1. Introduction

In ancient and modern political reflection, the archaic Greek polis is considered the cradle of political systems based on power sharing as well as on the citizens’ participation in collective decision-making.^[1] Yet, in the archaic just like in later periods, it was by no means uncommon for poleis to fall under the rule of a single individual. The Greeks called these individuals not by the standard word they used for ‘kings,’ basileus, but using a foreign word, tyrannos.^[2]

With the exception of few fragments of archaic poets, the vast majority of sources on archaic tyranny postdate the tyrants by centuries and consistently depict them as unlawful autocrats who took power by force and were responsible for all sorts of aberrations. The discourse on tyranny by which these sources are informed proceeds by stereotypes and follows precise conventions, making it almost impossible for us to reconstruct how tyranny worked in practice and what form of political thinking – if any – sustained it.^[3] The modern meaning of the word ‘tyrant,’ with all negative connotations attached to it, is ultimately built on that stereotyped conceptualization: we have all reasons to suspect that tyranny contrasts with our idea of the polis because it contrasted with the idea that Greek political theorists of the Classical period had of it.^[4]

As a result, tyranny remains one of the least understood phenomena in Greek political history. Crucial problems surrounding its origins as a political phenomenon have long been the focus of scholarly debate with little consensus: how and why did tyranny emerge? What was its institutional nature and relation to other polis institutions? The reasons why the Greeks used a foreign word to call city rulers are not fully clear either: is that because autocracy was imported from abroad?^[5] Should this conclusion reinforce the idea that autocracy is intrinsically alien to the polis, as later thinkers put it?

This paper seeks to answer these questions by addressing archaic tyranny from a new perspective, namely a Mediterranean one. Since most sources on tyranny address the phenomenon in retrospect and are thus of little value to understand its origins, I propose to reverse the way we look at it and follow the voyage of the concept of tyranny up to its arrival to Greece. As mentioned, tyrannos was not a Greek word: today the most plausible interpretation views it as a borrowing from the Iron Age Luwian term tarrawanni-.^[6] Surprisingly, however, Luwian tarrawanni- does not mean ‘ruler, autocrat,’ but is an adjective meaning ‘just, righteous.’ A homophonous abstract noun tarrawanni- with the meaning ‘justice’ also features abundantly in Luwian texts. So, what does Greek tyranny have to do with the Luwian concept of justice?

To my knowledge, the first and most developed attempt to find an answer to this question is a study by Franco Pintore, which appeared posthumously in 1983.^[7] By examining the social profile and political rhetoric of the individuals who employ the epithet tarrawanni-, ‘(the) just,’ in Luwian inscriptions, Pintore was able to identify a new, Iron Age type of political actor who set himself aside from forms of rulership of Bronze Age legacy and their traditional means of legitimation. By comparing the profile of the typical tarrawanni- power holder with that of the typical Greek tyrant, Pintore came to the conclusion that the former represents “the cousin of the Greek tyrant.”^[8] More recently, Pintore’s analysis has been resumed and further developed by Maurizio Giangiulio:^[9] observing that the model of rulership represented by tarrawanninzi, based on an emphasis on justice as a new way of claiming political legitimacy, has a significant counterpart in the figure of Cypselus, tyrant of Corinth and one of the earliest known to us, Giangiulio argues that that model, widespread in the Neo-Hittite world, must have been appealing to political actors, like Cypselus, who sought to establish new forms of autocratic power across the Aegean, too, resulting in the borrowing of the designation tarrawanni- as tyrannos.^[10]

Building on these observations, this paper aims to give a new look at archaic Greek tyrants alongside their Anatolian ‘cousins,’ and to further define the dynamics that made it possible for the Luwian concept of tarrawanni- to find its way into Greece. My contention is that only by understanding how and why this concept was received by the Greeks will we be able to cast light on the nature of early tyranny and its relation to polis institutions. Most importantly, given the retrospective nature of the vast majority of our sources on archaic tyranny, this approach enables us to look at tyranny from the perspective of how it originated and developed, rather than from the perspective of how later thinkers believed it originated and developed.

The argument of this paper is divided into three main parts. First (§ 2), I show how both Neo-Hittite and early Greek societies display a similar attitude towards justice as a key element that defines good leadership and legitimizes those in power, but I conclude that this commonality cannot per se explain why the Greeks borrowed the concept of tarrawanni- as tyrannos. Therefore (§ 3), I move from the semantics of tarrawanni- to its political significance and examine what specific way of doing politics is captured by the concept of tarrawanni- that is also captured by the concept of tyrannos. I conclude that the two terms identify political actors whose power could be exercised in any kind of institutional framework or in absence of one, being independent of institutional arrangements. Finally (§ 4), I ask whether it was just a word, a concept, or a political practice, too, that was transmitted from Anatolia to Greece, and lay out hypotheses concerning contexts and timing of the transmission.

2. The Call for Justice in Iron Age Anatolia and Early Greece

Unattested in the Bronze Age, the term tarrawanni- is very common in Iron Age Hieroglyphic Luwian inscriptions from central and south-eastern Anatolia as well as northern Syria, where it is found alone or in combination with political-institutional titles like ‘king’ or ‘Country Lord.’ Attestations cover almost the entire chronological span of Iron Age Hieroglyphic Luwian documents, from the 11th to the late 8th-early 7th century. For a long time, the term has been interpreted as a political title meaning ‘ruler.’^[11] It was Franco Pintore who, some decades ago, clarified that the term is not a political title, but an adjective meaning ‘just, righteous,’ sometimes substantivized and used as an epithet with the meaning ‘the Just (one).’^[12] Pintore’s analysis received little attention in Anatolian scholarship until a recent contribution of C. Melchert, who has reviewed all the evidence and confirmed his conclusions.^[13]

Thus, contrary to what has long been believed, tarrawanni- does not identify any political institution: there is no office of tarrawanni- in any Neo-Hittite polity.^[14] Rather, the word expresses a moral quality and serves the purpose of legitimizing the position in society of several categories of people. Individuals from a variety of socio-political backgrounds present themselves as being ‘just’ in their own inscriptions or are so defined in inscriptions dedicated by others: these include kings^[15] and political actors other than kings,^[16] in whose official titularies the adjective/epithet tarrawanni- usually occupies the first place;^[17] individuals who proclaim themselves ‘servants’ of rulers;^[18] as well as other categories of people, such as scribes,^[19] rulers’ wives,^[20] and more generally members of the elites or the sub-elites with no apparent share in political power.^[21] The high frequency with which the adjective/epithet tarrawanni- is attested in the Iron Age Luwian corpus shows that the theme of justice was a rhetorical topos, certainly among the most widespread ones:^[22] in Neo-Hittite society, there appears to have been a widespread concern for justice both among rulers and among people who were not in power.

Several inscriptions also attest to the corresponding abstract noun tarrawanni-, ‘justice, righteousness,’ most often found in the ablative in the phrase amiyadi tarrawannadi, ‘by / because of my justice,’ which further expands on the moral qualities of the dedicator or dedicatee by presenting their achievements on account of their ‘justice.’^[23] One of these inscriptions contains the passage that proved crucial to determining the meaning of the word, namely Azatiwada’s statement in the Luwian-Phoenician bilingual inscription KARATEPE 1, § 18, where the phrase amiyadi tarrawannadi corresponds to Phoen. b-ṣdqy, ‘because of my justice.’^[24] The root ṣdq, ‘justice,’ is frequently attested in West-Semitic inscriptions, both Phoenician and Aramaic, and in the Hebrew Bible.^[25] It stands for ‘justice’ in moral sense, identifying ‘just, justifiable conduct,’^[26] and should be distinguished from špṭ, which indicates, more technically, the act of dispensing justice, normally a royal prerogative.^[27] The equivalence between West-Semitic ṣdq and Luwian tarrawanni- confirms that the latter indicates a personal quality of the individual,^[28] and warns against interpreting substantivized tarrawanni- as a technical term meaning ‘judge.’^[29]

In political rhetoric, the personal quality ‘justice,’ which is not externally obtained but is intrinsically owned, is presented as the ultimate reason for the rulers’ success as rulers: because of their justice, rulers enjoy the gods’ favor and support, which is in turn a precondition for the prosperity and well-being of their country.^[30] In Hieroglyphic Luwian inscriptions that elaborate on the theme of justice, this logic takes the form of a cause-effect chain: the ruler’s justice triggers the gods’ favor, which triggers the land’s well-being.^[31] To my knowledge, there are only two cases, strictly related to each other, that explicitly depart from this logic: those of Halparuntiya III and his grandson Larama III, kings of Gurgum respectively in the late 9th century and in the early/mid-8th century. In their inscriptions, these kings emphasize that ‘justice’ was bestowed on them by the gods themselves, reversing the usual cause-effect chain:

MARAŞ 1, § 6 (Halparuntiya III)^[32]

|wa/i-mu-ta |lis+la/i/u-si-sá (deus)[sol]-ti-i^?-sá |i-mára/i-si-ha-ⁱ (deus)ru-ti-ia-sá-ⁱ |(“iustitia”)tara/i-wa/i-na-za-ta-’

“And the [Sun-]God of the lawsuit and Runtiya of the open country endowed me with justice.”^[33]

MARAŞ 16, § 2 (Larama III)^[34]

ego-wa/i || l[is+la/i/u-si-sa] (deus)tonitrus-hu-u-za-sá (“iustitia”)tara/i-wa/i-na-za-ta

“Me Tarhunza of the law[suit] endowed with justice.”

