–b7, 559a3–6, 559b8–c1).

In the Platonic Letters, Plato is presented precisely as one who aspired to be a teacher of young tyrants. Such an ambition is not incompatible with Plato’s view of tyranny in the dialogues. On the contrary, tyrants are most in need of instruction because they are most vulnerable to temptation and most capable of evil. But if they could be formed into philosophers, the Athenian suggests (Leg. 711b4–c2), they would equally have the power to do much good. According to the Seventh Letter, at Syracuse Plato had found in Dion just such a young man as that described in the Laws. Dion was also a youth when Plato first met him during the tyranny of Dionysius I. Moreover, Plato says, ‘being good at learning (εὐµαθής) in other things and in the arguments then propounded by me, he listened to me more closely than any of the young men (νέοι) I ever met’ (Leg. 327a6–b1). Dion would later persuade Plato to try to instruct the young Dionysius II, though here Plato met with little success. Conversions of tyrants are in fact rare and extraordinary achievements. Both Plato in the Seventh Letter and the Athenian of the Laws stress that the meeting of philosopher / law-giver with a suitable pupil is the result of rare good luck or even divine providence (Ep. VII 326e1–2; Leg. 710c7–d3). Nevertheless, the possible benefits of creating a philosopher king are sufficiently great as to make any attempt worthwhile. The similarities between these texts suggest that, even if the Seventh Letter is not by Plato, the education of tyrants could clearly be imagined as a Platonic project. This again suggests that Plato conceived of tyrants as varying in both their natural dispositions and according to the level of corruption to which their souls had been exposed.

Like Plato, Aristotle also believed that tyranny was a corrupt form of constitution, to be contrasted with monarchy. Yet although tyranny and kingship are different entities, it would be wrong to assume that Aristotle applies these categories inflexibly. An indication of his general method is provided by his discussion of the differences between democracies in Politics VI. Although he provides a list of characteristics that can be used to define a democracy, he acknowledges that they are not all found in all democracies. Indeed, it would be unwise to try to incorporate every democratic feature for the sake of creating a ‘complete’ democracy (1317a29–37). Instead, democracies differ according to the composition of the citizen body (1296b24–31) and the best democracy may closely resemble Aristotle’s preferred constitution of politeia (Eth. Nic. 1160b19–21), while the worst stands in the same relation to the best as does tyranny to monarchy (1292a17–18). Indeed, it is possible for one constitution to change into another, as the Syracusan politeia degenerated into a democracy (1304a27–9).

The same considerations appear to have applied in his discussion of tyranny. As with democracy, tyranny can resemble the better form of constitution, which is kingship. Indeed, a tyrant can consciously choose to become more like a king in order to preserve his regime (Pol. 1314a34–9). To gain legitimacy, tyrants pretend not to be tyrants (much as Plato’s despot does in the early phase of his development).

For just as one way of undermining a kingship is to make the rule more like that of a tyranny (τυραννικωτέραν), so the salvation of tyranny is in making it more kingly (βασιλικωτέραν), with one reservation which is power … It is necessary for this to remain on principle, but for the rest he should do some things and seem to do others in a play-act of a king.

Pol. 1314a33–40

In following this strategy, a tyrant could even achieve some level of excellence (ρετ, 1315b9). Cruel and stereotypically tyrannical behaviour is thus the result of only one possible policy by which an autocracy may be preserved. On the other hand, as tyrants can become more like kings, kings can become more like tyrants (Pol. 1310b18–23). The title of king is no protection against the smear of tyrant, since a tyrant is merely a ruler who exceeds the bounds of his office.

Possible similarities between kings and tyrants do not alter the fact that tyranny is the worst form of government. There are two reasons for this. First, once again, the fault lies not in the ruler himself, but rather the position, which is a corrupting influence. Aristotle believed that, as a general principle, magistrates should not be permitted to hold office for long: for the longer they do so, the more they will be tempted to exceed the limits of their post. It is for this reason that tyrannies typically arise in democracies and oligarchies (1308a20–3). It follows that an established tyrant would similarly be tempted to rule in a more tyrannical and openly illegal way over time. Second, even if a tyrant manages to remain a good ruler in his lifetime, it is very unlikely that his successors will be as fortunate and over time the tyranny is doomed to decay. Tyranny and oligarchy are as a rule the shortest-lived constitutions (1315b11), while the major exception, the Orthagorid dynasty, was preserved largely by a combination of uncharacteristically moderate rule, observance of the laws and military success (1315b15–19), which slowed the onset of decay. Most tyrannies are typically overthrown in the second or successive generations, as these later rulers lose power through their ‘loose’ style of life (1312b22–5). Aristotle significantly assumes that the young and rich are especially likely to be hybristic (Rhet. 1278b28–9) – and therefore, we may assume, young tyrants are more likely to provoke revolution.

Thus, Aristotle is not inconsistent: tyranny is indeed the worst form of government, but Aristotle’s understanding of the word tyrant allows for flexibility, as does Plato’s. Both are aware that there could be a spectrum of good and bad rulers, but that the most complete form of tyrant was a bad ruler. Any apparent inconsistencies need not be the result of changing or differing attitudes towards tyranny, but rather the consistent principle that the ‘tyrant’s progress’ is a gradual process.

3 The Tyrant’s Progress in Pindar, Herodotus, Xenophon and Isocrates

The model of tyranny proposed by Plato and Aristotle is more flexible than is sometimes supposed and allows for the seeming virtue and temporary popularity of certain historical tyrants (such as the Orthagorids or Pisistratids). Individual tyrants are not always archetypical monsters, but tyranny is always a corrupting influence that eventually turns rulers to evil, even if its effects are only felt over several generations. The tyrant’s progress can only be stopped or reversed with great difficulty, and the long-term effects are generally the ruin of the tyrant, his family and his city. Is a similar theory of tyranny in evidence elsewhere?

3.1 Pindar

The earliest reference to constitutional theory is found in Pindar, who distinguishes in his second Pythian three types of government: that of the single ruler, which he terms tyranny (παρὰ τυραννίδι, 87), popular rule and the rule of the ‘wise’ (that is democracy and aristocracy respectively). As no distinction is offered between good kings and bad tyrants, it has been supposed that Pindar viewed tyranny as ‘a neutral force’.21 Similarly, in Pythian 3, Pindar implies that Hieron is a tyrant:

A fate consisting of good fortune attends you. For great destiny watches over a leader of the people, a tyrant (λαγέταντύραννον), if over any man.

