• *  The oral version of this essay was presented in July 2008 at the University of Reading at the conf (...)
• 1  The classic definition of polis religion was provided by Sourvinou-Inwood (2000a) and (2000b), whi (...)
• 2 Sourvinou-Inwood (2000a), p. 22.
• 3 Sourvinou-Inwood (2000a), p. 27. Cult, of course, includes ritual.
• 4  Compare the comments of Woolf (1997), p. 76.

1For ancient Greeks, the polis was one of the fundamental units of their world and, as such, it provided an important context for religious activity, particularly for what has become known to modern scholars as polis religion.1 As Christiane Sourvinou-Inwood stated in her seminal study, ‘the Greek polis articulated religion and was itself articulated by it … Ritual reinforces group solidarity and this process is of fundamental importance in establishing and perpetuating civic and cultural, as well as religious, identities’.2 These dynamics were not limited to the city itself: ‘each significant grouping within the polis was articulated and given identity through cult’.3 As formulated here, the focus is on the group, rather than the individual, and this emphasis suggests the dominance of the single voice of the group. Indeed, the possibility of other positions is not discussed. If we wish to understand the dynamics of polis religion, we must ask, therefore, whether this system left room for alternative voices, for views which were not that of the dominant group, and to what extent it could accommodate multiplicity, issues which were not addressed by Sourvinou-Inwood in her examination of the religion of the city. Addressing these questions will also allow us to include some complexity in our understanding of the interrelationship between the city and its religion.4

• 5 Taylor (1991), p. 5-8; Garland (1992), p. 94-96; Parker (1996), p. 123, 136-137; see also e.g. Kear (...)

3Various scholars, including Michael Taylor, Robert Garland, and Robert Parker, have discussed the rituals for Harmodios and Aristogeiton; despite the small number of testimonia, there is general consensus that the Tyrannicides were, indeed, figures of cult.5 The occasion for the cult, however, has inspired rather less agreement. If we are to understand how the rituals served to articulate the city and its various identities, then we must place them in their correct setting and context. As we shall see, the rituals first seem to have been instituted at the end of the sixth century B.C., when the well-known statues were erected, and they took place annually at the Panathenaia, the occasion of the Tyrannicides’ deed and of Harmodios’ death.

• 6  Aristotle, Athenaion Politeia, 58, 1 (Chambers), repeated by Pollux, Onomastikon VIII, 91; Demosth (...)
• 7  As, for example, Pelops at Olympia (scholia on Pindar, Olympian, I, 149a; cf. Pindar, Olympian I, (...)
• 8  Honours equal to gods: Habicht (1970), p. 195-205, 212; Fishwick (1987), p. 21-31; Chaniotis (2007 (...)
• 9 Ekroth (2002), p. 74-128.
• 10 Ekroth (2002), p. 89, 127-128.
• 11 Ekroth (2002), p. 88-89, 96-98, 126-127.
• 12  Victim type: Ekroth (2002), p. 170.
• 13  Cicero, Pro Milone, 80.

4The clearest evidence for the cult of Harmodios and Aristogeiton comes from the third quarter of the fourth century. At this time, they received annual enagismata from the polemarchos and ‘libations and wine-offerings at the sacrifices in all the sanctuaries’; they were also celebrated in song and honoured (τιμᾶτε) equally with the heroes and gods.6 Multiple ritual occasions seem to be described here because the enagismata should be separate from the libations and wine-offerings. The libations at sacrifices find parallels in occasions when offerings for heroes accompanied offerings to specific divinities.7 The enagismata should map on to the honours equal to those of heroes and gods because honours equal to those of gods are usually understood as sacrifice and other attendant rituals, while the words τιμάω and τιμή, when used in connection with heroes (and gods), identify them as recipients of cult.8 As Gunnel Ekroth has shown, the term enagismata in connection with heroes or individuals other than the ordinary dead is rare at this time.9 Such offerings were completely destroyed; heroes typically received an animal victim, while the ordinary dead were given non-fluid offerings such as cakes, fruits, and prepared food.10 Enagismata particularly emphasised the dead state of the recipients and were very often connected with individuals who died violently.11 As such, they were particularly appropriate for Harmodios and Aristogeiton who probably received some sort of animal victim.12 In the middle of the first century B.C., Cicero describes the honours which Greek cities awarded to tyrant-slayers: although he does not mention Harmodios and Aristogeiton by name, he locates the honours at Athens among other cities and recalls the ‘divine rites’, the cantus, and the carmina.13 These men, he states, ‘are deified almost to the sanctity and memory of immortality’ (prope ad immortalitatis et religionem et memoriam). The invocation of Athens points towards Harmodios and Aristogeiton as among the tyrant-slayers honoured in this fashion and the rituals described find good parallels in Demosthenes’ speech.