As one can see, except for the deities involved, the phrasing is quite the same in both inscriptions, suggesting that Larama III was echoing his predecessor and that his scribes were working with a rhetorical formulation that had been established in Gurgum at least a couple of generations earlier.^[35] As far as we can tell, within the Neo-Hittite world such formulation is unique to Gurgum: the verb tarrawannazza-, ‘to make [sb/sth] have the quality of tarrawanni-,’^[36] thus in this context “(the gods) endowed me with justice,” has no further parallels in the Luwian corpus, despite the widespread use of the rhetoric of justice. The association of tarrawannazza- with the Sun-God (Tiwaz, here in the form Tiwadis) and the Storm-God (Tarhunza) ‘of the lawsuit’^[37] suggests that Halparuntiya’s and Larama’s divine endowment is not limited to moral ‘rectitude, righteousness,’ but also includes ‘equity, fairness, honesty’ in the administration of justice. This seems to reflect aspects rendered by West-Semitic špṭ. Therefore, although there is no doubt that tarrawanni- corresponds to the concept of ṣdq rather than špṭ, we cannot exclude that it could encompass, on occasion, a broader or different range of nuances.^[38] One should also keep in mind that the West-Semitic dialects distinguish between at least three nuances of ‘justice, righteousness, uprightness:’ alongside the roots špṭ and ṣdq, Ugaritic, Phoenician, and Hebrew frequently employ, too, the root yšr, which has the meaning of ‘being straight,’ thus ‘rectitude.’^[39] By contrast, from what we can tell Luwian only shows the concept of tarrawanni-. What is more, in West-Semitic the root špṭ has not only the technical meaning ‘to judge, dispense justice,’ but is also associated with the sphere of rulership:^[40] the logic underpinning this use is not simply that the administration of justice was usually a ruler’s prerogative, but, more generally, that ‘ruling,’ like ‘judging,’ implies ‘deciding.’^[41] The choice of Halparuntiya III and Larama III to speak of ‘justice’ not as an inherent quality of their own but rather as a divine gift has precise implications: their divinely-bestowed ‘justice, fairness, righteousness’ in their capacity as kings and dispensers of justice places them on a level different from people who equally boasted of their ‘justice’ without being rulers, or more specifically kings.^[42] As a matter of fact, it is not moral righteousness per se, but the (fair) administration of justice which distinguishes a ‘just’ ruler from a ‘just’ non-ruler.

If we consider the semantics of tarrawanni- and the rhetoric that was built around this concept, we may observe that the concern for justice in Neo-Hittite society is well paralleled in early Greece. That the theme was a matter of political debate is evident as early as the Homeric poems and the works of Hesiod.^[43] This should be contextualized within a broader set of expectations concerning the well-being of the people, the preservation of which is central to the legitimacy of political leaders – the basileis. In the ‘Iliad,’ leaders such as Agamemnon, Achilles, and Hector all recognize that their main duty is to keep their people safe.^[44] Yet, epic leaders typically fail to fulfill this duty.^[45] The contrast between expectations and the leaders’ failure to meet them reveals ongoing debates on the nature and limits of rulership.^[46] Most notably, it reveals the emergence of a set of values that identify the ‘good ruler,’ the definition of which is negotiated between the rulers and the ruled. A key value that leaders should uphold to fulfill their duties towards the people and ensure their well-being is precisely justice. A passage of the ‘Odyssey’ (19.107–114), put in the mouth of disguised Odysseus when he meets Penelope in his palace, best outlines the relation between justice, well-being, and good rulership: a blameless basileus – this is the model that Odysseus advertises – is one who upholds justice, which, in turn, ensures the prosperity of the land and the well-being of the people.^[47] By contrast, according to another passage of the ‘Iliad’ (16.384–393), those who do not give straight judgment incur the gods’ wrath, which manifests itself in the form of natural calamities, causing the disruption of human subsistence activities and thus affecting, by implication, the people’s well-being.

A similar picture emerges from the works of Hesiod: failing to uphold justice is the distinctive trait of ‘gift-devouring’ basileis (Hes. erg. 35–39, 219–224, 263–264); while ‘just’ basileis rule over prosperous lands and the people flourish under them (Hes. erg. 225–237), the leaders’ injustice triggers divine punishment and causes the ruin of the entire populace (Hes. erg. 238–251). Hesiod’s people are ready to honor their basileis like gods when they uphold justice and adjudicate disputes by means of persuasion, a gift of the Muses to leaders in order for them to compensate the people when they are wronged and thus to ensure their well-being (Hes. theog. 81–93). While in the Homeric world there is still scope for human leaders to be models of good rulership (see the passage of the ‘Odyssey’ cited above), in Hesiod’s cosmos the ideal basileus is divine, not human: the model is condensed in Zeus, presented as the king of the gods (Hes. theog. 886: θεῶν βασιλεύς), the one who restores order and establishes legitimate rulership in heaven. Earthly basileis are far removed from this model, which can only be projected into heavens: justice is Zeus’s gift to humans (Hes. erg. 276–280; cf. 1–8), and leaders are expected to make good use of it, though more often than not they fail to do so.

As is evident, the cause-effect chain linking the rulers’ justice, the gods’ favor, and the land’s prosperity as well as the people’s well-being in early Greek thought is in all respects similar to the logic observed in Iron Age Luwian inscriptions – even though early Greek thought puts a special emphasis on the administration of justice, that is, on the expectation that rulers give straight judgment. As seen earlier, however, this aspect cannot be deemed entirely absent from the notion of tarrawanni- and the rhetoric that was built around it: the case of Halparuntiya III and Larama III in MARAŞ 1 and 16 respectively seems to suggest that giving straight judgment was part of the expectations surrounding Neo-Hittite rulers (and namely in their institutional capacity as rulers) just as it was in early Greece. We may also note that, by presenting their justice as a divine endowment, Halparuntiya III and Larama III project the model of good rulership into heavens just as Hesiod does by presenting justice as Zeus’s gift to mortals.

The discourse on justice keeps characterizing Greek political thought throughout the archaic period.^[48] As the evidence shows, a concern for justice often lies behind the practice of conferring absolute powers on a single individual as a means to solve situations of crisis in poleis: in sources contemporary to them, whether favorable or hostile, archaic tyrants are presented as champions of justice and order, appointed or acclaimed by civic bodies, or at least part of them, in the hope that conflicts internal to poleis might be resolved.^[49] The evidence suggests that the practice of conferring special powers to an individual based on this kind of beliefs must have been relatively common, but also signals that it was not uncontroversial: its opponents seem to emphasize how this practice, too, could easily translate into injustice, although purposeful demonization of political adversaries, and of the people who endorsed them, should be taken into account when analyzing critical stances.^[50]

A case in point is the tradition on the rise of Cypselus as tyrant of Corinth, the original nucleus of which undoubtedly dates back to the 7th century, when the tyrant was still in power.^[51] Cypselus is presented as the just, divinely-blessed straightener of Corinth’s wickedness, represented by the dominant elite of the Bacchiads. This is especially evident in the first Delphic oracle reported by Hdt. 5.92β, which was almost certainly contemporary to the tyrant: the oracle ‘predicted’ that Cypselus would ‘make Corinth just’ or ‘bring justice’ to it (δικαιώσει δὲ Κόρινθον), crushing the Bacchiads like a stone. The resulting image is that of a predestined, dreadful hero who would rise to power to meet a demand for justice that had arisen in Corinth due to the wicked, unjust rule of the Bacchiads.^[52] Cypselus apparently had this image translated into iconography: according to Pausanias (5.18.2), the chest that he dedicated at Olympia displayed a scene of Justice choking Injustice and striking her with a staff.^[53]

A similar image of archaic tyrants emerges from sources hostile to them. For example, the poet Theognis was preoccupied that his city, Megara, would give birth to an εὐθυντήρ, a man with exceptional powers who would ‘correct, straighten’ wickedness and injustice in the polis by ruling as a μόναρχος (Thgn. 39–52 W):^[54]

“Cyrnus, this city is pregnant and I fear that it will give birth to a man who will straighten (εὐθυντῆρα) our evil outrageousness (κακῆς ὕβριος). For these townsmen are still sound of mind (σαόφρονες), but the leaders (ἡγεμόνες) have changed and fell into much baseness. Good men (ἀγαθοί) have never destroyed any city, Cyrnus, but whenever it pleases the base ones (κακοῖσιν) to behave outrageously, they ruin the people and give justice to the unjust (δίκας τ’ ἀδίκοισι διδοῦσιν), for the sake of their own gain and power. Do not expect that that city will stay quiet for long – not even if now it lies in great tranquillity – whenever such thing shall become dear to the base ones, gain that comes with the people’s evil. For from these things arise civil strife (στάσιες) and internecine murders (ἔμφυλοι φόνοι ἀνδρῶν), and monarchs (μούναρχοί). May (all this) never please to this city.”

In this poem the word μόναρχος describes in a straightforward way, and in the Greeks’ own language, a figure which they used to call by a foreign term, tyrannos.^[55] Theognis makes it clear that the rise of a ‘monarch’ is contingent upon special circumstances, namely when the κακοί, for the sake of profit and power, give “justice to the unjust” and thus ruin the people.^[56] The opposition between ἀγαθοί and κακοί should not be intended as class conflict, but rather as a way to formulate elite competition in a context characterized by a high degree of social mobility, where ἀγαθοί and κακοί continuously swap places.^[57] As the leaders (ἡγεμόνες) turn to wickedness, elite conflict produces civil strife (στάσιες) and internecine murder (ἔμφυλοι φόνοι ἀνδρῶν). Monarchs, that is, tyrants, emerge to ‘straighten’ or ‘correct’ the situation, as a way out of injustice. As has been noted, the analogy between Theognis’s εὐθυντήρ and Cypselus, dreadful restorer of justice in Corinth, is obvious.^[58] Theognis makes it clear that the rise of a monarch is a possibility envisaged by the people – though an undesirable one, in his own view.^[59] While the city is ‘pregnant,’ the crucial factor is the people’s attitude: the leaders’ evil behavior creates the conditions for a monarch to rise, but the actual ‘birth’ of the monarch would not take place as long as the townsmen (ἀστοί) are sound of mind (σαόφρονες). The monarch’s rise would therefore be accompanied by the support of the people.^[60]

Solon gives us more insights into the circumstances which surround the emergence of a ‘monarch’ or ‘tyrant.’^[61] Not unlike Theognis, in his famous Eunomia poem (fr. 4 W, esp. ll. 5–29) he argues that the citizens’ foolishness and greed bring a city to ruin, and so do the “unjust mind” (ἄδικος νόος, l. 7), arrogance (ὕβρις, l. 8), and greed (ll. 11–13) of the people’s leaders. The combination of these factors triggers the punishment of Dike, Justice, which befalls the entire community: one of its consequences is that the city falls into “wretched slavery” (κακὴ δουλοσύνη, l. 18).^[62] A more concise, yet more explicit, treatment of this cause-effect chain is found in another fragment, where the slavery that befalls the community when leaders are wretched and the people lose their mind is described as that of a “monarch” (fr. 9 W):^[63]

“From a cloud comes the might of snow and hail, thunder is born from a bright lightning, a city is ruined by great men (ἀνδρῶν δ’ ἐκ μεγάλων), and for ignorance the people falls into slavery of a monarch (ἐς δὲ μονάρχου … δουλοσύνην). Having raised one too high, then it is not easy to restrain him afterwards. So now is the time to consider everything.”