Pyth. 3.84–6

Mitchell argues that Pindar could not have made this claim ‘if tyrannos was a negative term’, while Kathryn Morgan similarly suggests that ‘there can be no pejorative connotations here’.22 Indeed, Morgan has argued that Pindar was making the case for the possibility of virtuous kingship in the face of ‘developing Greek ideas on freedom in the wake of the Persian wars’.23

Such a conclusion is not inevitable, however, if it is possible to conceive of a spectrum of rulers who can collectively be termed tyrant. All are vulnerable to the possibility of corruption and moral decay, but not all have necessarily succumbed to it. It would then be the duty of the ruler’s wise advisor to warn against the dangers of power and to steer the tyrant towards virtue – something that Aristotle believed was achievable in theory for the king or kingly tyrant and which Plato had attempted, supposedly, to inculcate in Dion and Dionysius II. The author of the second Platonic Letter (311a) in fact compared Plato’s relationship with Dionysius to that between Simonides and Hieron and so suggested that Plato was following in the same tradition as the earlier lyric poets.

Such a view of autocracy and the role of the poet is in evidence in the text of Pindar, while the case for a competing Pindaric ideology of tyranny is by no means conclusive. Hieron is not addressed directly as ‘tyrant’ in Pythian 3 or anywhere else, but at most Pindar only implies that he is one. Where Hieron himself is named with a title, the poet notably prefers the term basileus (Ol. 1.23, Pyth. 3.70; cf. Pyth. 1.60). Pindar is thus conceivably aware of some negative connotations associated with the term, which are sufficiently serious to dissuade him from calling Hieron ‘tyrant’. At the same time, the implication that his patron belongs to the general category of tyrant is clearly less offensive. Why is this? Here we may postulate a general assumption that not all tyrants epitomise the stereotype. Pindar warns us, however, that even a ‘king’ as successful as Hieron could potentially degenerate into a tyrant.

In the first Pythian, Hieron is presented with Croesus as a model, the wise king whose ‘kindly virtue (arete) does not perish’ (94). Croesus is then contrasted with the brutal Phalaris (95–6), infamous for baking his subjects in a bronze bull. The wise poet instructs his master to pursue a good reputation by spending money lavishly (90), as Croesus did, and not to be ‘tricked by shameful gain’ (µὴ δολωθῇς, ὦ φίλε, κέρδεσιν ἐντραπέλοις, 91–2). Pindar is aware that virtuous kingship is possible but also that rule presents many temptations. He implies that Hieron could instead choose to augment his wealth and power further through unjust and tyrannical means. Through generosity and liberality, on the other hand, he will become an object of praise (rather than envy and hatred). Unlike Plato, Pindar does not suggest that the transformation from Croesus into Phalaris is inevitable (that would have been a most injudicious suggestion in this context), but it is still possible.

In other contexts, where Pindar’s patron is not a ruler, the dangers of tyrannical excess are even more clearly expressed. In a conventional note of caution to the Theban athlete Thrasydaeus, Pindar explicitly juxtaposes the benefits of moderation with tyranny:

May I desire fine things that come from the gods, struggling for things that are possible in my time of life. As I find that, of the things in the city, the middling part flourishes longer in prosperity (τὰ µέσα µακροτέρῳ ὄλβῳ τεθαλότα), I criticise the lot of tyrannies (µέµφοµαἶσαν τυραννίδων): I strive for excellence that serves the commonweal, and the envious are kept away.

Pyth. 11.50–4

Pindar’s rejection of tyranny is not an apology for accepting the patronage of tyrants. Rather, he follows a tradition, stretching back to Solon and Archilochus, in which the wise poet eschews excessive and potentially unjust gain.24 He does not here directly criticise tyrants, but merely states that their position is unenviable. His conviction that the ‘middling part flourishes longer’ suggests that tyranny, and the excess with which it is associated, is unstable, a view shared later by Aristotle.

To both of these victors, Thrasydaeus and Hieron, Pindar speaks a note of warning. Both are encouraged in effect not to become victims of their own success. For the former, a private citizen, the virtues of a middling fortune are praised; in the case of the latter, a tyrant who has already achieved great wealth and power, Pindar urges his patron not to cling to his possessions and to avoid the fate of Phalaris (that is, not to progress further as a tyrant). While in Pythian 11 the envious are warded off by moderation (φθονεροὶ δἀµύνονται, 54), in Pythian 1 Pindar accepts that envy is inevitable, but at least it is ‘better than pity’ (κρέσσων γὰρ οἰκτιρµοῦ φθόνος, 85). The difference is one of perspective, not principle. Pindar does not explain the negative consequences of envy, but these are clear from a reading of Aristotle’s Politics. Aristotle indicates that both kings and tyrants are often overthrown or murdered because of their good fortune: ‘monarchs have great wealth and honour, which all desire’ (1311a30). One might suppose that fear of such an overthrow could instil in a ruler, if he is not careful, the covetousness, paranoia and oppressive tendencies that are characteristic of tyrannies.

Ironically, however, such warnings are only appropriate for those who are truly excellent. Tyranny, though an evil in itself, is also a negative consequence of innate excellence and superiority. It is not merely bad men, but those who are good who can be tempted by excessive good fortune. As Aristotle recognised (Arist. Pol. 1311a15–22; cf. Hdt. 1.99.2, Isocr. 9.24), good men are for this reason a threat to existing tyrants, since they are in a position to take their place. Pindar thus praises Thrasydaeus by implying that he is, in theory, the sort of person who could seek a tyranny because of his fortune and virtue. This view is not fundamentally at variance with those of Plato, for whom the outwardly virtuous timocratic man was a direct precursor of the tyrant. Equally, Pindar’s warning to Hieron to avoid the example of Phalaris, and the implication that he is a tyrant, can paradoxically be seen as a form of praise. The poet’s conventional note of warning is appropriate for one as wealthy and successful as Hieron – and is generally to be expected from sage advisors to princes.

3.2 Herodotus

The earliest sustained theoretical treatment of tyranny is found in Herodotus and there has been much debate as to what he understood tyrannos to mean.25 By comparison with Herodotus’ characterful and idiosyncratic portraits of the archaic tyrants, Plato’s single model of tyrannical behaviour can appear excessively prescriptive and reductive. Yet, I suggest, this again can be explained by the fact that the tyrannos is a man undergoing a gradual process of corruption.