• 14  E.g. Mommsen (1898), p. 302-303, 307; Deubner (1932), p. 230; Calabi Limentani (1976), p. 11-12, 2 (...)
• 15  Aristotle, Athenaion Politeia, 58, 1: ὁ δὲ πολέμαρχος θύει μὲν θυσίας τήν τε τῆι Ἀρτέμιδι τῆι ἀγρο (...)
• 16  Above note 6.
• 17 IG I³, 523-525: ἆθλα ἐπὶ τοῖς ἐν τōι πολέμοι; Vanderpool (1969), p. 4 note 7.

5While there is general agreement that the Tyrannicides were the recipient of a cult, the occasion for the enagismata, the primary manifestation of the cult, has proved difficult to identify securely. Scholars have frequently associated these honours with the Epitaphia.14 Doing so requires us to rely on the text of the Athenaion Politeia as transmitted in the papyrus, but this version was unknown to Pollux, who presented the text in a slightly different form: for him, the games for the war-dead were a separate occasion from the enagismata for Harmodios and Aristogeiton.15 Recent editors have accepted his version of the text.16 Furthermore, the known prizes awarded to victors in the Epitaphia specify that they are ‘for those who (died) in the war’ and they do not mention Harmodios and Aristogeiton.17 There is, accordingly, no evidence to connect the rituals for the Tyrannicides with the games for the war-dead and their enagismata were almost certainly not offered on that occasion.

• 19  Note, however, that the same verb is used in Demosthenes, 19, 280.
• 20  Philostratos, Vita Apollonii I, 14, 1; 30, 1; III, 14, 3; 17, 2; cf. V, 42, 2.
• 21  Philostratos, Vita Apollonii V, 8, 1: εὐαγέλια θύειν τρισολυμπιονίκην Νέρωνα ᾄδοντας.

7Although forms of the verb to sing (ἀείδω) are used in a variety of ways in the Life of Apollonios, there is no exact parallel for this passage.19 A number of examples concern the singing of songs or hymns in a ritual setting, a context analogous to the situation at the Panathenaia.20 The closest parallel, however, is provided by the people of Gadeira (modern Cadiz) who are ordered ‘to make a sacrifice of thanksgiving for good news, celebrating in song Nero as a triple Olympic victor’.21 At this ritual occasion, the songs make the reason for the sacrifice explicit, while they also commemorate the emperor’s achievements. Similarly, at the festival of the Panathenaia, Harmodios’ and Aristogeiton’s deed is also the subject of song. The biggest difference between the two passages is in the nature of the occasion: the Panathenaia was held regularly, but the sacrifice for Nero’s victory was a unique event.

• 22  Compare Philostratos, Vita Apollonii, V, 34, 3; VIII, 16 (Panathenaia as occasion of their deed).
• 23  The Panathenaia is independently attested at this time by a number of inscriptions: e.g. SEG XLIII (...)

8Furthermore, in the first passage, Apollonios could encourage the governors without mentioning the Panathenaia or the songs; instead, Philostratos has deliberately chosen to juxtapose the festival, the songs, and the men.22 The association should indicate that the Panathenaia was the occasion for their cult which ought to have been well known to Philostratos’ readers; if the rituals were obscure or had lapsed, then Apollonios’ point would have been lost.23 The songs described here also map on to those mentioned by Demosthenes and Cicero. This passage, accordingly, provides the context for the rituals and it indicates that they continued to be part of the festival in the early third century A.D. The Panathenaia is a particularly appropriate venue for the cult because it was the occasion of the Tyrannicides’ deed and Harmodios’ death. Adding the annual cult to the festival maintained precisely these connections.

• 24  Harmodios skolia PMG nos. 894, 896 = Athenaeus, XV, 695a-b, nos. 11, 13. For a parallel, see Simon (...)
• 25  E.g. Ostwald (1969), p. 121-136; Fornara (1970), p. 158, 178-180; Brunnsåker (1971), p. 23-24, 121 (...)
• 26  Pliny, Historia Naturalis XXXIV, 17.
• 27 Polybios and Cicero: Walbank (1957), p. 340, 665-669 with further references; cf. Castriota (1998), (...)
• 28  Creation of Agora: Shear, Jr. (1994), p. 228-245 with further bibliography; Shear, Jr. (1993), p.  (...)