While signaling the responsibilities of the elites (ἄνδρες μεγάλοι), Solon remarks, too, those of the people: it is they who raise someone too high and then fail to restrain his power.^[64] The support of the people is crucial to a tyrant’s rise to power.^[65] Actually, Solon himself had the opportunity to become a tyrant, but refused, as he himself says in one of his poems (fr. 32 W):^[66]

“If I spared (he says) my fatherland, and did not grasp tyranny and relentless violence (τυραννίδος δὲ καὶ βίης ἀμειλίχου οὐ καθηψάμην) ruining and dishonoring my reputation (μιάνας καὶ καταισχύνας κλέος), I am not ashamed: for this way I think I shall rather win everyone’s favor.”

Solon associates tyranny with violence, the ruin of the state, and the loss of personal reputation.^[67] In another poem, he gives voice to his critics and shows that for members of the elite tyranny was in fact an appealing prospect (fr. 33 W):^[68]

“Solon was not born a deep-minded man nor one of good counsel: for when the god gave him good gifts, he did not accept them. Having cast a big net around the prey, remaining in wonder, he did not pull it tight, as he was led astray in his heart and mind at once. For, if I were to gain power (κρατήσας), take endless wealth (πλοῦτον ἄφθονον), and rule Athens as a tyrant even for a single day (τυραννεύσας Ἀθηνέων μοῦνον ἡμέρην μίαν), afterwards I would be willing to be flayed like a wineskin and have my family destroyed.”

These poems should be read as a response to those (presumably unsatisfied political supporters)^[69] who criticized Solon for refusing to make full use of the special powers which were conferred on him. From his own poetry, as well as from the account of the Athenaion Politeia (5–13.1), we infer that Solon was offered special powers to solve conflicts regarding land distribution and the problem of debt.^[70] According to Plutarch, Solon was encouraged to take up such powers by a Delphic oracle, which recalls in tone and style the oracles that accompanied the rise of Cypselus at Corinth.^[71] Yet, while he fulfilled his tasks presenting himself as the restorer of order, justice, and good government (eunomia),^[72] Solon chose to remain in an institutional capacity envisaged by the existing constitution, that of archon, despite having the possibility of turning the powers that were offered to him into an opportunity to seize absolute power over Athens. To be sure, his archonship was not an ordinary one, as it was characterized by exceptional powers; however, Solon did not push them to their extreme consequences (οὐκ ἐπέσπασεν μέγα | δίκτυον, he “did not pull the big net,” as per fr. 33.3–4 W) and did not let them go unrestrained, which would have made him a tyrant. Instead, according to the Athenaion Politeia (5.2), he served as a διαλλακτής, a ‘mediator:’ this definition is consistent with Solon’s own claims of equity in his political activity,^[73] but we may suspect that the term was introduced by the author of the Athenaion Politeia precisely because it could appropriately summarize the spirit of Solon’s poems.^[74]

As we have seen, Theognis and Solon suggest that tyranny could result from the practice of appointing an individual endowed with exceptional qualities or skills to an extraordinary role within the polis. ‘Justice,’ or the individual’s ability to put an end to injustice, features prominently among the qualities of the appointee. This picture is further confirmed by the evidence on the rise of Pittakos at Mytilene, known as a tyrant but himself a lawgiver,^[75] one of the Seven Sages,^[76] and even considered the one who liberated Mytilene “from the three greatest evils: tyranny, civil strife, and war.”^[77] The picture of Pittakos drawn by one of his fiercest opponents, Alcaeus, goes in the opposite direction: a ‘devourer’ of Mytilene, Pittakos brought about civil strife and violence; and yet, he owed his glory to the people, though according to Alcaeus the latter had been driven to folly (Alk. fr. 70.5–6, 9–12 LP):^[78]

κῆνος δὲ παώθεις Ἀτρεΐδα[.].[

δαπτέτω πόλιν ὠς καὶ πεδὰ Μυρσί̣[λ]ω̣[

“But that one, having married into the Atreidai[’s house], let him devour the city as (he) also (did)^[79] with Myrsilos …”

χαλάσσομεν δὲ τὰς θυμοβόρω λύας

ἐμφύλω τε μάχας, τάν τις Ὀλυμπίων

ἔνωρσε, δᾶμον μὲν εἰς ἀυάταν ἄγων

Φιττάκωι δὲ δίδοις κῦδος ἐπήρ[ατ]ο̣ν̣.

“And let us relax from heart-devouring anger and internecine strife, which some of the Olympians aroused, driving the people to folly, but giving lovely glory to Pittakos.”

While the subject of the participle δίδοις in line 12 is τις Ὀλυμπίων, the underlying implication is that the people’s folly, which comes about as a consequence of civil strife, resulted in Pittakos’s glory, that is, in his rise to power.^[80] Not only is this interpretation consistent with the dynamics we observed in Theognis’s and Solon’s poems, but also, and most importantly, with the picture that emerges from Alcaeus’s fr. 348 LP, where the poet explicitly confirms that Pittakos was established and acclaimed by the Mytilenaeans as tyrannos of the city:^[81]

                                          … τὸν κακοπατρίδαν

Φίττακον πόλιος τὰς ἀχόλω καὶ βαρυδαίμονος

ἐστάσαντο τύραννον, μέγ’ ἐπαίνεντες ἀόλλεες

“[…] they established base-born Pittakos as tyrannos of the city, which lacks bile and is pressed by a heavy fate, all acclaiming him loudly.”

Based on this very fragment, Aristotle (pol. 3.1285a29–b1) cites Pittakos as an example of a form of ‘monarchy’ which – he maintains – existed among the early Greeks, that of the aisymnetai, ‘arbiters,’ which he interprets as an ‘elective tyranny’ (αἱρετὴ τυραννίς).^[82] Aristotle considers aisymneteia a fully legitimate form of rulership: he compares it with barbarian monarchy, arguing that the only difference between the two is that aisymneteia is not inherited; the legitimacy of both, however, is not in question, as both are compliant with the law.^[83] It is doubtful whether Aristotle had access to any sources that specifically called Pittakos an aisymnetes.^[84] The fact that Pittakos is the only example of aisymnetes that the philosopher is able to cite makes one suspect that he introduced this ambiguous category of ‘monarchy’ precisely to explain Pittakos’s ambiguous position. What Aristotle is trying to do is to make sense of the paradox, posed by Alcaeus’s fragment, that someone could be established as a tyrant (ἐστάσαντο τύραννον) with the people’s approval (μέγ’ ἐπαίνεντες ἀόλλεες): the concept of aisymneteia or ‘elective tyranny’ is his way to solve it.^[85] Aristotle must have been aware that in some parts of the Greek world certain civic officers held the title of aisymnetes,^[86] and that in some cases, such as in early 5th-century Teos, the appointment of an aisymnetes, to be understood as an officer with special powers, was considered a disgraceful possibility:^[87] establishing an analogy between the latter situation and a tyranny, namely an ‘elective’ one, must have been an easy step.^[88] Incidentally, one may wonder why Aristotle does not have a similar issue with Solon: indeed, Pittakos’s role is in all respects comparable to Solon’s role as archon-diallaktes (and also, for that matter, to what Theognis calls an εὐθυντήρ).^[89] The answer must be that Solon explicitly refused to be considered a tyrant, whereas Alcaeus, in Aristotle’s eyes, provides firm evidence that Pittakos was considered one. From the perspective of his categorizations, Aristotle was allowed to exclude Solon from his treatment of ‘monarchies’ or ‘tyrannies,’ but he had to include Pittakos. After all, Aristotle presumably had no other source on Solon than Solon himself, and one wonders how he would have categorized Pittakos if he had not had access to Alcaeus’s view of Pittakos but rather to Pittakos’s own voice.

As clearly emerges from the examples discussed so far, the Neo-Hittite and the early Greek world shared the same concern for justice as a key value in social and political life. Justice sets the standards of good rulership and of political legitimacy: most notably, the concern for justice underpins the legitimacy of tyrants and of political actors who called themselves tarrawanni-. To obtain the people’s support, tyrants and tarrawanninzi relied on the people’s belief, instilled through political rhetoric, that they were endowed with the quality of justice to an extraordinary degree.^[90] A rhetoric of divine legitimation, or divine ‘election,’ could add to their perceived extraordinary nature.^[91] However, it is important to note that, in both societies, standards of good rulership based on justice were advertised not only by those who were in power but also by those who were not: upholding justice was expected of political leaders, and adopting a rhetoric of justice in political discourse was also a means for political leaders to communicate that expectations were met. The recourse to a standardized rhetoric of justice ensured that the rulers and the ruled were, so to speak, all on the same page.^[92] This conclusion is important as it shows that in both Greek and Neo-Hittite societies there was political debate about what a good and legitimate political leader should look like. I emphasize this point because the idea of political debate taking place has normally been associated with Greek society, as one which leans towards collective government, rather than with Near Eastern societies, traditionally considered to be bound to autocratic government.^[93] By contrast, the concern with justice in political debate puts Greek and Neo-Hittite societies on a comparable scale. I shall return on this point later in this paper.