This tendency of tyrannical power to corrupt those who wield it is clearly outlined in the debate on the constitutions in Book III. Otanes’ rejection of monarchy and tyranny is motivated by his experience of the hybris of individual bad rulers, especially Cambyses and the Magus (3.80.2). However, Otanes understands this hybris to be an inherent feature of autocratic rule. This is due to both a lack of checks and balances and excessive good fortune that together lead the tyrant to hybris:

And how could monarchy (µουναρχίη) be a suitable constitution, in which the power to act as one wishes with impunity resides (τῇ ἔξεστι ἀνευθύνῳ ποιέειν τὰ βούλεται)? For it would turn even the best of men from his accustomed ways of thinking if he assumed this office (καὶ γὰρ ἂν τὸν ἄριστον ἀνδρῶν πάντων στάντα ἐς ταύτην τὴν ἀρχὴν ἐκτὸς τῶν ἐωθότων νοηµάτων στήσειε). For hybris is engendered in him as a result of his present good fortune (ἐγγίνεται µὲν γάρ οἱ ὕβρις ὑπὸ τῶν παρεόντων ἀγαθῶν) and envy is part of human nature from the start. With both of these he possesses every bad quality: many outrageous things he does when glutted with hybris (βρι κεκορηµένος), others when glutted with envy (φθόν). True, a man who is a tyrant (τύραννος) should not be envious, since he possesses every good thing; yet the opposite happens in his conduct towards the citizens.

Hdt. 3.80.3–4

Here, the personal qualities of the ruler are not the point at issue. Hybris is not something naturally inherent to the tyrant but to tyranny. Envy, on the other hand, is a vice from which all humans naturally suffer. Here it not just the tyrant’s subjects who are envious (as in Pindar), but the tyrant himself who (whether through fear, greed or pride) will not permit anyone to equal him in fortune. Moderate rule is thus impossible, whether the tyrant intends to rule in such a manner or not. Like Plato, Otanes sees tyranny as something that is ultimately bad for both the tyrant and his subjects. This conclusion seems to have informed his final decision not to enter the contest to become king, since he wished ‘neither to rule nor be ruled’ (Hdt. 3.83.2).

Excessive good fortune, which engenders hybris, leads to and explains the tyrant’s moral decline. Otanes’ warning is not merely the personal opinion of an eccentric Persian nobleman, but instead restates received Greek wisdom (much as Pindar did when he censured the ‘lot of tyrants’). The danger of excessive good fortune and consequent insatiability is a common theme in both the poetry of Solon and the Theognidea:

There is no limit to wealth revealed for men. For those of us who now have the greatest livelihood strive twice as hard. What could sate everyone (τίς ἂν κορέσειεν ἅπαντας)? Possessions are indeed a source of folly for mortals, and ruin rises up from folly, which one or another possesses whenever Zeus sends it upon the wretches.

Thgn. 227–32, cf. Solon fr. 13.71–6

Surfeit (κόρος) has destroyed many men in their foolishness, for it is hard to understand moderation whenever good things are at hand.

Thgn. 693–4

Surfeit (or excessive prosperity) – who appears either as the mother of Hybris (Thgn. 153–4; Solon fr. 6.3–4) or Hybris’ child (Pind. Ol. 13.9–10) – here is seen actively to cloud the judgement of the rich and is consequently one of the main causes of the injustice that fuels civil strife (cf. Solon fr. 4.7–9 West; Thgn 45–52, 605–6).26 We should assume on this basis, as Otanes does, that the tyrant, the wealthiest individual, is also the most insatiable and the most unjust.

It is not entirely surprising, therefore, that Otanes uses the words µούναρχος and τύραννος interchangeably as synonyms. Otanes does not believe that tyrannos is a neutral synonym for monarchos, but rather that the bad qualities of a tyrant can take hold of all autocrats. As Ferrill noted, his ‘confusion’ of the terms is also partly due to the dramatic context.27 When Darius makes his speech in favour of monarchy, the word tyrannos is notably absent – even if good autocrats exist, they are not called tyrants, it seems. Darius does not deny that there are bad rulers (the example of Cambyses makes that impossible) but merely that hybris is an inherent characteristic of all monarchies. The best rule is that of an excellent man (τοῦ ἀρίστου, 82.2), which is preferable to the strife caused by popular or oligarchical rule. He notably ignores Otanes’ contention that even such a man could not avoid being corrupted by his position (a truth that will be borne out by Darius’ later treatment of Intaphernes and his disastrous expeditions to Scythia and Greece). Otanes, on the other hand, appears to have deliberately employed both words in his bid to show that autocratic rule as a whole is corrupting, regardless of the personal qualities of the ruler: ‘what Otanes and Megabyzus were saying in essence was that a monarchos could easily become a tyrannos’.28

The tyrant’s progress, moreover, is evident in many of the narratives of historical Greek tyrants provided by Herodotus. We will here restrict ourselves to considering one well-known representative case: the Pisistratids. Pisistratus’ rise to power is perhaps the most gradual of any tyrant, in that it spans twenty years and involves two periods of temporary exile. Pisistratus is presented in our sources as a popular tyrant who rules with the consent of at least some of his people. In Herodotus’ account, he makes his first attempt at tyranny following his success in the war against Megara and uses this reputation to win a bodyguard (1.59.4). Pisistratus is not without personal qualities and on the first occasion that Pisistratus seizes power he maintains the laws unchanged (Hdt. 1.59.6), a style of government which the author of the Aristotelian Constitution of the Athenians characterises as ‘moderate and more statesmanlike than tyrannical’ (16.2, cf. 14.3). In this, he conforms to Aristotle’s type of kingly tyrant, and indeed Aristotle cites Pisistratus in the Politics (1315b29–34) as an example of a moderate ruler who thus preserved his tyranny longer than most.

However, like Plato’s tyrant, Pisistratus in Herodotus’ account becomes more tyrannical in response to opposition. Herodotus notes that during his first period as tyrant, in which he remained a largely benign ruler, his position was not yet firmly secured and so he was easily driven out (τὴν τυραννίδα οὔκω κάρτα ἐρριζωµένην ἔχων ἀπέβαλε, 1.60.1). In both the first and second attempts he employs deception more than force to achieve his aims and at this stage his bodyguards only carry clubs, rather than spears as is normal for a tyrant (1.59.5). He only finally ‘made his tyranny secure’ (ἐρρίζωσε τὴν τυραννίδα, 1.64.1) after the third attempt, with the help of foreign allies, mercenaries and money. Unsurprisingly, this final return from exile is the most violent and destructive. The narrative ends with a brief summation of the consequences of Pisistratus’ rule: some Athenians had died in battle and others had fled into exile with the Alcmaeonidae (1.64.3). It is in this context that the omen given to Pisistratus’ father Hippocrates, and recorded at the start of Herodotus’ narrative, makes most sense: the seer Chilon had advised him never to take a wife or have children (1.59.1–2). The implication is that Pisistratus’ tyranny, when taken as a whole, was a disaster, even if aspects of his rule were not especially tyrannical. Once again, while a tyrant may not always begin as a violent ruler, he is likely to end as one.