9The Athenians probably instituted the cult not long after the events. The Harmodios skolia refer to the kleos of Harmodios and Aristogeiton and the Isles of the Blest, which are inhabited not only by Harmodios, but also by swift-footed Achilles and Diomedes; the Homeric diction marks the equation of the Tyrannicides with Homeric heroes, many of whom had cults.24 If these songs do go back to the end of the sixth century, as has been suggested by various scholars,25 then they would point towards an early date for the start of the rituals. Further evidence is provided by the bronze statues made by Antenor and set up in the Agora. According to Pliny, they were erected in the same year as the expulsion of the kings from Rome.26 While Varro placed this event in 510/9, Polybios and Cicero located it in 508/7 and Cornelius Nepos put it in 507/6.27 These alternative dates suggest that the statues of the Tyrannicides were erected in the Agora in 507 when they would have been part of the project to erect the first public buildings and to make the area fully public space as the city’s Agora.28

• 29  Konon and Euagoras: Demosthenes, 20, 69-70; Isokrates, 9, 56-57; Pausanias, I, 3, 2-3; Shear (2007 (...)
• 30 Ma (2007), p. 203; Dillon (2010), p. 12-13.
• 31  Herodotos, V, 77, 3-4; IG I³, 501A.
• 32  Herodotos, I, 31, 5. Whether the pair of early kouroi found at Delphi (Delphi Inv. 1524, 467) repr (...)
• 33  Herodotos, I, 30, 5; Nagy (1990), p. 244; cf. Currie (2005), p. 144. This example raises the quest (...)
• 34  Pausanias, VIII, 40, 1-2; Krumeich (1997), p. 202 with further bibliography.
• 35 Shear (2007a), p. 109. On the phenomenon, see Currie (2005), p. 120-157 with further references.

10These statues, however, were anomalies: public honorary statues did not exist in Athens until 394/3 when the Athenians awarded bronze figures to Konon and Euagoras of Cypriot Salamis for their role in the naval victory at Knidos;29 beyond Attica, such statues appear to be a fourth-century development.30 On the late archaic Akropolis, the only (vaguely) similar monument is the bronze chariot celebrating the victories over the Boiotians and Chalkidians in 506, but it apparently did not include human figures.31 Otherwise, the monuments most comparable to the Tyrannicides are the statues of Kleobis and Biton and of Arrachion. Herodotos tells us that the Argives dedicated the figures of Kleobis and Biton at Delphi, again a sanctuary context.32 The two young men may also have been the recipients of cult, as was Tellos the Athenian with whom Hero­dotos links them.33 Even in sacred settings, however, public figural monuments were rare in the late archaic period. The archaic statue of Arrachion was seen by Pausanias in the Agora at Phigalia.34 Since Arrachion died winning his third Olympic victory in the pankration, he was probably the recipient of a city cult as a victorious athlete.35 The best parallel for Antenor’s group, consequently, is the statue of a divinity or hero located in sacred space, a connection presupposing the existence of a cult when the figures were set up in the Agora.

• 36  For an early date for the institution of the cult, see also e.g. Fornara (1970), p. 157; Taylor (1 (...)

11Collectively, our evidence suggests that the institution of the Tyrannicides’ cult and the erection of their statues belong together in 507 B.C. when they must have been approved by the demos of the Athenians.36 The figures will also have been part of the larger project to construct public buildings north of the Areiopagos and to create a new Agora for the Athenians. The primary occasion for their rituals was the Panathenaia when the polemarchos made the enagismata and songs were sung for them. These rites continued to be carried out as late as the early third century A.D. By the 340s B.C., the Tyrannicides also received libations at all the sacrifices. Together the rituals and the statues will have kept the Tyrannicides’ deed ever present in the minds of the Athenians.

• 37  Harmodios skolia, PMG, nos. 893, 896 = Athenaeus, XV, 695a-b, nos. 10, 13.
• 38  Harmodios skolia, PMG, no. 895 = Athenaeus, XV, 695a-b, no. 12.
• 39 Raaflaub (2000), p. 253-254, 264; Raaflaub (2004), p. 58-102, 117.
• 40  Aristophanes, Acharnians, 979-981, 1094; Wasps, 1225; Lysistrata, 632-634; fr. 444 (PCG).
• 41  Since Furley and Bremer have described the first four skolia in Athenaeus’ collection as ‘hymnic’ (...)
• 42 Furley and Bremer (2001), p. 19-20.