Although the link with justice that characterizes both tarrawanninzi and tyrannoi may seem to be a promising avenue to understand the origins of tyranny, the semantics of tarrawanni- (and thus the etymology of tyrannos) is not enough to explain why the Luwian term was borrowed into Greek and why the concept of tyrannos emerged in Greek political life. The Greeks had their own language of justice and needed no extra term or concept for it. Most notably, the concept of tyranny does not appear in the Homeric poems and in the works of Hesiod, although they attest that a lively debate on the role of justice in political leadership had been going on in the Greek world at least as early as the 8th century. Moreover, the cause-effect chain that links the ruler’s justice, the gods’ favor, and the land’s prosperity in Neo-Hittite political rhetoric appears to be quite close to the Homeric and Hesiodic notion of justice, which presupposes supernatural intervention when it comes to describing the consequences for the people when their leaders fail to uphold justice.^[94] But again, despite these analogies, the Neo-Hittite concept of tarrawanni- does not seem to have made its way into the world of Homer and Hesiod, where political leaders are called basileis. In sum, while it seems easy to associate the Neo-Hittite with the Homeric and Hesiodic notions of justice, including its connection to political leadership, the fact that tyranny is neither a Homeric nor a Hesiodic concept but only emerges in later evidence^[95] needs an explanation.

So, why, at some point, did autocrats emerge in Greece who were called after the Luwian word for ‘just, justice?’ A promising way to solve the impasse is to explore what kind of political practice, or political culture, tarrawanni- captures in the Neo-Hittite world that tyrannos captures in the Greek world, too. We therefore need to go beyond the surface of the rhetoric of justice and more in depth into the ways of doing politics of Neo-Hittite tarrawanninzi and Greek tyrants.

3. The Ambiguity of Ruling Alone

As we have seen, the epithet tarrawanni- was employed in various kinds of contexts where justice was perceived as a key element for legitimizing an individual’s position in society or their actions. In contexts more specifically political, the term was employed both by figures who held institutionalized positions of power, for example by kings, and by ones who exercised political power in their communities either in an institutional capacity that would normally not allow them to do so, or in no institutional capacity at all. In a world which had been dominated by precise notions of rule ranking (see further below), this distinction is particularly important. A king could exercise certain powers and claim specific prerogatives for the sheer fact of being a king, that is, by virtue of occupying an office, namely kingship, in which the legitimacy of his actions as king resided. If, by contrast, someone came to exercise the same or similar powers and/or to claim the same or similar prerogatives as kings while occupying another type of office or (even!) no office at all, then this poses a problem of legitimacy.

A situation like the one I just described points to a discrepancy between the community’s institutional arrangement – that is, the institutional capacity of the community’s power holders – and the actual type of political power that is exercised by the power holders themselves. Building on a suggestion of Franco Pintore, I will refer to political actors who find themselves in such situations as ‘para-institutional figures:’ this definition emphasizes their ability of exercising political power informally, de facto, without enjoying the institutional capacity to do so but as if they enjoyed it.^[96] In addition, I propose to refer to contexts in which political power is exercised by para-institutional figures as ones characterized by ‘institutional ambiguity,’ where the ‘ambiguity’ results from the dissonance between the institutional capacity of those who exercise power and the substance of the power they exercise. It is precisely para-institutional figures and the notion of institutional ambiguity that allow us to explain how tarrawanni- could become a new political concept capturing a specific way of doing politics. I will illustrate the point by discussing a selection of examples.

Some of the earliest attestations of the concept of tarrawanni-, from mid-10th century Karkamiš, cast light on its political significance in contexts of institutional ambiguity. Since the collapse of the Hittite empire around 1200 BC, Karkamiš had been ruled by a dynasty of ‘Great Kings’ who could claim descent from the Hittite king Suppiluliuma I.^[97] In the 10th century the royal dynasty coexisted with one of ‘Country Lords,’^[98] an institutional arrangement that reproduced a hierarchy of powers that went back to the imperial period: in the Hittite empire, Country Lords were figures of non-royal rank with administrative functions, who usually served as governors of provinces; while being subordinate to the king, they were often connected to the royal family by kinship ties.^[99] After the collapse of the empire, this distinction in rank and prerogatives between ‘Kings’ and ‘Country Lords’ was still in place at Karkamiš, and the same is true for Malatya further north, as the two cities were presumably united under the rule of (Great) King Kuzi-Tešub or his immediate descendants:^[100] the Country Lords were likely royal appointees in charge of administering the respective urban centers. At least for the 11th and part of the 10th century, this distinction in rank was also recognized at a supra-regional level: in the inscription ALEPPO 6 (§§ 4–12), the 11th-century king Taita I of P/Walastin(a/i) established ‘ranked’ offering regulations for the Storm God of Aleppo based on a hierarchy that clearly distinguishes ‘Kings’ from ‘Country Lords’ (as well as other figures);^[101] similarly, the curse section of the inscription MARAŞ 8 (§§ 13–16), authored by a mid-10th century ruler of Gurgum, Larama, recognizes the distinction in rank between ‘Kings’ and ‘Country Lords’ assigning to each a different prerogative.^[102]

As time passed, the situation evolved in both Karkamiš and Malatya, though in different directions: while at Malatya some of the Country Lords claimed the royal title for themselves,^[103] at Karkamiš they replaced the royal dynasty as the new ruling institution of the polity but never adopted the royal title.^[104] The process by which the Country Lords became the new rulers of Karkamiš was gradual and, most importantly for our purposes, characterized by an overt use of the rhetoric of justice, thus of the concept of tarrawanni-, in political discourse. The rhetoric of justice was employed as a mark of distinction by the first Country Lord of Karkamiš known to us, Suhi I, a contemporary of Great King Ura-Tarhunza in the first half of the 10th century. Suhi dedicated a monument commemorating Ura-Tarhunza’s victory in war against the land of Sura, signing it as follows (KARKAMIŠ N1, § 7):^[105]

|a-w[a/i]-tá || zi/a |stele ^Isu-hi-sa |ponere ^Imagnus+ra/i–tonitrus-sa rex bonus-sa-mi-sa || |*462-sa iudex-ní ka+ra/i-ka-mi-sà(urbs) |regio |dominus

“And this stele Suhi erected, dear kinsman(?) of King Ura-Tarhunza, the Just, Country Lord of the city Karkamiš.”^[106]

The dynasty of the Country Lords coexisted with that of the Kings for at least three generations under formalized arrangements likely sanctioned by intermarriage.^[107] However, the evidence shows that such coexistence resulted in a situation of institutional ambiguity: while the Kings represented the ruling institution of Karkamiš de iure, the Country Lords de facto began to exercise political power in the city. From the perspective of kingship, therefore, the Country Lords can be defined as figures of a para-institutional type who flanked the ruling institution and slowly encroached upon it.

Indeed, inscriptions commissioned by the Country Lords at a time when they were still formally subordinated to the Kings show that the title of ‘Country Lord’ slowly, and subtly, became a ‘mask’ for a power of a type different from that indicated by the title itself. For one thing, the building projects that the Country Lords commissioned for Karkamiš would have been hardly conceivable for Country Lords of the imperial period. Astuwalamanza, son of Suhi I, was directly involved in a (costly) building project for the erection of the city gates of Karkamiš (KARKAMIŠ A14b),^[108] while Suhi II, Astuwalamanza’s son, was the commissioner of another ambitious building project in the city, the Long Wall of Sculptures, ornamented with relief orthostats of ca. 2 meters high.^[109] But it is political rhetoric that reveals the new connotations that were subtly attached to the title of Country Lord. For example, Suhi II de facto appropriates elements of the political rhetoric of the kings, most notably the claim to divine support, the idea that political power was hereditary, and the belief that political authority was identifiable in the ruler’s persona and, at the same time, in the (inscribed) monuments erected by the ruler himself,^[110] as is testified to by the curse against whoever would erase Suhi’s name from the orthostats (KARKAMIŠ A14a).^[111]

Three generations after Suhi I, the para-institutional Country Lords finally took over power from the Great Kings and the dynasty of Country Lords replaced the royal dynasty as the rulers of Karkamiš: under Katuwa, son of Suhi II, the crisis of traditional monarchy becomes overt and the new Country Lord could claim to have “acquired” the city from the hands of the legitimate heirs to the throne, the grandsons of Ura-Tarhunza (KARKAMIŠ A11b+c).^[112] Katuwa’s rise to power is legitimized by the usual rhetoric of justice: in the inscription KARKAMIŠ A11a, Katuwa, who like his predecessors presents himself as “the Just, Country Lord of Karkamiš” (§ 1), emphasizes that his “justice” triggered divine support for him in the context of a revolt stirred by his “kinsmen,” in all likelihood to be identified with the grandsons of Ura-Tarhunza from whom he claimed to have “acquired” the city (§§ 4–8):^[113]

1. *a-wa/i-mu deus-ní-zi *a-mi^!?-ia-ti <“>iustitia”-wa/i-ní-ti pugnus.mi.la/i/u^[114] |pugnus-ri+i-ta

2. *a-mi-zi-pa-wa/i-mu-ta |20-tá-ti-zi arha crus+ra/i

3. [*a-wa/i-m]u-tá ⸢|⸣regio-ní-ia |*314(-)sá-pa-za |kwa/i-a-ti sub-na-na arha (pes₂)tara/i-za-nu-wa/i-tá

4. *a-wa/i-mu *a-mi-i-sa (dominus)na-ni || (deus)tonitrus-sa (deus)kar-hu-ha-sa (deus)ku+avis-pa-sa-ha *a-mi-ia-ti |“iustitia”-na-ti (lituus)á-za-tá

5. *a-wa/i-mu-tá á-ma |tá-ti-ia avus-ha-ti-ia |regio-ní-ia (*33(1))mi-tà-sa₅+ra/i-i-na kwa/i-a-ti a-tá i-zi-ia-tá

6. “Me the gods because of my justice raised strongly, but my kinsmen stood against me, and caused the lands to turn away from under me like a …(?). Me my lord Tarhunza, Karhuha, and Kubaba loved because of my justice, so that they made the lands of my father and great-grandfather(s) (vel ancestors) a reward for me.”

By that time, the Country Lords had acquired royal prerogatives and de facto exercised royal powers, though maintaining their institutional ‘façade’ of ‘Country Lords.’ The office of Country Lord, which traditionally came with administrative powers of a local scope, was repurposed as to uphold powers of a royal type without its trappings being altered: as far as we can tell, this new arrangement became permanent in Karkamiš.^[115] The Country Lords had found the way to increase their political power in the city without claiming for themselves the institutional title traditionally associated with the power they now wielded. As Pintore put it, this testifies to the separation of (factual) power from (institutional) authority: political power became distinct from the institution which formally conferred legitimate authority on it and could be obtained independently.^[116] Since the times of Suhi I, the political action of the Country Lords had been legitimized by an innovative rhetoric of justice, condensed in the epithet tarrawanni-: typically the first element in the Country Lords’ titularies, accompanying their political title, the epithet passed on from father to son as if it signaled a hereditary prerogative, flagging a new way of wielding power.