Herodotus suggests that the tyranny continued to decline further following the death Pisistratus. His successor, Hippias, ruled yet more tyrannically following his brother Hipparchus’ murder in 514 (5.55). After his narration of the assassination of Hipparchus, Herodotus returns to his account of the exile of the Alcmaeonidae (5.62.2). Their banishment is thus a sign of increasing oppression in both Books I and V. Later authors present a similar narrative of decay. Thucydides believed that the Pisistratids had initially governed the city well and virtuously (6.54.5), just as in Herodotus’ account Pisistratus had ruled within the laws in his first period. The regime became increasingly brutal, however, after Hipparchus’ death (6.59.2). Before this point, they had ruled according to the laws, but had maintained their grip on power, as Aristotle suggests a kingly tyrant should do, by making sure that one of the archonships was held by a member of the family (6.54.6).

This broad narrative is also adopted by fourth-century works such as the Aristotelian Constitution and the Platonic Hipparchus, in which Pisistratus’ rule appears as the ‘age of Cronos’ in comparison to the harsher regime of Hippias (Ath. Pol. 16.7–8; cf. 19.1, Hipparch. 229b). These authors adapt many of the historical facts to suit their prior assumptions. In the Hipparchus, for example, the Pisistratid Hipparchus is said to be the eldest of Pisistratus’ sons (228b), even though Thucydides (6.54.1) clearly showed that he was in fact a younger son and not tyrant of Athens. As in Herodotus, in the three years after Hipparchus’ death Hippias is said to have ruled in a more tyrannical fashion, but the author of the Hipparchus goes further in asserting that ‘it was in these years only that there was a tyranny (τυραννίς) in Athens’ (229b). The underlying assumption is that older rulers are less hybristic and that dynasties such as the Pisistratids become more tyrannical over time. This same narrative of decline is, however, also present in, and is partly derived from, Herodotus.

3.3 Xenophon

Xenophon, Plato’s contemporary, provides a theoretical discussion of tyranny in his Hieron, a treatise inspired both by contemporary attitudes and those of the earlier lyric poets. In this work, Pindar’s rival Simonides takes the part of wise advisor to Hieron. In the first part (chapters 1–7), Hieron explains why the ‘lot of tyrants’ is indeed blameworthy, as Pindar suggested. In the second (8–11), Simonides outlines how Hieron can nonetheless be a good ruler. We may be tempted to imagine that the treatise thus juxtaposes two opposing views of tyranny, with Simonides acknowledging ‘the possibility of the good tyrant’.29 And yet what Xenophon in fact acknowledges, like Plato and Aristotle, is the possibility of a good ruler (archōn) and the conversion of a tyrant into such a ruler.

Xenophon consistently uses the word tyrannos in a negative sense, both in the Hieron and in his other works. In the Estate Manager (21.12) Xenophon contrasts ‘being a ruler over willing subjects’ (τὸ ἐθελόντων ἄρχειν) with ‘being a tyrant over unwilling subjects’ (τὸ δὲ ἀκόντων τυραννεῖν). Similarly, in the Memorabilia (4.6.12) rule over the willing according to law is termed basileia and rule over the unwilling tyrannis. The character of Simonides in the Hieron is equally careful in his choice of language. His purpose in the dialogue is twofold: first, he prompts Hieron into providing an explanation as to why tyrants are fundamentally and always unhappy; then he suggests ways in which Hieron can in fact rule happily over willing citizens (11.12). The passages from the Estate Manager and Memorabilia cited above suggest that this second better form of government cannot be a tyranny, which is defined as rule enforced on unwilling subjects. When, therefore, Simonides embarks on his discussion of good rule in chapter 8, he introduces for the first time a neutral term for ruler (archōn) – the preferred title, it seems, of Hieron’s eventual successor Dionysius I.

The topic of the dialogue at its commencement was a comparison between the life of the tyrant and the private citizen (ὁ τυραννικός τε καὶ ὁ ἰδιωτικὸς βίος, 1.2). At chapter 8, however, Simonides subtly changes the terms of the discussion to a comparison between the ruler and the private citizen (ὅ τε ἄρχων καὶ ὁ ἰδιώτης, 8.3). From this point, Simonides uses the words tyrannos or tyrannis only three times and in a negative sense, when he is referring to what Hieron should not do. At 8.6, Simonides remarks ‘when it comes to boyfriends, in regard to which you especially criticised tyranny (ἐν οἷς δὴ καὶ σὺ µάλιστα κατεµέµψω τὴν τυραννίδα), the age and ugliness of a ruler (ἄρχοντος) hardly displeases them’. Hieron, like Pindar, had criticised the lot of tyrants, but here Simonides argues that a ruler can in fact be happy. At 11.1, the Greek text is as follows: καὶ γὰρ ἔµοιγε δοκεῖ τὰ εἰς τὴν πόλιν ἀναλούµενα µᾶλλον εἰς τὸ δέον τελεῖσθαι ἢ τὰ εἰς τὸ ἴδιον ἀνδρὶ τυράννῳ. The Loeb translation by Marchant and Bowerstock renders this ‘for in my opinion the sums that a great despot spends on the city are more truly necessary expenses than the money he spends on himself’. And yet this translation changes the word order of the Greek in a significant respect by moving the word ‘despot’ (ἀνδρὶ τυράννῳ) to the start of the sentence rather than the end. The original word order suggests that we should understand τυράννῳ as a dative of advantage following τὸ ἴδιον, rather than of agent following ἀναλούµενα.30 This passage therefore should be translated as ‘for in my opinion sums spent on the city are more truly necessary than those spent on private matters for the benefit of a tyrant’. Finally, at 11.6 Simonides urges that ‘it is not fitting for a tyrant (ἀνδρὶ τυράννῳ) to compete [in equestrian contests] with private citizens’. Instead, Simonides continues at 11.7, ’I declare that your contest, Hieron, is with other state representatives’ (πρὸς ἄλλους προστάτας πόλεων). The word tyrant is used, again, for behaviour that is worthy of reproach and can be contrasted with an equivalent neutral term.