13Already in the Harmodios skolia, Harmodios and Aristogeiton are described as killing the tyrant and making Athens isonomos, a place of equal rights.37 In this version, the death of the tyrant leads directly to isonomia for Athens, as if no time at all had elapsed between these two events in 514 and 508. In the third poem, the tyrant is specifically named as Hipparchos.38 Significantly, the term freedom, eleutheria, does not appear in these texts. As Kurt Raaflaub has shown, this noun and the concept of freedom as a political value developed only in the years immediately after 480 in the aftermath of the Persian Wars.39 For these reasons, the poems very probably go back to the period before 480 when they must have been in fairly wide circulation, hence their preservation and Aristophanes’ references to them in the later fifth century.40 The sentiments expressed are appropriate both for men supporting the new Kleisthenic political system and also for the city’s rituals for Harmodios and Aristogeiton. The likely source for this depiction of the events would appear to be the cult of the Tyrannicides and particularly its songs.41 This association would explain the focus of the skolia on slaying the tyrant and bringing equal rights to Athens because hymns for heroes often concentrate on the crisis leading to the individual’s status as a figure of cult.42

• 43  Antenor’s group: Pliny, Historia Naturalis XXXIV, 70; Arrian, Anabasis III, 16, 7-8; VII, 19, 2; V (...)
• 44 IG I³, 502; Hephaistion, Enchiridion, 4, 16 = Simonides fr. 131 Bergk (fr. 1, Page, FGE). I would s (...)
• 45  ἰσόνομον: Peek in SEG X, 320; cf. Raaflaub (2003), p. 64; ἐν ἐλευθερίαι: Friedländer and Hoffleit (...)
• 46  Aischylos, Choephoroi, 809-810, 863, cf. 1046; Raaflaub (2003), p. 64.
• 47  Herodotos, VI, 123, 2; Thucydides, VI, 53, 3; 54, 3; 56, 3; on the last two passages, see Hornblow (...)
• 48  Herodotos, VI, 109, 3; cf. 109, 6.
• 49  Aristophanes, Lysistrata, 630-635.
• 50  Andokides, 1, 96-98.
• 51  The eldest male descendants of Harmodios and Aristogeiton received sitesis, proedria, and ateleia; (...)
• 52  There is no support, accordingly, for Thomas’ claim that this role for the Tyrannicides belongs to (...)

14Harmodios and Aristogeiton, however, did not remain simply the killers of the tyrant and the bringers of isonomia; with the invention of political freedom, they also became the bringers of freedom. This version may have been reflected on the now fragmentary base for the second statue group made by Kritios and Nesiotes to replace Antenor’s earlier group and erected in 477/6.43 The poem says that Aristogeiton and Harmodios killed Hipparchos and also that they did something for the fatherland.44 Since the crucial section of the epigram is not preserved, we do not know exactly what these two men did and both freedom and isonomia have been restored in this section.45 The parallel between the great light born from their deed in the first line of the epigram and Aischylos’ use of the same metaphor in the Choephoroi shows that Harmodios and Aristogeiton could already have been understood as bringers of freedom in the 470s.46 Certainly, both Herodotos and Thucydides knew this tradition.47 The equation is particularly clear in Herodotos’ report of Miltiades’ speech to Kallimachos on the eve of the Battle of Marathon. By way of encouragement, Miltiades offers the polemarchos a choice: he has the opportunity either to enslave Athens or to make her free and to leave a memorial greater than that of the Tyrannicides.48 Harmodios and Aristogeiton provide an apt comparison because they, too, are bringers of freedom. They are so depicted in other late fifth-century texts. In Aristophanes’ Lysistrata, the leader of the men’s chorus equates resisting tyranny with carrying a myrtle branch and the statue of Aristogeiton in the Agora which he will imitate.49 These connections emphasise the Tyrannicide’s role as someone who frees the city from tyranny. In 410/9, Demophantos’ decree specifying how the Athenians should act if the democracy were overthrown in the future speci­fically invokes Harmodios and Aristogeiton in the oath taken by all the Athenians.50 They are to kill with impunity anyone who overthrows the democracy and anyone who dies in the process and his sons will receive the Tyrannicides’ benefits.51 Here, Harmodios and Aristogeiton are specifically associated with democracy and freedom and the Athenians swear to be(come) tyrannicides. By the late fifth century, accordingly, Harmodios and Aristogeiton were certainly associated not only with the end of the tyranny, but also with the bringing of freedom to the city.52 This role likely goes back to the 470s and the period immediately after the Persian Wars.

• 53  Demosthenes, 20, 69-70.
• 54  Demosthenes, 19, 280. On Epikrates, see MacDowell (2000), p. 323-324. For other fourth-century exa (...)
• 55  Above note 18.