A similarly telling case of a para-institutional figure who claims to be a tarrawanni- is that of Saruwanni, dedicator of the 8th-century inscription ANDAVAL:^[117]

1. ego[… s]a-ru-[w]a/i-ni-sa iudex-wa/i-ni-sa na-hi-ti-ia-wa/i-ni-sá(urbs) dominus-ia-sa

2. wa/i-mu |wa/i+ra/i-[pa]-si-sá […-s]a ||[…]x

3. ⸢a⸣-wa/i (terra+la+la)wa/i-li-ri+i-tà-ti |rel arha (pes)u-sa-wa/i

4. a-wa/i |(equus)á-zú-wa/i-za za-ti la-pa-ni-wa/i

5. |wa/i-ma-tà x-[…]-ta[ …|| … ]x[…] arha […] (figure) wa/i+ra/i-pa-la-wa/i-ha [… || …] |magnus+ra/i-nu-[ …

6. “I am Saruwanni, the Just, Lord of Nahidiya. And me [Tarhunza(?)] of the Weapon [loved(?)].^[118] And when I bring (it) out of the plains, I pasture the horse-herd here. And (for) me it/they [… …] away […] (figure) and Warpalawa (acc.?) [… …] make/made great […”

Saruwanni’s institutional capacity is not immediately evident. The term with which he defines himself, niya-, ‘Lord,’ logographically rendered with the sign dominus, is never employed elsewhere as a self-standing ruling title, but is always found in compound titles like ‘Country Lord’ (regio.dominus), ‘River-Country-Lord’ (flumen.regio.dominus), and ‘River-Lord’ (flumen.dominus).^[119] The reduplicate form nanni-, also meaning ‘lord’ and logographically indicated by the sign dominus,^[120] is used as a generic way to indicate a king or ruler, to describe the relation between an ‘overlord’ and a ‘subordinate,’^[121] and to indicate the ‘master’ of an economic activity or a ‘house-lord,’^[122] but does not identify any institutional position. Saruwanni’s use of the term in combination with a geographic indication (Nahidiyawanni-, ‘of Nahidiya/Niğde’) is unparalleled in the corpus. Although Saruwanni is sometimes believed to be an early king of Tuwana, nothing recommends this interpretation.^[123] The relief accompanying the ANDAVAL inscription, which no doubt portrays the dedicator, shows no royal attributes, nor is the individual’s attire typical of a ruler (he is bare-headed and beardless).^[124] Another dedication authored by Saruwanni, NİĞDE 1, confirms the impression. The piece is a drum-shaped statue base with a simple linear decoration, and in the one-line text the dedicator presents himself with no political title at all.^[125] Saruwanni’s self-presentations suggest that he was simply a local dignitary. ‘Warlord’ is the term which defines him best. This emerges from ANDAVAL §§ 1–2, according to the restoration of I. Yakubovich: amu[… S]aruwannis tarrawannis Nahidiyawannis niyas | (a)=wa=mu war[pa]ssis [Tarhunza]s [azzatta], “I am Saruwanni, the Just, Lord of Nahidiya. And me [Tarhunza] of the Weapon [loved].”^[126] The mention of a deity who ‘loves’ (azza-) the commissioner immediately after the latter’s titulary is expected in Iron Age Luwian inscriptions. Yakubovich persuasively makes the case that the deity in question is a Storm-God of the weapon, based on the frequent association between the Storm-God and the divine ‘weapon’ bestowed on rulers at war.^[127] Saruwanni, a beloved of the Storm-God of the Weapon, rose to his position from military ranks, and his para-institutional appellative ‘Lord of Nahidiya’ suggests that his authority on Niğde was not institutionalized, but of an informal nature. The mention of Warpalawa in § 5 of the inscription, which presumably alludes to the king of Tuwana, suggests that Saruwanni had some relations with the local ruling dynasty: the circumstances point to a form of interaction between an institutional figure (the king) and a para-institutional one (the warlord) who claims power on a more local basis.

A final example I would like to discuss is that of Tarhunazza, ‘servant’ of king Warpalawa and commissioner of the 8th-century inscription BULGARMADEN:^[128]

1. á-mu-wa/i-mi-i |tonitrus-hu-na-(lituus)á-za-sá-’ |iudex-ni-sa |tonitrus-hu-wa/i+ra/i-*273-sa |(infans)ni-mu-wa/i-za-sá |wa/i+ra/i-pa-la-wa/i-si-sa |rex-ti-sa |heros-ti-i-sá |iudex-ni-sa servus-la/i-sa

2. |a-wa/i-ta |á-mi-i |dominus-ni-i || wa/i+ra/i-pa-la-wa/i-ia-’ |rex-ti-i |(“bonus”)wa/i-sà-za-ha

3. wa/i-mu-u (deus)mons-ti-na |mu-ti-na |pi-ia-ta

4. |wa/i-ma-na (deus)tonitrus-hu-za (deus)ku-avis-pa-pa-sa-ha |pa+ra/i-na arha |la+ra/i-ta

5. |wa/i-ta-’ |tara/i-zi-ha

6. |a-wa/i regio-ni-ha |prae-i |zí-ra/i-la-mi-i || |(“scalprum.argentum”)su-ha-pa-na-ti |ta-ta-ha |á-mi-ti |iudex-na-ti |á-mi-ia-ti-ha-’ |ha-tà-sà-tara/i-ma-ti

7. |“mára/i”-i-sa-pa-wa/i |(deus)cervus₃-ti-ia-sá-ti-i |(bonus)wa/i-sa₅+ra/i-ti-i |á-mi-i |dominus-ni-i ⸢(bestia)⸣hwa/i-sa₅+ra/i-’ |pa(+ra/i^?)-ti-i u-ta-ti na-ha-’

8. |á-mi-ha-wa/i-ta-’ |dominus-ni-na wa/i+ra/i-pa-la-wa/i-na-’ |w[a/i-s]u-u || u-sa-nu-sá-ha

9. |á-p[a]-sa-pa-wa/i-mu-u |(“asinus”)tara/i-ka-sa-ni-ia-za (*92)za-la-la |(“argentum.dare”)pi-ia-ta-’

10. |hwa/i-sa-pa-wa/i-ti-i mu-ti-ia (deus)mons-ti |ha-⸢zi⸣-ia-ni-sá-’ |⸢i-zi⸣-ia-ti-i

11. a-wa/i |tonitrus-hu-na-(lituus)á-za-sa-za-’ deus-na-za |“ovis”-ru-pi |sa₅-sa₅+ra/i-la-i |“annus”-na annus-na ||

12. wa/i-ru-ta |mu-ti-ia-wa/i-ni-zi deus-ni-zi |wa/i-su-u |pes-wa/i-i-<tu^?>

13. |za-pa-wa/i-ta-’ |“capere”-ma-z[a] hwa/i-sa |arha-’ |ha+ra/i-ri+i

14. |á-pa-⸢x⸣ caput-ti-na (deus)tonitrus-z[a^?] deus-ni-zi-ha |arha |“delere”-nú-tu

15. |(deus)luna+mi-pa-wa/i-na |ha+ra/i-tu

16. (deus)ni-ka-ru-ha-sa-pa-wa/i-na ar[ha] edere-t[u]

17. (deus)ku-pa-wa/i-na “xx”-tu-i

18. “I am Tarhunazza, the Just, son of Tarhuwarri, servant of king Warpalawa, the Hero, the Just. I was dear to my lord Warpalawa, the king. He gave me the divine Mountain Mudi. Tarhunza and Kubaba blessed it for me, and I took care (of it).^[129] And I administered also the land – a fertile and rich one – by my justice and my influence.^[130] By the favor of Runtiya of the open country I …ed wild beasts there(?) for my lord,^[131] and I pleased my lord Warpalawa well.^[132] And he sold^[133] to me carts for the mules. Whoever becomes mayor over the divine Mountain Mudi, let him offer a sheep year by year to the deities of Tarhunazza, and for him may the deities of Mudi come well. But whoever smashes this contract, that man may Tarhunza and the gods destroy! May Arma smash him! May Nikaruha devour him! May Kubaba … him!”

The inscription shows that the ‘servant’ shares attributes with his ‘overlord,’ the ‘para-institutional’ figure with the ‘institutional’ one: both Tarhunazza and Warpalawa are called ‘just.’^[134] In this case, we may imagine that the former’s ‘justice’ was a prerequisite for him to obtain from Warpalawa a position of power: the king gave him Mountain Mudi, with the surrounding territory, which Tarhunazza administered by his justice and influence. Tarhunazza’s appellative hudarli- (‘servant’) is not per se a political title but rather a generic marker of subordination: it does not qualify an institutionalized position but rather an informally-defined role which could be complemented, on occasion, by an institutional ‘façade,’ expressed by a proper political title. In the case of Tarhunazza this was presumably haziyanni-, ‘mayor,’ as can be inferred from § 10. This title evokes a situation of institutionalized subordination: it is a loanword from Akk. ḫazannu, the title adopted by the rulers of Late Bronze Age city-states in the Levant under Egyptian domination. From the perspective of the polity’s ruling institutions, represented by King Warpalawa, Tarhunazza is a para-institutional figure: the legitimacy of Tarhunazza’s position does not reside in a formally established institution, but in Tarhunazza’s moral qualities (justice) and in the good will of the king, who ‘alienated’ to him a share of royal property in exchange of other benefits: note that the relation between Warpalawa and Tarhunazza is defined a ‘contract, agreement’ (lalamman, § 13). Yet, Tarhunazza himself tries to institutionalize his para-institutional position. This is clear in the injunction of §§ 10–11: “Whoever becomes mayor over the divine Mountain Mudi, let him offer a sheep year by year to the deities of Tarhunazza.” The fact that future ‘mayors’ are bound to bring yearly offerings to the deities of Tarhunazza (!) suggests that Tarhunazza presents his appointment in that place as foundational: he binds future haziyanninzi to his own patron deities! Significantly, Tarhunazza legitimizes his position in the same way as the king, as if there were no qualitative difference between the two: ‘justice’ allows both king (the institution) and ‘servant’ (the para-institution) to exercise their powers, and the shared epithet tarrawanni- remarks that both of them are qualified power holders regardless of their specific institutional capacity.