While Xenophon clearly saw tyranny as a word for bad rule, he nonetheless believed that tyrants could change into good rulers. This is not incompatible with the understanding of Aristotle, who thought that autocracies could metastasise into less or more extreme forms of one-person rule. Simonides aims to teach Hieron to make the change to good government, just as the character of Plato in the Letters sought to do with Dionysius. It may be objected that the character of Hieron is by no means the ‘ogre’ of Plato’s Republic: he is aware of the wrong tyranny does and regrets it. Yet Xenophon’s dialogue suggests that Hieron will eventually become an ogre in effect if he does not follow Simonides’ advice. Hieron states that the masses are deceived by appearances into believing that tyranny is something enviable (2.3), but that his experience has shown that the life of a tyrant is a wretched one. This is not unlike the circumstances of the man who chooses the lot of a great tyranny in the Myth of Er in Republic X: once he notices that it has ‘fated him to eat his children among other evils’ (619c2–3), he also regrets his choice. For Hieron, tyrants are not so much fated as compelled to act in unjust and heinous ways: they are ‘forced (ἀναγκάζονται)’ by a need for cash ‘to rob unjustly temples and mortal men’ (4.11); forced (ἀναγκάζεσθαι, 5.2) to make use of bad men because good ones cannot be trusted; and forced by tyranny itself ‘to be a trouble to their own countries’ (ἡ δὲ τυραννὶς ἀναγκάζει καὶ ταῖς ἑαυτῶν πατρίσιν ἐνοχλεῖν, 5.3). The lot of the tyrant, in Hieron’s estimation, does not explicitly lead to cannibalism but it does include filicide: ‘you will find many tyrants who have killed their own children, many who were themselves killed by their children’ (3.8).

Hieron is not obviously maddened by an overwhelming desire for tyranny, as Plato’s tyrant in the Republic is, but this difference is due more to the distinct agendas of these works than any major divergence in doctrine. Plato aims to provide a portrait of the most unjust man possible, in order to compare this individual with the most just, whereas Xenophon’s tyrant has to be one who could potentially be reformed by a wise advisor. Plato’s tyrant is, in his final form, the most complete tyrant at the end of the ‘tyrant’s progress’, while Hieron might perhaps be turned back mid-way in the other direction. And, of course, Xenophon does not try to claim that Hieron actually took Simonides’ advice: the historical record would seem to suggest otherwise.

Much the same pattern of the tyrant’s progress is observable in Xenophon’s historical narratives. Here we will consider the Thessalian tagoi Jason of Pherae and his successors Polyphron and Alexander. Aristotle devotes surprisingly little attention to these contemporary rulers and only mentions Jason as a tyrant once (Pol. 1277a23–5). Lewis suggests that the reason for this is that Jason, as described in Xenophon’s account, does not fit Aristotle’s conception of tyrant and, in addition, that ‘the model of tyranny with which Xenophon operates is thus different from that of Aristotle in the Politics, which tends to dominate modern thought’.31 Yet ‘the tyrant’s progress’, which provides Aristotle’s theory of tyranny with the flexibility that Lewis believes is wanting, can equally be observed in Xenophon’s account of Jason’s history.

The first description of Jason in Xenophon’s Hellenica is provided by Polydamas of Pharsalus (6.1.2–16) as he appeals to the Spartans for aid. Polydamas emphasises that a considerable force of men is needed if Pharsalus is not to fall under Jason’s control. He is a general and ruler of considerable talent and a man of great ability and ambition. Polydamas clearly admires his personal stamina and the way he encourages and trains his men (6.1.6, 15–16). Jason’s strategy of rewarding his soldiers for excellence resembles Simonides’ suggestion in Hieron 9 that a ruler should award prizes to his most successful subjects. Furthermore, Jason possesses that most untyrannical of virtues, great self-control over bodily pleasures (6.1.16), which allows him to achieve whatever he desires. As in the Politics, Jason has the strength, he has assured Polydamas, to enforce his will but he would prefer the Thessalians to be his willing vassals (6.1.7), and thus rule in the manner prescribed by Simonides in the Hieron. His aim is to become tagos of Thessaly, and so possess absolute power legitimately (τῷ νόµῳ, 6.4.28), which he achieves following Polydamas’ capitulation. And indeed not just Pharsalus but many other cities, according to Xenophon, willingly sought to be his allies at the time of his death (6.4.28).

Jason is thus an excellent example of a good ruler over willing subjects, equivalent to Aristotle’s king or kingly and virtuous tyrant. Once Jason had achieved the office of tagos and united Thessaly, he had become, Xenophon tells us, ‘the greatest man of his time’ (6.4.28). This is not unalloyed praise, however. As Pindar had noted, it is the moderate who prosper longest, and so Jason’s fortune proved to be short-lived. Xenophon indicates that, even though he did not rule by fear, he inspired fear nonetheless. Polydamas told the Lacedaemonians that Jason’s allies would revolt if Sparta came in force against him, since all the cities were ‘afraid of where his power would lead’ (6.1.14). The answer, it would seem, is to tyranny. Jason was assassinated by seven young men as he was reviewing his army. The conspirators who survived were honoured by the Greeks, since they feared that Jason was about to become a tyrant (ἰσχυρῶς ἔδεισαν οἱ Ἕλληνες αὐτὸν µὴ τύραννος γένοιτο, 6.4.32). There was much trepidation regarding his planned appearance at Delphi and even a question posed to the oracle asking what would happen if Jason looted the sanctuary (6.4.30–1). The Greeks clearly feared that Jason, though a man of iron self-control on campaign, would nonetheless yield to the temptations afforded by his great power. It is significant here that the one passage in Aristotle’s Politics that refers to Jason mentions his ‘hunger’ for tyranny (Ἰάσων ἔφη πεινῆν ὅτε µὴ τυραννοῖ, 1277a24) – this hunger is the insatiability against which Solon and Theognis warned and the desire that maddens Plato’s tyrant.

Aristotle believed that a tyranny could become more unjust and unstable over time, something which occurs in Xenophon’s narrative of the next generation of Thessalian rulers. The tagoi declined further under Jason’s successor Polyphron (who was suspected of the murder of Jason’s brother Polydorus) and Polydorus’ son Alexander, who murdered Polyphron in turn. According to Xenophon, Polyphron made the Tageia resemble a tyranny (κατεσκευάσατο δὲ τὴν ταγείαν τυραννίδι ὁµοίαν, 6.4.34). It is Polyphron who does actually seek to abuse power by murdering or exiling many of the leading Thessalians, including Polydamas, whom Jason had accepted as a vassal.