15After the end of the fifth century, the image of the Tyrannicides as liberators is, of course, common and well known. When Demosthenes came to speak against Leptines, he emphasised that Konon was the first man after Harmodios and Aristogeiton to receive a bronze statue because his contemporaries thought that ‘he, too, had ended no small tyranny in breaking up the Spartan empire’; on the stele with the decree, they made the same point by stating that he had ‘freed the allies of Athens’.53 As the juxtaposition indicates, both Konon and the Tyrannicides brought freedom to the Athenians after tyranny. In his speech on the false embassy, Demosthenes associated Harmodios with Thrasyboulos of Steiria, who led the demos back from Phyle in 403, and Epikrates, who was active in the Peiraieus in the same year.54 In this way, the men who overthrew the Thirty and restored freedom and democracy to the city are connected with the man who freed Athens from the tyranny of the Peisistratidai. In the early third century A.D., Philostratos could still equate the Tyrannicides, freedom, and the return from Phyle.55 Despite the intervening centuries, the associations had lost none of their potency and Harmodios and Aristogeiton remained the liberators of Athens. Both Demosthenes and Philostratos link this tradition with contexts mentioning the cult, an association which suggests that this version was being disseminated through the rituals for the Tyrannicides. Such a setting with its annual repetition would further explain why this tradition was so dominate for so long.

• 56 They ought to have been already associated when the Athenians decided to locate the new cult of Zeu (...)

16According to our evidence, the cult for Harmodios and Aristogeiton seems to have promulgated a very specific version of their deed and the events surrounding the assassination of Hipparchos. The killing of the tyrant by these two men led directly and immediately to the foundation of the democracy. This tradition set the rule of the demos in opposition to tyranny and, after the Persian Wars, associated it firmly with freedom, which the Tyrannicides brought to the city. Indeed, the cult may have been instrumental in causing the Athenians to link democracy and civic liberty.56 Together, the cult and the statue group in the Agora continually kept this foundational moment present for the Athenians and they stressed that keeping Athens democratic might cost a citizen his very life.

• 57 Herodotos, V, 55; 66, 1; 69, 1-73, 1; Aristotle, Athenaion Politeia, 19, 1-2; 20, 1-21, 2; Thucydid (...)
• 58 Spartans: Herodotos, V, 62, 1-65, 2; Aristotle, Athenaion Politeia, 19, 2-6; Thucydides, VI, 53, 3; (...)
• 59  Alkmeonidai: Herodotos, V, 62, 2-63, 2; VI, 123, 1-2; Aristotle, Athenaion Politeia, 19, 3-4; Thuc (...)
• 60 Davies (1971), p. 472-474. Other members of these families are attested from the middle of the fift (...)
• 61  Compare Castriota (1998), p. 202; Rausch (1999), p. 47. Oikistai and agora: e.g. Pindar, Pythian, (...)

17For the polis in the last decade of the sixth century, this imagery would have been of particular benefit because it provided an authorised account, as it were, of how the isonomic system came into being. This version overlooked the four years of tyranny after Hipparchos’ assassination and the stasis which followed the overthrow of Hippias.57 It also eliminated the Spartan participation which actually allowed the Athenians to remove Hippias so that the events were (re)constructed as purely an Athenian affair.58 Similarly, the role of the Alkmeonidai was eliminated so that one family was not favoured at the expense of others, a situation which might then cause further disunity.59 Since the men singled out for honour were conveniently dead, they could not make special demands. Their families do not seem to have been particularly prominent at this time60 and, therefore, the institution of the cult would not have given them the power to destabilise the city. The cult and the statues in the Agora recall the honours given to oikistai of colonies, particularly their monuments in agorai and their rituals, so that the Tyrannicides are configured as the founders of the new political order and of a new start for the city.61 Through the cult, the city is articulated in a completely new way: now it is isonomos and tyranny has been banished.

• 62 Ober (1996), p. 32-52; Ober (2007).
• 63  For discussion of the dynamics involved and some further examples, see Shear (2011), p. 15-16, 135 (...)

18Of all the events between the killing of Hipparchos in 514 and the institution of the rule of the demos in 508/7, the assassination was perhaps the only one which all Athenians could (potentially) share. If, as Josh Ober has argued, the spontaneous uprising of the Athenians against the Spartans and Isagoras was the crucial moment for the creation of the Kleisthenic democracy, then this event would not have been easy to commemorate.62 The actual implementation of the new system would have required a series of decisions made over a period of time, a situation which precludes celebration within regularly repeated ritual. In contrast, commemorating the actions of the Tyrannicides involved none of these complications. Celebrating in ritual the story of how Athens came to be ruled by the demos both created and reinforced unity among the Athenians and gave them a moment which they could agree to commemorate together.63 The disunity which marked the years after the overthrow of Hippias would not be allowed to continue.

• 64  Above note 48.
• 65  Above note 49. According to Sommerstein and Ober, when the male chorus leader said ‘I’ll stand bes (...)
• 66  Above note 50.
• 67  See further Shear (2007b) and Shear (2011), p. 75, 96-106, 136-141, 147, 160-161.