The examples I discussed above show that in political contexts tarrawanni- was much more than an adjective-epithet and much more than just a flag of a moral quality: it was a tool of legitimation for a way of exercising political power outside institutional authority or beyond its limits. As a concept fluid enough to apply to a variety of political actors, whether holding institutional positions or not, tarrawanni- captures the notion of ‘power’ and detaches it from that of ‘institution,’ attaching it, instead, to the individual: in political language, being a tarrawanni-, a ‘just’ one, meant to be entitled to hold a position of power in virtue of one’s own personal qualities, regardless of whether and how that position was defined from an institutional perspective. At the same time, political actors who define themselves as tarrawanni- often try to provide their para-institutional position with an institutional ‘façade,’ either by repurposing existing political titles (‘Country Lord,’ ‘mayor’) or by endowing generic terminology with new meanings (‘lord’). These ‘façades’ can be interpreted as attempts to institutionalize situations of ‘institutional ambiguity’ – to make the ambiguity less ambiguous, as it were.

Not only does this framework help us better understand the political significance of tarrawanni-, but it allows us to understand the origins of the concept of tyranny. As has been pointed out, tyranny is fundamentally characterized by a low degree of institutionalization.^[135] The earliest tyrants obtained and exercised power in their poleis when civic institutions and the competences of public officers – if one can speak of ‘public officers’ at all – were overall poorly defined.^[136] For example, the constitutional framework that preceded Cypselus in Corinth, in spite of late sources claiming that the Bacchiads expressed first a basileus and then an annual prytanis, could hardly have been more than an informal arrangement where elite families had monopolized aspects of public life:^[137] Cypselus, in other words, did not find any established institutional framework to subvert or modify, but created a new form of power in a still undefined system.^[138] Other tyrants appear to have never substantially altered the institutional apparatus of the poleis over which they ruled, if there was one. On the contrary, they operated within existing constitutional frameworks and are depicted as respectful observers of existing laws.^[139] For example, this is how Aristotle characterizes the Orthagorids of Sicyon (pol. 5.1315b12–16):

“For the tyranny in Sicyon was the one which lasted for the longest time, that of the sons of Orthagoras and of Orthagoras himself: it lasted one hundred years. The reason for this is that they used to treat the subjects with measure (μετρίως) and in many respects they served the laws (τοῖς νόμοις ἐδούλευον).”^[140]

Elsewhere Aristotle says that Pittakos of Mytilene introduced new laws but did not modify the city’s constitution (pol. 2.1274b18–19): “Pittakos, too, was a maker of laws, but not of a constitution (ἐγένετο δὲ καὶ Πιττακὸς νόμων δημιουργὸς ἀλλ’οὐ πολιτείας).”^[141] This comment, which Aristotle makes in the context of his treatment of early Greek lawgivers, comes right after a statement about Drakon (pol. 2.1274b15–16): “There are laws of Drakon, but he established the laws for an existing constitution (Δράκοντος δὲ νόμοι μὲν εἰσί, πολιτείᾳ δ’ ὑπαρχούσῃ τοὺς νόμους ἔθηκεν).” Pittakos is associated (καὶ Πιττακός) with Drakon in that both introduced new laws within existing constitutional frameworks, leaving them unaltered. A fragment of Alcaeus, a fierce political adversary of Pittakos, indirectly seems to confirms this. Speaking from exile, Alcaeus is nostalgic about the assembly and council of Mytilene (fr. 130B.16–20 LP): ὀ τάλαις ἔγω | ζώω μοῖραν ἔχων ἀγροϊωτίκαν | ἰμέρρων ἀγόρας ἄκουσαι | καρ̣υ̣[ζο]μένας ὦγεσιλαΐδα | καὶ β̣[ό]λ̣λ̣ας … – “Wretched me, I live with a wild fate, longing to hear the assembly being summoned, Agesilaidas, and the council …” When read together with Aristotle’s testimony, Alcaeus’s statement seems to suggest that Pittakos had not abolished the institutions of the city, the assembly and the council, and that these preserved their prerogatives even under the tyranny.^[142]

Similar considerations apply to the Pisistratids. Speaking of Pisistratus’s tyranny, Herodotus (1.59.6) affirms that he ruled respecting Athens’s constitution:

“Hence Pisistratus ruled over the Athenians, without overturning the existing offices nor altering laws (οὔτε τιμὰς τὰς ἐούσας συνταράξας οὔτε θέσμια μεταλλάξας), and governed the city according to the established ones (ἐπί τε τοῖσι κατεστεῶσι), administering it fairly and well.”

The Athenaion Politeia (14.3 and 16.2) agrees on this judgment, affirming that Pisistratus ruled “in a way that befits the polis” (πολιτικῶς), rather than “in a tyrannical fashion” (τυραννικῶς): in the political theory of Aristotle’s school, this distinction is extremely meaningful, as tyranny is viewed in profound contrast with the world of the polis.^[143] Elsewhere, the Athenaion Politeia (16.8) says that “for the rest, he [Pisistratus] wanted to administer everything according to the laws (ἐβούλετο πάντα διοικεῖν κατὰ τοὺς νόμους).” Thucydides makes a similar comment on his successors (6.54.6):

“For the rest, the city itself used the laws already in place (τοῖς πρὶν κειμένοις νόμοις ἐχρῆτο), except as concerns the fact that they took care that always one of their own be in the offices (ἐν ταῖς ἀρχαῖς).”^[144]

A fragmentary Athenian archon list of the late 5th century adds details to the picture: it confirms that archonship was not abolished, but it seems to contradict Thucydides’s statement, as it appears that under the Pisistratids archons were elected from other prominent Athenian families, too, such as the Philaids (Miltiades) and the Alcmaeonids (Cleisthenes).^[145]

Although their position of power shows a low degree of institutionalization, tyrants do try to institutionalize it in the form of a monarchy. Several practices associated with tyrants go towards that direction: extending the tenure of special powers; transmitting special powers to descendants and establishing dynasties; resorting to violence as a means of monopolizing legal force. These practices are all symptomatic of the nature of tyranny and of its main problem: a form of political power which constantly needs strategies to institutionalize, or stabilize, itself, as it is exercised within poorly defined constitutional frameworks, or coexists with established constitutional frameworks while being at the same time external to them. From what we know, however, the process of institutionalization was never accomplished by any tyrant in the archaic Greek world: no tyranny was ever supported by a coherent institutional apparatus. Furthermore, dynasties of tyrants are typically very short, lasting two or three generations, and the death of the last representative puts an end to tyranny itself.^[146] There is no ‘vacant’ office of tyrant after that: tyranny is not a position which must be filled, and tyrannos never became a political-institutional title. It is a term which describes a position of power, not an institution.^[147] As far as we can tell, no tyrant ever called himself that way;^[148] we simply have no idea of what institutional title – if any – tyrants held during their rule: we may suppose that those who rose to power from offices belonging to the constitutional apparatus of the polis retained their previous title,^[149] and that those who were appointed to a special office held the relevant title.^[150] Others may just have had no title at all. Some tyrants claimed for themselves the title of basileus, to which the meaning ‘king’ was clearly attached:^[151] this can be seen as an attempt to provide an ambiguous form of monarchic power with the most straightforward institutional façade.^[152]

In sum, archaic tyranny is characterized by a high degree of what I proposed to call ‘institutional ambiguity:’ tyrants de facto exercise monarchic power in poorly institutionalized contexts or within pre-existing constitutional frameworks of a non-monarchic type without substantially modifying them; while they could formally serve in various institutional capacities, they place themselves outside the limits imposed by institutions; they can be described as para-institutional figures who continuously seek to institutionalize their para-institutional position.^[153] It is important to note that the para-institutional nature of tyrants as power holders has no bearing on their perceived legitimacy: as we have seen, evidence from archaic Greece suggests that appointing an individual to a position of a para-institutional nature was considered, notwithstanding the opposition of particular groups, a perfectly viable option under extreme circumstances.

Now, the Greeks knew no term in their own language to describe all this at once when a descriptor was needed. The title of basileus, which could have been, in principle, a good candidate for describing individuals who ruled alone, was inadequate at least for three reasons: first, because it evoked an early form of leadership which, as emerges from the Homeric poems and from Hesiod’s works, was perceived as unsatisfactory, resulting precisely in situations of injustice and discontent such as those that needed to be healed by a skilled ‘straightener;’^[154] secondly, because in the constitutional framework of archaic poleis the title basileus came to indicate accountable officers who shared power with other civic institutions;^[155] and thirdly, because a term which elsewhere indicated an established institution like ‘kingship’ could not appropriately define a form of rulership intrinsically characterized by a lack of institutionalization.^[156] The latter point is also the reason why any foreign word meaning ‘king’ was inadequate to the purpose: while kingship, as an institution, is a relatively static notion, tyranny is a non-institution and as such it is fluid, unstable, and mutable.^[157] It should now have become clear that Luwian tarrawanni- provided all that the Greeks needed to find a name for their autocrats. Tarrawanni-, a fluid concept that expressed standards of legitimacy and political behavior for rulers and non-rulers alike, was a marker for positions of power that, like those of tyrants, were poorly defined, and ‘ambiguous,’ from an institutional perspective.