But the tagos who most closely resembles Plato’s ‘stereotypical’ tyrant is Alexander. All ancient authorities, beginning with Xenophon, agree that he was a paragon of depravity. Sprawski and Lewis have attempted to restore his reputation somewhat by arguing that Alexander achieved some military success, despite facing greater challenges and more opposition than Jason.32 However, though the decision to quell opposition with violence may indeed be an expedient policy, it is also one that will lead to moral decay and further decline. Thus, Plato’s werewolf of a tyrant begins by killing in self-defence and ends by developing a taste for bloodshed. Though some of his actions may indeed have been expedient, Alexander nevertheless appears as an ‘unjust robber’ (Hell. 6.4.35) in Xenophon’s account. Moral decay is also demonstrated by the tyrant’s lifestyle. Xenophon subtly remarks on Alexander’s lack of self-control when he describes how his wife Thebe had him murdered as he lay drunk in his bed (6.4.37). He may invite here a comparison with his disciplined predecessor Jason.

3.4 Isocrates

Although Xenophon ends his Thessalian narrative with the death of Alexander, that was not in fact the end of the tyranny. Alexander was succeeded by the sons of Jason, who demonstrated a superficial interest in virtue by inviting Isocrates to Thessaly. The philosopher declined the offer and instead wrote a letter in which he advised them against pursuing tyranny. Like Plato and Xenophon in the Hieron, he states that the life of the private citizen is preferable to that of those ruling as tyrants (τῶν τυραννούντων, Epist. 6.11). He warns against those who are urging them towards tyranny (πὶ τὴν τυραννίδα, 12), since such bad counsellors (the equivalents of Plato’s ‘tyrant-makers’) can only see the attractions of wealth and power and not the difficulties and dangers of rule. These advisors are like those who ‘undertake the most shameful and lawless acts (οἱ τοῖς αἰσχίστοις καὶ παρανοµωτάτοις τῶν ἔργων ἐπιχειροῦντες)’, since they hope to benefit from crime with impunity.

The rest of the letter is lost, but Isocrates is likely to have imitated Simonides in Xenophon’s dialogue in attempting to provide a didactic treatise on how to rule well – and so avoid the dangers and temptations of tyranny. His advice does not seem to have been heeded. Diodorus comments that these rulers were praised initially for assassinating Alexander, but that later they changed their temper and, by hiring mercenaries and executing political opponents, they also ‘showed themselves to be tyrants’ (ἀνέδειξαν ἑαυτοὺς τυράννους, 16.14.1). The tyranny only ended when the dynasty was finally driven out of Thessaly by Philip of Macedon. One might suppose that Jason had doomed his family to misfortune and crime when he first sought greatness and tyranny.

Thus far Isocrates does not appear to have held a neutral view of tyranny. His views on autocracy are, however, more fully articulated in three discourses ostensibly composed for the ruler of Cypriot Salamis, Nicocles the son of Euagoras. These are Oration 2 (a treatise on how to rule delivered to Nicocles), Oration 3 (an address to the Cyprians and an argument for the benefits of monarchy delivered in the persona of Nicocles) and Oration 9 (apparently a funeral oration for Euagoras but also a didactic work aimed at exciting in Nicocles a desire to emulate his father). In these works, Isocrates has a high opinion of the character and virtues of Euagoras and his sons, but he also describes the Cypriot monarchy both as a tyranny (tyrannis) and a monarchy (basileia). Nicocles freely describes Isocrates’ address (Oration 2) as an account of ‘how to be a tyrant’ (ὡς χρὴ τυραννεῖν, 3.11). This does not of course mean that Isocrates is advocating tyrannical rule, but its opposite: how tyrants can in fact be good rulers. Does Isocrates regard these two terms as synonymous and is he indeed less precise in his use of language than Xenophon or Plato and Aristotle?

One exception to Isocrates’ apparently indiscriminate use of these terms is when the character of Nicocles is referring to his own rule. His argument for the general benefits of monarchy does make reference to tyranny (3.16, 22, 23): here kingship and tyranny clearly belong to the same type of constitution. From chapter 27 onwards, however, the discourse becomes less general and more personal, as Nicocles begins to justify his own rule and that of his father. Euagoras had restored to his family their ancestral rule ‘with the result that Phoenicians are no longer tyrants (τυραννεῖν) over Salaminians, but these men now hold the kingship (τὴν βασιλείαν), as it was in the beginning’ (28). Here the descendants of a barbarian usurper, and not the original Aeacid rulers, are said to be tyrants. When Nicocles progresses to discuss his own reign, he consistently uses the term basileia (ἐπὶ τῆς ἐµῆςβασιλείας, 32; cf. 36–7, 38, 41, 56, 60).

This suggests that while Nicocles believes kingship (basileia) and tyranny (tyrannis) to be the same type of constitution, autocracy, the Aeacid dynasty re-established by Euagoras is most certainly a kingship. Here Isocrates is imitating Nicocles’ and his father’s actual practice. An Athenian honorary decree, passed in 394/3 after the victory at Cnidus (which had been won in part thanks to Euagoras’ assistance), describes Euagoras as ‘king of the Salaminians’ (ὁ Σαλ]αµινίων βασιλεύς, R – O 11, ll. 6 and 16). We learn from Isocrates that one of the honours Euagoras received was a statue: this was described later by Pausanias as an image of the ‘king of the Cyprians’ (βασιλεὺς Κυπρίων, 1.3.2). It is conceivable that Pausanias was echoing the inscription on the statue base. Nicocles also styled himself officially as king. According to Harpocration (α 118), Aristotle reported in the Constitution of the Cyprians that the sons and daughters of the king (βασιλεύς) were known as ‘princes’ and ‘princesses’ (ἄνακτες καὶ ἄνασσαι). Isocrates refers to this practice in Oration 9.72 (the subject of Harpocration’s note), where he states that none of Euagoras’ children were known by the titles of private citizens, ‘but rather one is called king, and some princes and others princesses’.

Why then does Isocrates, in propria persona, nonetheless associate the kings of Salamis with tyranny? In part, this is because Euagoras’s rule technically could be termed a tyranny, since he was a usurper who had taken absolute power by violent means. Thus Euagoras ‘established himself as tyrant (τύραννον) of the city’ (9.32). Isocrates does not avoid this fact (he cannot, since Euagoras’ daring assault on the city with only fifty men is a major achievement that Isocrates aims to celebrate). Nonetheless he stresses that ‘divine providence took thought on his behalf as to how he might take the basileia in a fine manner, so that another man performed the deeds that by necessity had to be prepared with impiety (ὅσα µὲν ἀναγκαῖον ἦν παρασκευασθῆναι δι᾽ ἀσεβείας)’ (9.25–6). Euagoras usurped the position of a usurper. This individual is described as a tyrant (9.31) and as one of those Phoenicians who had deprived the Aeacids of their ancient kingship – and therefore established a tyranny (9.26, 27). Isocrates here indicates not that tyrants are bad men – in fact Euagoras was marked out for rule because of, and not despite, his natural virtue (9.24) – but that in order to establish a tyranny one must necessarily commit some crime or impiety. Euagoras’ great good fortune, according to Isocrates, was that he avoided having to commit such impiety. The term tyrannos thus still has negative connotations (as in Letter 6 to the sons of Jason), but Isocrates avoids making any direct criticism of Euagoras.