19These processes of commemoration also established a paradigm for the good Athenian citizen: he was a democrat modelling himself on the Tyrannicides. Our fifth-century texts demonstrate the force of this model. As we have seen, in Herodotos’ narrative of the Battle of Marathon, Miltiades invokes the Tyrannicides to encourage Kallimachos to cast to the deciding vote to engage with the Persians; like them, Kallimachos can free Athens.64 Here, Harmodios and Aristogeiton show how a good Athenian ought to behave and Kallimachos ought to model himself on them. He does, indeed, do so and the Athenians defeat the Persians, while Kallimachos himself gains lasting renown. In Lysistrata, the leader of the men’s chorus is going to stand up to the tyranny of the women by imitating the example of Aristogeiton.65 Doing so requires him to act against tyrants and his conduct identifies him as a proper democratic Athenian. Swearing the oath provided in Demophantos’ decree makes all the Athenians into potential tyrant slayers who kill men overthrowing the democracy.66 In this way, it delineates the appropriate behaviour for citizens who are democrats and act against those overthrowing the rule of the demos.67 In all three of these cases, the Tyrannicides explicitly serve as models for Athenian men. For those individuals who were not committed democrats, it would have been a rigid definition of what it meant to be a citizen of the city.

• 68 Thompson and Wycherley (1972), p. 157-158; Ajootian (1998), p. 3-7. In the winter of 2011, rescue e (...)
• 69  Marathon painting: Pausanias, I, 15, 1, 3.
• 70 Lysistrata: Gomme, Andrewes, and Dover (1981), p. 184-190, 193; Sommerstein (1977); Henderson (1987 (...)
• 71  On the oath and the creation of Athenian unity, see Shear (2007b), p. 158-159; Shear (2011), p. 13 (...)

20These three examples use the imagery promulgated by the rituals in other civic and cultural contexts which then serve to perpetuate it. The dynamics of Herodotos’ speech are particularly striking because, from the middle of the fifth century, they were mirrored in the topography and monuments of the Agora. Harmodios’ and Aristogeiton’s statues stood somewhere in the centre of the square, probably just west of the Panathenaic Way (Fig. 1).68 To the north, the Stoa Poikile with its painting of the Battle of Marathon faced south on to the Agora (Fig. 2).69 Viewers, particularly those standing on the stoa’s steps, were able to look from one monument to the other and to see both the inspiring models of the Tyrannicides and the results of imitating them in the victory celebrated by the painting. With Lysistrata and Demophantos’ decree, the contexts also involved rituals, but not ones connected with the Tyrannicides: the play is thought to have been performed at the Lenaia of 411, while the oath of Demophantos was sworn by the Athenians shortly before the Dionysia of 409.70 This oath shows the potency of the Tyrannicides: in the aftermath of the oligarchies of 411, they were instrumental in bringing the divided city back together again.71 At the moment when democracy’s continuing existence had been brought into question, invoking its foundation after the overthrow of the tyrant suggested that this system would indeed continue. With the addition of libations at sacrifices in all sanctuaries, Harmodios and Aristogeiton were imported into the city’s other religious rites. The repetition of their imagery in different parts of Athens at different times of year united the polis repeatedly in time and space.

• 72  Demosthenes, 20, 69-70; Gauthier (1985), p. 92.

21The Tyrannicides’ cult, consequently, created a particular set of images for the city. Harmodios and Aristogeiton were presented as the slayers of the tyrant and the bringers of isonomia and democracy to Athens. In the aftermath of the Persian Wars, their image changed and they were identified as the men who freed Athens from tyranny. Their role as liberators certainly goes back at least to the 450s when the monuments in the Agora made visible the same dynamics which Herodotos would later use in Miltiades’ speech to Kallimachos. This identity may very well have been invented in the 470s when freedom became a political value. The Tyrannicides also promulgated a very particular image of the good Athenian: he was a male citizen who acted against tyrants and so protected the freedom of the democratic city. After 411, these good citizens were understood as prepared to become tyrannicides themselves. This imagery also set tyranny and democracy in opposition to each other and it stressed that the city was ruled by the demos. After of the overthrow of the Peisistratidai and the subsequent events, the cult served to create unity among the divided Athenians and these dynamics would subsequently be used again in 409 in the aftermath of the oligarchies of 411. In these ways, the polis created a cult which served very well to articulate a particular identity both for herself and her citizens. In turn, the rituals continually reinforced the images of the city and the good Athenian so that the polis was also articulated by them. The influence of the Tyrannicides spread beyond rituals because they also provided the model for honouring the city’s most important benefactors: as Demosthenes noted, Konon was the first man after Harmodios and Aristogeiton to receive a bronze statue, voted because of the victory at Knidos in 394/3.72 Ritual and civic spheres are closely intertwined here and these developments further articulated the identities of the city and the good citizen.