Thus, the notions of institutional ambiguity and para-institutional figures provide a conceptual framework for understanding how and why Luwian tarrawanni- was borrowed into Greek and how the concept of tyranny came about. While this concept had already been in use for a few decades by his time, Solon offers – as far as we know – the first lucid conceptualization of the para-institutional nature of tyranny and its inclination to institutional ambiguity as a political problem. The danger that Solon sees in the practice of conferring special powers upon a single individual is that these powers, which included the right to introduce laws and reforms and potentially a monopoly on violence, could go unrestrained: the real problem, for Solon and those who, like him, considered a μόναρχος or τύραννος something undesirable, was that this practice set these individuals outside the reach of political accountability.^[158] A rising concern, and a matter of political debate, already in the world of Homer and Hesiod (see above), by the time of Solon political accountability was slowly finding its way into emerging polis institutions. In fact, it can be argued that notions of accountability underpin processes of institutionalization of public offices since their very inception.^[159] Before the emergence of formal examinations of the magistrates’ conduct (called euthynai in Athens),^[160] regulations on the tenure of offices can be considered first attempts to institutionalize positions of power according to a principle of accountability. One of the earliest examples is the so-called ‘constitutional law’ of Dreros, Crete (second half of the 7th century), which prohibits whoever has served as chief magistrate (kosmos) from serving again in the same capacity for ten years. The very fact that a norm was issued by the community to limit the iteration of the office of kosmos indicates that, however vaguely, the activity of chief magistrates had become institutionally subject to scrutiny: whether it was issued to prevent individuals from seizing absolute power or elite families from monopolizing public offices, the resolution suggests that in absence of regulations chief magistrates, or more generally top-ranking power holders, were prone to transcend expectations concerning the scope and limits of their powers.^[161] Viewed against efforts to institutionalize political accountability, which, between the 7th and 6th centuries, we may infer were being made in various parts of the Greek world,^[162] legitimizing the activity of para-institutional figures could easily be interpreted as a step back from the process: not only is the ‘tyrant,’ or ‘monarch,’ one who rules alone, but he is one who does not account for what he does, as he does not put himself (or is not put) in an institutionalized position to do so. Solon’s solution to preserve the principle of accountability envisioned exercising supreme powers in the framework of an established, and accountable, office of the state, the archonship. In Solon’s view, this solution would preserve the city’s own legal authority, which would not be alienated to a monarchos.^[163]

To be sure, Solon did much more than what several tyrants did in their careers, as he did modify Athens’ constitution.^[164] However, by exercising his powers within Athens’ institutional framework, he could carry out such a large reform of Athens’ constitution ‘from the inside,’ as it were: his refusal to become (or to be considered) a tyrant allowed him to present himself as someone who acted as part of the institutions, not outside them.^[165] Yet, Solon’s solution was a ‘façade’ one: from an institutional perspective, his position of power is just as ambiguous as that of any other tyrant, but the key difference is that he tried formally to solve the ambiguity by insisting on remaining an archon and by explicitly drawing a divide between his own solution and ‘tyranny.’ In other words, while in practice Solon’s solution does not set him apart from what the Greeks considered a ‘tyrant,’^[166] it aims to do so from a theoretical, conceptual perspective. Solon draws the divide between tyranny and himself in his poetry, which should in all respects be considered a form of political action:^[167] in his poems, he captures the problem of institutional ambiguity that was at the heart of the concept and practice of tyranny; as a solution to it, he creates a figure that could appear alternative to the ‘tyrant’ at least in form if not in substance.^[168] To push the argument further, one can maintain that Solon created the prototype of the ‘lawgiver’ – the figure who, in the retrospective view of most sources on archaic tyranny, appears as the ‘positive’ avatar of the tyrant.^[169]

4. Words, Concepts, and Practices

What I discussed in the previous section shows that the borrowing of tarrawanni- into Greek was, so to speak, selective: tyrannos captures neither the semantics of tarrawanni- nor its rhetorical potential to advertise ‘justice;’ rather, it captures its political function as a tool for legitimizing and identifying positions of power which lacked a coherent institutional apparatus. Tyrannos comes out of what tarrawanni- represents politically, not of what it signifies semantically. Yet, the borrowing could happen precisely because the way of doing politics of Anatolian tarrawanninzi and Greek autocrats in-need-of-a-name found its legitimation in justice: as I noted earlier, when taken out of context, the concern for justice shared by the two political cultures does not per se explain why tarrawanni- ended up in Greece as tyrannos; but when viewed in the hands of para-institutional figures and in contexts of institutional ambiguity, such concern turns out to be the catalyst of the borrowing. In other words, the political concept of tarrawanni- found in Greece a fertile soil for diffusion because the two political cultures were aligned along similar concerns, debates, and practices.

This conclusion seems to suggest that the Greeks learned a name and a concept for a political practice they had developed on their own and which was, by coincidence, similar to a practice that had developed in the Neo-Hittite world. But was there, too, any transmission of political practices from the Neo-Hittite world to Greece? To answer this question, a distinction should be made between the more standard practice of conferring special powers on a single individual, based on the belief in his extra-ordinary sense of justice, and practices that reveal attempts to turn such special arrangements into a proper institution in the shape of a monarchy. In retrospect, the Greeks themselves distinguished between the two by calling representatives of the former situation ‘lawgivers’ and representatives of the latter ‘tyrants,’ but the distinction had already emerged, as we have seen, by the time of Solon. The idea of legitimizing political action by means of a rhetoric of justice, which underpins the ‘tyrannical practice’ but is also at the heart of political reflection in the Homeric poems and in Hesiod, characterizes various societies of the Iron Age Eastern Mediterranean and Near East and can be considered a common development in a koiné of political thought and practice,^[170] fostered by a general reorientation in the criteria that sustained political legitimacy after the crisis of the Late Bronze Age palatial societies.^[171] In my approach to it, the koiné model is alternative to diffusionistic explanations, but also to models that posit similar independent, and thus unrelated, developments. Rather, it works at the intersection between the two options, emphasizing that similar starting circumstances foster a common ground that facilitates the circulation, exchange, and reception of ideas and practices and favors the development of shared forms of political action and thought, which emerge as common solutions for common problems.^[172] According to this model, it does not really make sense to say that the Greeks independently developed the idea that absolute power is legitimized through a rhetoric of justice, nor does it make sense to say that they imported it from elsewhere: rather, the idea, and the political practice associated with it, developed as part of this koiné in Greece just like in other societies, with distinctive traits in each of them. What the Greeks derived from the Neo-Hittite world is the means to conceptualize this practice, something that was made possible by the fact that the two societies were part of a common ground. Besides this, however, the Greek world may have been receptive of strategies oriented to the institutionalization of the informal power that is inherent to the ‘tyrannical practice:’ these strategies were deployed by both tarrawanninzi and tyrants with varying degrees of success, but it is significant to note that it is precisely the use of such strategies that, in retrospect, ended up distinguishing tyrants from lawgivers.^[173]

This scenario brings us to a final set of questions: who introduced the concept of tarrawanni- into Greece and named the tyrants ‘tyrants?’ To what extent were the Greeks aware of political practices in the Neo-Hittite world and of the political significance of concepts like tarrawanni-? What routes did the concept take to reach Greece, and around what time? On the one hand, as a political concept that describes a position of power but was not used as a title by power holders, it seems unlikely that the autocrats themselves imported it from the Neo-Hittite world. On the other hand, the earliest attestations of tyranny as a phenomenon in Archilochus (frr. 19 and 23 W)^[174] and Semonides (fr. 7.67–70 W),^[175] which are devoid of condemnation but, on the contrary, confirm that in popular imagination tyranny was considered appealing,^[176] suggest that the concept was not purposefully introduced by opponents of the ‘tyrannical practice’ either. The best way to formulate an answer to the question of ‘who’ brought the concept to Grece is probably that it was introduced as part of a political debate that involved not only the rulers but also the ruled – various players, whether winners or losers, who were experimenting with solutions to institutional issues that emerged as political communities formed and new institutions were created. Were they aware of what was going on in the Neo-Hittite polities? The answer is probably ‘yes, somehow,’ but not because they could observe or do research on how any individual tarrawanni- rose to power or ruled. Rather, such awareness comes, I would argue, from the existence of similar political debates about ways of doing politics in the Neo-Hittite world just like in the Greek one. The fact that a rhetoric of justice and the related tarrawanni- lexicon were widespread among a wide variety of socio-political players shows that the role of justice in society and its potential as a legitimizing force for new forms of power were matter of discussion, not only in the palaces, but also in the cities, among the people. It is in these debates that the Greeks found the concept of tyranny.

We can imagine several ways in which such debates could reach their ears, though it is difficult to decide which one is the most likely.^[177] In the last couple of decades, several studies have emphasized the role and importance of the Anatolian world in the realm of cultural interactions between the Aegean and the Near East.^[178] More specifically, the kingdoms of Hiyawa in Cilicia and of P/Walastin(a/i) in the ‘Amuq are prominent examples of intercultural zones that arose as the product of Aegean-Anatolian-Levantine interactions.^[179] The identification of Al Mina, the well-known ‘Aegean(izing)’ port of trade in the Levant, with the port of Ahtâ in the Neo-Hittite kingdom of Patin, heir of that of P/Walastin(a/i),^[180] adds to the picture of intense Aegean-Anatolian interactions,^[181] and so does the proposal that the Greeks learned the alphabet in the context of multilingual scribal schooling, for which the Luwian-Phoenician milieu of Cilicia represents the best candidate.^[182] In this context, it seems appropriate to note that scribes, as the ones who composed the pieces of rhetoric that we can now read in royal and other public or private inscriptions, were certainly among the most conversant with the implications of the tarrawanni- lexicon and were no doubt involved in the debates around it – as is suggested by the 8th-century funerary inscription of the scribe Ilali, remembered by his family precisely for his ‘justice.’^[183] In this regard, if we accept that Syro-Anatolian multilingual scribal workshops played a role in the transmission of the alphabet, it would not be far-fetched to imagine, too, that the Greeks may have been exposed to the debates about justice and politics in the Neo-Hittite world in the very same contexts.^[184]

If, however, an intermediary existed between the Neo-Hittite and the Greek world for the transmission of the concept of tarrawanni-, Phrygia or Lydia would be the most suitable candidates, in light of the relations that we know they entertained with the Greeks of western Asia Minor and of the mainland as well as with polities in the East.^[185] The route of diplomacy is certainly a fertile one for political ideas and terminology to circulate, and during the 8th and 7th centuries Phrygians and Lydians made full use of the land route that passes through Anatolia and connects the western, central, and eastern parts of the peninsula^[186] – for example when the Phrygians clashed with Great King Hartapu in the Konya Plain^[187] and established diplomatic relations with Warpalawa of Tuwana in southern Cappadocia,^[188] or when Lydian ambassadors went on that way up to Assyria.^[189] Pursuing the idea of diplomacy as a vehicle of political concepts, one may imagine that the Greeks acquired ‘second-hand’ knowledge of the concept of tarrawanni- via western Anatolian polities such as Phrygia or Lydia because Syro-Anatolian political actors employed the tarrawanni- rhetoric in diplomatic exchange with their western Anatolian interlocutors.^[190]