Yet this only explains how Isocrates is able to use this term without seeming to give offence to his patron Nicocles. Isocrates might still have avoided any implication that Nicocles or his father were tyrants. Why did he not do so? The answer lies, as with Pindar, in the didactic purpose of these orations. Like Herodotus’ Otanes, Isocrates blurs the distinction between tyrant and king because he knows that a king can become a tyrant, regardless of the title he assumes. Isocrates later claimed in his Antidosis (15.70) that, in Oration 2, his intention was to speak freely (ἐλευθέρως) to Nicocles and so benefit his subjects by moderating the regime. Isocrates’ presents himself not as a flatterer, but as a wise advisor willing to tell uncomfortable truths. This presumably explains his bold use of the term ‘tyrant’. That Nicocles seemingly permitted such license is also explicable given Isocrates’ intention to better his pupil.

Isocrates’ claim in the Antidosis is borne out by the text of To Nicocles itself, in which Isocrates warns Nicocles of the dangers of rule. Tyrants are more in need of instruction that private citizens and yet generally receive no such education (2.4). Once again, monarchy is not something that should be accepted lightly, since it involves fear and danger, as some rulers ‘have been destroyed by those from whom they least deserved to suffer, others have been forced (ἠναγκασµένους) to do wrong against those closest to them’ (2.5). Isocrates believes that anyone who ‘could turn rulers to virtue’ (ἐπ᾽ ἀρετὴν προτρέψειεν, 2.8) through education would thereby make the regimes of his pupils safer and more benevolent. This is precisely Isocrates’ aim in these discourses.

He is fortunate that he has in Nicocles a pupil who was the ‘first and only one of those possessing tyranny, wealth and luxury to study philosophy’ (9.78). Isocrates describes Nicocles as something of an enlightened despot who, like Dion under the instruction of Plato, follows carefully the advice of his wise counsellor. This is shown to be the case during Nicocles’ address in Oration 3. The king begins with a preamble refuting potential criticisms of those who study philosophy (τοὺς φιλοσοφοῦντας, 1). He further believes that while a good nature is praiseworthy, it is even more commendable to act well through reason (48). Nicocles is clearly of one mind with his teacher and has learned his lesson well. He was, moreover, aware, even when he came to the throne, of the dangers of rule. When he was in need of funds, he noted that others in similar circumstances had been ‘forced to do many things contrary to their nature’ (πολλὰ παρὰ τὴν φύσιν τὴν αὑτῶν πράττειν ἀναγκαζοµένους, 31): that is forced to rob their subjects. Yet he had resisted these temptations: ‘by none of these things was I corrupted’ (οὐδ᾽ ὑφ᾽ ἑνὸς τούτων διεφθάρην). Under his rule there were no ‘banishments, deaths, confiscations of property’ (32) or sexual impropriety (36–7): all crimes typically associated with tyranny. Yet, even so, the success of a ruler depends not just on the individual potentate but also his subjects. Nicocles claims that other tyrants have been ‘forced (ἠναγκάσθησαν) through the bad behaviour of their subjects to rule more harshly than they themselves intended’ (55). If both ruler and ruled strive to live virtuously, then they can lessen the risks associated with autocracy and together enjoy its benefits.

Isocrates thus produces a series of works that belong roughly to the same genre as Xenophon’s Hieron: advice to tyrants on how to rule well. Xenophon had chosen a deceased ruler who was widely regarded as a tyrant as the focus of his discussion. Isocrates instead addressed a living contemporary monarch who styled himself as king. To view the words tyrannos and basileus in the Cypriot Orations as synonyms is overly simplistic. Isocrates’ message is rather that all rulers can be led to behave tyrannically. However, as in Pindar’s odes, this warning is not a criticism, but rather part of an exhortation to continued efforts towards virtue: Isocrates is like a spectator at the games cheering on a champion runner (9.79). Nicocles need not be overly concerned by his association with tyranny, since he has taken the necessary steps to forestall, by means of education and philosophy, his further progression as a tyrant.

4 Conclusion

The Greek view of tyranny is coherent and consistent. By envisaging the development of the tyrant as a continuous process of inevitable moral decay, the Greeks were able to identify numerous leaders of different characters and from different periods, regions and backgrounds as a single type of ruler. The theoretical schemata Plato and Aristotle developed can account for such diversity, since individual tyrants need not have passed all the stages in their development. A comparison with other authors reveals different nuances and points of emphasis, largely determined by the genre and purpose of each work, but few or no conceptual differences. Positive portrayals of tyrants need not be informed by early or positive views of tyranny. While a tyrant can be called a king, this does not necessarily indicate that tyrannos was a neutral term, but rather that a king can decline into a tyrant. It is the tyrant who evolves (or rather degenerates), not the Greeks’ understanding of the word tyrannos.33


See especially G. Anderson, ‘Before turannoi were tyrants: rethinking a chapter of early Greek history’, Classical Antiquity 24.2 (2005), pp. 173–222, at pp. 175–7. Those called tyrants by the Greeks generally adopted titles from constitutional magistracies and in no case is a ruler known to have styled himself as tyrant. This seems to have been as true for the fourth-century tyrant Dionysius, whom the Athenians diplomatically named ‘ruler (ρχων) of Sicily’ (RO 10.7, 33.19–20, 34.8; on the earlier title ‘general plenipotentiary’ cf. H. Berve, Die Tyrannis bei den Griechen (I) (Munich: C.H. Beck’sche Verlagsbuchhandlung, 1967), pp. 236–9), as it was for Pittacus in the seventh century, who claimed the title of αἰσυµνήτης (Arist. Pol. 1285a29–40), even though (as we know from Alcaeus frr. 75.13, 348.3 Voigt) ‘tyrant’ was available as a sobriquet.


See especially V. Parker, ‘Τύραννος: the semantics of a political concept from Archilochus to Aristotle’, Hermes 126.2 (1998), pp. 145–72.