• 73 Herodotos’ and Thucydides’ sources have been extensively discussed in the scholarly literature; par (...)

22In the Tyrannicides’ cult, accordingly, we can see the polis using religion for its own ends and being articulated by it. The institution of these rituals by the city, however, did not prevent the circulation of other versions about how the tyranny of the Peisistratidai was overthrown and how the city came to be ruled by the demos. These alternative traditions are now preserved in Thucydides, Herodotos, Aristophanes, and the orators and these sources indicate that different versions of the events not only circulated, but also flourished in the fifth century.73 In some case, these variants can be traced to specific sub-groups of the polis which seem to have promulgated their accounts in ritual settings. For them, these alternative traditions created a narrative about what it meant to be a member of that particular group. This story then served to create an identity for this community. For us, these variants bring out the multiplicity of traditions in fifth-century Athens about the end of the tyranny and the beginning of the democracy.

• 74  Thucydides, I, 20, 2; VI, 54, 1-2; 55, 1.
• 75  Thucydides, I, 20, 2; VI, 54, 2; 55, 1-3. That he knew the version in which Hipparchos was tyrant (...)
• 76  Thucydides, VI, 54, 1; 59, 1; cf. VI, 57, 3. As Fornara notes, Thucydides is the earliest source t (...)
• 77  Thucydides, VI, 54, 3; 56, 3; Hornblower (2008), p. 443, 449.
• 78  Thucydides, VI, 53, 3; 59, 2, 4.
• 79  Thucydides, VI, 53, 3; 59, 4.

23Thucydides discusses the assassination of Hipparchos at length and he is notoriously hard on the Athenians for their mistaken beliefs.74 In creating his narrative, he is evidently drawing on a number of different versions. As we saw earlier, he knew the tradition in which Harmodios and Aristogeiton freed Athens from tyranny. In this version, Hipparchos, not Hippias, was identified as the tyrant; Thucydides, in contrast, emphasises that Hippias, as the eldest, was the ruler.75 In book VI, he stresses that the assassination was the result of a love affair and, by implication, it was not undertaken for political reasons.76 In his narrative, however, this tradition is in tension with the story promulgated by the cult so that Aristogeiton plots to overthrow the tyranny and the conspirators later hope that the bystanders at the Panathenaia will want to join in liberating themselves.77 In Thucydides’ rendition, Peisistratos and his sons were not harsh rulers until after the death of Hipparchos and Hippias managed to stay in power for another four years.78 About the end of the tyranny, the historian knows a variety of stories: that it was put down by the Spartans, not Harmodios and the Athenians; that it was ended by the Spartans and the Alkmeonidai in exile; and that Harmodios and Aristogeiton were responsible, as articulated by the cult.79 In creating his narrative, Thucydides is reacting against the version promulgated by the Tyrannicides’ rituals, but he is also drawing on several other traditions, the sources of which are not readily identifiable.

• 80  Herodotos, V, 55.
• 81  Herodotos, V, 63, 1-65, 2.
• 82  Herodotos, V, 64, 2. If these Athenians are the Alkmeonidai, they are not so identified.
• 83  Herodotos, V, 66, 1-2.
• 84  Herodotos, V, 70, 1 – 72, 2.
• 85  Herodotos, VI, 123, 2. The reference is back to V, 63, 1.
• 86 Thomas (1989), p. 247-251, 262, 264-265, 272, 280-281.
• 87  Aristophanes, Lysistrata, 1150-1156.
• 88  Aristophanes, Lysistrata, 273-282.

24Herodotos also knows several different stories about Hipparchos’ death and the end of the tyranny. Miltiades’ speech before the Battle of Marathon reflects the version of the rituals, as we saw earlier. Herodotos’ narrative of the events themselves presents a more complicated picture. There, Harmodios and Aristogeiton do, indeed, kill Hipparchos, but the tyranny continues for another four years.80 Its end begins when the Alkmeonidai bribe the Pythia to tell the Spartans to free Athens; they duly invade and, on their second attempt, they force out Hippias.81 The crucial players in so doing are described as Kleomenes ‘with those of the Athenians who wished to be free’.82 The institution of the democracy comes out of the subsequent stasis between Isagoras and Kleisthenes.83 In his attempts to gain the upper hand, Isagoras calls in Kleomenes and the Spartans who end up being besieged on the Akropolis by the Athenians.84 The defeat of the Spartans then makes it possible for democracy to flourish. Later in book VI, in discussing the actions of the Alkmeonidai after the Battle of Marathon, Herodotos states that, in his opinion, the Alkmeonidai did more than the Tyrannicides to free Athens, ‘if, indeed, they really bribed the Pythia to tell the Lakedaimonians to free Athens, as I showed earlier’.85 Like Thucydides, Herodotos was evidently aware of multiple versions about how the tyranny was brought to an end. Various groups of Athenians seem to have assisted or joined the Spartans and the variations probably originate in different versions of the events. The information about the Alkmeonidai in book VI has often been identified as coming from that family, but, as Rosalind Thomas has argued, the emphasis on the bribery of the Pythia is not a credit to the Alkmeonidai and we are more likely seeing here a tradition hostile to that family.86 Herodotos and Thucydides together attest to the existence in the late fifth century of a number of different versions about the end of the tyranny and the liberation of Athens; the story promulgated by the cult of the Tyrannicides was simply one tradition in competition with a series of others. This impression is reinforced by Aristophanes’ Lysistrata: in addition to the liberation by the Tyrannicides, Lysistrata refers to a version in which the Spartans help the Athenians to drive out Hippias.87 The men’s chorus also mention the Athenians’ siege of Kleomenes on the Akropolis, the same story which Herodotos presents, but without connecting it to the institution of the democracy.88 In these three authors, there is a certain amount of overlap between the variants not connected with the cult, but their origins are not identifiable. They must, however, have been circulating fairly widely at this time, hence their use, and so preservation, by these authors.