It must be emphasized, though, that the existence of an intermediary should be taken only as a possibility, not as a necessity. The evidence offers no firm basis for preferring the intermediary option to direct transmission, and this is especially true for the case of Lydia. Most notably, one should be wary of the misguided assumption, which is often found in previous as well as in recent scholarship, that the Greeks imported the concept of tyranny through Lydia because the earliest sources on the phenomenon associate it with Gyges.^[191] Archilochus’s fr. 19 W, which associates Gyges with wealth and tyranny, is taken as proof that Gyges was the prototype of the tyrant. The chronological coincidence between the emergence of tyrannies in the Greek world and the dynastic turn in Lydia, which culminates with Gyges murdering the previous king Kandaules, adds further scope to this reading, and so does the fact that Herodotus refers to Gyges by the word tyrannos or derivatives of it.^[192] All this is believed to find additional support in a fragment of Euphorion, who claims that Gyges “was the first to be called tyrannos.”^[193] However, none of the conclusions drawn from these sources is warranted. Firstly, whether or not Archilochus was specifically referring to Gyges when he speaks about tyrannis,^[194] the fragment simply suggests that in the imagination of 7th-century Greeks tyranny was something out of the ordinary, not that the archetypal tyranny was Lydian.^[195] Secondly, in his Lydian logoi, Herodotus does not consistently distinguish between basileus and tyrannos, for he uses both terms of Kandaules and his dynasty as well as of Gyges and his dynasty.^[196] Finally, Euphorion’s statement that Gyges “was the first to be called tyrannos” certainly derives from a combined reading of Herodotus and Archilochus and is thus historically unreliable.^[197] Therefore, although Lydia might have been an intermediary, there is no need to assume that it must have been so because Gyges features in Greek sources as a model of autocratic rulership.^[198] In fact, it may well be the case that Gyges came to be viewed as a ‘tyrant’ after the Greeks had already borrowed and developed this concept. Indeed, Gyges himself fully qualifies as a para-institutional figure, and stories about the rise of the dynasty establish a qualitative distinction between the two types of political power represented, respectively, by the Mermnads themselves (a ‘new’ power of a para-institutional nature) and by their predecessors (a ‘traditional’ power of a royal nature).^[199] According to the analysis carried out in this paper, the interpretation of Gyges as a ‘tyrant’ may therefore be viewed as the most natural way the Greeks could conceptualize the ‘Gyges phenomenon.’

As to the last unanswered question, that is, ‘when’ the concept of tarrawanni- found its way into Greece, the time period between the late 8th and the early 7th century seems to me to be the best chronological framework. The conditions for Anatolian political debates to reach the ears of the Greeks, which I outlined above, were most favorable in that time frame. This conclusion is not in conflict with the fact that our earliest testimonies of tyranny date to the mid-7th century. In fact, these testimonies allow pushing back of some decades the diffusion of the concept of tyranny in political debate. According to the 5th-century sophist Hippias, the word tyrannos, which he evidently recognizes as a foreign one, spread among the Greeks by the time of Archilochus (BNJ 6 F 6 = DK 86 B 7, apud Soph. Oid. T. argumentum). Hippias’s argument is based on textual research, for he could have no means to verify when the word entered the spoken language: as he could not find the word tyrannos in those who were considered the earliest Greek poets, Homer and Hesiod, but found it for the first time in Archilochus, he deduced that the word spread among the Greeks around the latter’s times.^[200] Hippias’s testimony coincides with what we moderns, too, know on this issue: Archilochus’s fragments (frr. 19 and 23 W) are, for us just as for Hippias, the first testimonies of the phenomenon. However, this does not allow us to exclude that the concept had already spread in the Greek world one or two generations before Archilochus in the spoken language or in contexts other than works of literature.^[201] In fact, Archilochus’s reflection on tyranny suggests that by his time the form of rulership associated with figures called ‘tyrants,’ the way it could be established, what it looked like, as well as its potential consequences, were already well-known in the Greek world and a matter of debate: indeed, in both of the relevant fragments the poet refers to the abstract concept of ‘tyranny,’ the ‘rule by a tyrant’ (called tyrannis in fr. 19 W and tyranniē in fr. 23 W), and not to the ‘tyrant’ as a ruler. The fact that by the time of the poet the rule of tyrants had already been conceptualized in abstract terms makes it unlikely that tyranny was by then a complete novelty or a very recent phenomenon.^[202]

One final piece of evidence seems to support the hypothesis that the conceptualization of tyranny occurred between the late 8th and the early 7th century. In the ‘Works and Days,’ while describing the injustice that characterizes the age of the Race of Iron, Hesiod says that in that age (his own) people “rather honor a doer of evil and an outrageous man” (Hes. erg. 191–192: μᾶλλον δὲ κακῶν ῥεκτῆρα καὶ ὕβριν | ἀνέρα τιμήσουσι). This statement, which resembles later ones in the poems of Theognis and Solon (see above), seems to allude to the practice of bestowing special powers on a single individual, which Hesiod, just like Theognis and Solon, views as controversial. As noted earlier, Hesiod never uses the word and concept of tyrannos, but it would not be far-fetched to imagine that that practice was being conceptualized in political debate precisely in the decades immediately following his poetic activity, namely when the issues that had been informing political reflection in the unstable world of Homeric and Hesiodic basileis were entering the emerging world of civic institutions. Among communities experimenting with new ways of organizing themselves politically, tyranny was part of the experiment.

5. Conclusions

This paper has addressed yet unsolved questions surrounding the emergence of tyranny in the archaic Greek world. Instead of examining tyranny solely from the perspective of Greek sources, most of which present a stereotyped account of it, I focused on the conceptual roots of the phenomenon by investigating the political significance and implications of Luwian tarrawanni-, ‘just, justice,’ the term from which Greek tyrannos derives. A shared concern for justice in political thought and practice emerged in both Neo-Hittite and Greek societies: Luwian tarrawanni- was widely employed by political actors as well as by other categories of individuals in a rhetoric of justice that advertised standards and expectations of good rulership, while in the Greek world a similar rhetoric lay behind the practice of assigning special powers to single individuals with the aim of putting an end to injustice and civil strife in poleis. However, the semantics of Luwian tarrawanni- cannot per se explain why the term was borrowed into Greek to indicate a form of sole rulership. Therefore, I investigated what specific way of doing politics is captured by the concept of tarrawanni- that is also captured by the concept of tyrannos. There emerged that the two concepts capture ways of wielding political power proper of ‘para-institutional’ figures and in contexts characterized by what I propose to call ‘institutional ambiguity:’ the two concepts describe political actors that exercised powers which they would formally have no institutional right to exercise within the constitutional arrangements in which they operated, but which they tried to present under some institutional ‘façade.’ The existence of similar political debates in the Neo-Hittite and the Greek world about what a good ruler should look like as well as about new ways of wielding political power irrespective of established institutional trappings favored the diffusion of the concept of tarrawanni- in Greece as tyrannos: it is in the alignment of those two worlds that we should look for tyranny’s deep roots.

Appendix: Linguistic Aspects of the Borrowing tarrawanni- > τύραννος

According to the most recent proposals, from a phonetic perspective, the word tarrawanni- may have passed into Greek according to one of the following sequences:

(1) tar(r)awannV- > *tarwannV- (syncope of the second syllable) > *tarunnV- (contraction -wa- > -u-) > turannV- (metathesis of vowels);^[203]

(2) tar(r)awannV- > *tarwannV- (syncope of the second syllable) > *trwannV- (syncope of unaccented [a], presupposing accent on the syllable -wán-)^[204] > *tr^wannV- (leftward labialization of [r]) > turannV- ([r^w] > [ur], cf. Old Irish ‘u-infection’);^[205]

(3) tar(r)awannV- > *tarwannV- (syncope of the second vowel, in Lydian) > *turwannV- (shift [a] > [u] triggered by following [w]) > turannV-; or from a putative original Luwian form *tǝrrawannV- > *tǝrwannV- (syncope of the second syllable, in Lydian) > turannV- (Greek rendering of Lyd. [r] + consonant, cf. Lyd. PN šrkaštus > Gr. Syrgastos or Syrgastor, epithet of Bithynian Zeus).^[206] Internal difficulties aside,^[207] both options posit Lydian as an intermediary, which is theoretically possible but not necessary (see above, § 4).

However, other possibilities can also be envisaged:

(4) tar(r)awannV- > tar(r)aunnV- (a syncopated form with contraction -wa- > -u- attested in the ablative tara/i-u-na-ti in KULULU 1, § 15)^[208] > *tarunnV (syncope of the second syllable) > turannV- (metathesis of vowels);

(5) tar(r)awannV- > *tar(r)unnV- (contraction -awa- > -u-) > turannV- (metathesis of vowels).

The contraction -awa- > -u- posited in the transmission sequence (5), which differs from the regular Luwian contraction -uwa- > -u-,^[209] is attested in a loanword from Luwian into Phrygian: the personal name Luw. Urawanni- > Phryg. Urunis (G-346).^[210] The metathesis posited in the sequences (1), (4), and (5) finds typological parallels in possible borrowings from other Anatolian languages into Phrygian: NPhryg. σαυναμαν (16.1 = 116) < (?) Hitt. samana-, ‘foundation(s), foundation deposit;’^[211] Phryg. PN siidos (G-105, HP-110, G-346) < (?) Sidetic PN śdis;^[212] Phryg. tesan (T-02b) < (?) Lyd. tasẽn.^[213] Compare also the Phrygian PN saragis, attested as Σαγαρις in Greek inscriptions and as sgry in Aramaic on a bulla from Daskyleion, for which a metathesis internal to Phrygian has been postulated.^[214] These parallels are only indicative of the plausibility of the phenomena involved and should not lead one to conclude that Phrygian need to be an intermediary language between Luwian and Greek; again, this is no more than a possibility.