L. Mitchell, The Heroic Rulers of Archaic and Classical Greece (London: A&C Black, 2013), p. 32. See also M. White, ‘Greek tyranny’, Phoenix 9 (1955), pp. 1–18, at pp. 3–4; A. Andrewes, The Greek Tyrants (London: Hutchinson, 1956), p. 30; Berve, Die Tyrannis (I), p. 190–206; C. Mossé, La Tyrannie dans la Grèce antique (Paris: Presses Universitaires de France, 1969), p. 139; Anderson, ‘Before turannoi were tyrants’, p. 174; S. Lewis, ‘Introduction’ in S. Lewis (ed.), Ancient Tyranny (Edinburgh: Edinburgh University Press, 2006), pp. 1–14, at p. 5; M. Koiv, ‘Basileus, tyrannos and polis: the dynamics of monarchy in early Greece’, Klio 98.1 (2016) pp. 1–89, at p. 21; K. Morgan, Pindar and the Construction of Syracusan Monarchy in the Fifth Century BC (Oxford: Oxford University Press, 2015), p. 13.


S. Lewis, Greek Tyranny (Bristol: Bristol Phoenix Press, 2009), p. 90.


Berve, Die Tyrannis (I), p. 3; cf. J.L. O’Neil, ‘The semantic usage of tyrannos and related words’, Antichthon 20 (1986), pp. 26–40, at p. 30.


Mitchell, Heroic Rulers, pp. 15–16; cf. L. Mitchell, ‘Political thinking on kingship in democratic Athens’, Polis 36 (2019), pp. 442–65, at p. 448; Lewis, Greek Tyranny, p. 32; S. Lewis, ‘καὶ σαφῶς τύραννος ἦν: Xenophon’s account of Euphron of Sicyon’, Journal of Hellenic Studies 124 (2004), pp. 65–74, at p.73.


Parker, ‘Τύραννος’, pp. 168–71.


E.R. Dodds, The Greeks and the Irrational (Berkeley and Los Angeles: University of California Press, 1951), p. 49.


See e.g. Andrewes, Greek Tyrants, pp. 27–8; Berve Die Tyrannis (I), p. 194.


Andrewes, Greek Tyrants, p. 23.


See e.g. White, ‘Greek tyranny’, p. 1 and, more recently, K.A. Raaflaub, ‘Stick and glue: the function of tyranny in fifth-century Athenian democracy’, in K. Morgan (ed.), Popular Tyranny (Austin: University of Texas Press, 2003), pp. 59–93, at p. 61.


P.J. Rhodes, ‘Tyranny in Greece in the fifth and fourth centuries’, Polis 36.3 (2019), pp. 419–41, at p. 420.


See M. Meulder, ‘Est-il possible d’identifier le tyran décrit par Platon dans la République?’, Revue Belge de Philologie et d’Histoire 67 (1989), pp. 30–52.


Andrewes, Greek Tyrants, p. 29. The term ‘ogre’ is similarly adopted by V.J. Rosivach, ‘The tyrant in Athenian democracy’, Quaderni Urbinati di Cultura Classica 30 (1988), pp. 43–57, at pp. 44, 53, and 56 n.52.


C. Arruzza, A Wolf in the City: Tyranny and the Tyrant in Plato’s Republic (Oxford: Oxford University Press, 2018), p. 7.


Lewis, ‘καὶ σαφῶς τύραννος ἦν’, p. 69; cf. Ancient Tyranny, pp. 4–6; Greek Tyranny, pp. 8–11, 91; cf. Mossé, La Tyrannie, p. 135.


See A. Heuss, ‘Aristoteles als Theoretiker des Totalitarismus’, Antike und Abendland 17 (1971), pp. 1–44, at pp. 39–40; A. Meister, ‘Das Tyrannenkapitel in der ‘Politik’ des Aristoteles’, Chiron 7 (1977), pp. 35–41, pp. 36–8; A. von Kamp, ‘Die aristotelische Theorie der Tyrannis’, Philosophisches Jahrbuch 92 (1985), pp. 17–35, at pp. 20–24.


I. Jordović, ‘Aristotle on extreme tyranny and extreme democracy’, Historia 60.1 (2011), pp. 36–64, at p. 63.


ὑπὸ ἀφροσύνης τε καὶ λαιµαργίας, 619b8–c1; cf. Plt. 301c and Soph. 230d–e.


Arruzza, The Wolf in the City, p. 58 n.18.


Morgan, Pindar, pp. 197–8.


Mitchell, Heroic Rulers, p. 23; Morgan, Pindar, pp. 13–14.


Morgan, Pindar, pp. 2–4.


Esp. Archil. fr. 19 West; cf. the passages cited by D.C. Young, Three Odes of Pindar (Leiden: Brill, 1968), pp. 6–18 and P.J. Finglass, Pindar: Pythian Eleven (Cambridge: Cambridge University Press, 2007), p. 117–18.


See especially A. Ferrill, ‘Herodotus on tyranny’, Historia, 27.3 (1978) pp. 385–98; V.J. Gray, ‘Herodotus and images of tyranny: the tyrants of Corinth’, American Journal of Philology 117.3 (1996), pp. 361–389; C. Dewald, ‘Form and content: the question of tyranny in Herodotus’, in K.A. Morgan (ed.) Popular Tyranny (Austin: University of Texas Press, 2003), pp. 25–58.


On insatiability and hybris, see further D. Cairns, ‘Hybris, Dishonour and Thinking Big’, Journal of Hellenic Studies 116 (1996), pp. 1–32, at pp. 23–4 and P.J. Finglass, Sophocles: Oedipus the King (Cambridge: Cambridge University Press, 2018), pp. 436–7.


Ferrill, ‘Herodotus on tyranny’, pp. 391–2.


Ferrill, ‘Herodotus on tyranny’, p. 393.


G.J.D. Aalders, ‘The date and intention of Xenophon’s Hiero’, Mnemnosyne 6.3 (1953), pp. 208–15, at p. 213.


Cf. οὐκ εἰς τὸ ἴδιον κατεθέµην ἐµο, ‘I did not put aside [this money] for my own private benefit’ (Xen. An. 1.3.3).


Lewis, ‘καὶ σαφῶς τύραννος ἦν’, p. 69.


S. Sprawski, ‘Alexander of Pherae: infelix tyrant’, in S. Lewis (ed.), Ancient Tyranny (Edinburgh: Edinburgh University Press, 2006), pp. 135–47, at p.138 and p. 142; Lewis, Greek Tyranny, p. 68.


I am most grateful to Stefanos Apostolou, Patrick Finglass, Edward Harris, James Kierstead, David Lewis and Elodie Paillard for reading and commenting on drafts of this essay. I would also like to thank Josh Webb and Graham Shipley for inviting me to deliver a version of this paper at the University of Leicester Archaeology and Ancient History research seminar and for the helpful suggestions that followed from those present.

Content Metrics

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 3864 2336 378
PDF Views & Downloads 4102 2248 233