• 89 Davies (1971), p. 16-19, 376, 379.
• 90  Isokrates, 16, 25.

25In addition to these traditions, there were also stories about the end of tyranny which were promulgated by specific families and celebrated the actions of their ancestors. In the case of the Alkmeonidai, we have not only the hostile version(s) in Herodotos’ narrative, but also the family’s own tradition about its role in the events, as we learn from the younger Alkibiades, who was related to the family through his paternal grandmother, Deinomache, the daughter of Megakles of Alopeke.89 He begins his recitation of their history with Alkmeon, who was the first Athenian to win a chariot race at Olympia, and then refers to the family’s goodwill towards the people which it demonstrated in the time of the tyrants.90 He continues:

• 92  This history, incidentally, demonstrates that the family did not overlook Kleisthenes’ role in the (...)

26As presented here, the family’s history stresses the democratic credentials of its members, even before the democracy had been invented. The forty-year exile and the tyrants’ retaliation against the Alkmeonidai emphasise the family’s hostility to tyranny. We must understand that these feelings finally led to the actions of Kleisthenes and Alkibiades, who was not related to the Alkmeonidai: the overthrow of the tyrants, the return of the demos, and the establishment of democracy.92 In this version, there is no mention of Harmodios, Aristogeiton, and Hipparchos nor of the Spartans’ involvement nor of the four years between the expulsion of Hippias and the institution of the rule of the demos. Instead, the focus is on members of the family and their services to the demos. They are configured as good democrats who fight against tyranny and who make important contributions to the city’s political life. For later generations of the family, this history provides a model to imitate and a standard of behaviour to be reached: it indicates how good Alkmeonidai should act. Repeated in the context of family rituals, this history would reinforce the expected behaviour for a member of this family. For the younger Alkibiades and his father, this history also demonstrates that their democratic credentials are long standing and come from both sides of the family.

• 93  Thucydides, VI, 59, 4.
• 94  Herodotos, I, 64, 3; VI, 123, 1; cf. Thomas (1989), p. 248, 263.
• 95 Thomas (1989), p. 248-249, 250, 265-266, 267, 269.
• 96  Demosthenes, 21, 144. He mentions the family’s exile for championing the demos in stasis, its borr (...)

27Some elements of this history appear elsewhere. Thucydides attributes the overthrow of Hippias to the Spartans and the exiled Alkmeonidai.93 Herodotos reports that the Alkmeonidai were in exile for the whole period after Peisistratos gained the tyranny for the third time in 546 and he also describes the family as ‘tyrant-haters’.94 Much of Herodotos’ narrative, however, does not reflect well on the Alkmeonidai.95 Stressing the bribing of the Pythia and the involvement of the Spartans would easily have thrown mud on the family and undermined their own positive version of the events. If they particularly wanted to emphasise their contributions to Athens, then they could also have stressed that they acted because of their concern for the Athenian demos and not out of personal pique over a love affair gone wrong. One of the influences on Herodotos’ and Thucydides’ narratives, accordingly, will have been the interplay between the Alkmeonidai’s own history and the other versions in circulation. That Demosthenes could summarise the essential aspects of the family’s version in 347/6 even though he was not related to the family indicates just how well known their tradition had become by the middle of the fourth century.96

• 98  This event may be the battle of Pallene in 546, but its identity is disputed; see MacDowell (1962) (...)
• 99  Andokides, 2, 26. The appearance of the exiled demos in this speech shows that this characterisati (...)
• 100  On the problems of this relationship, see MacDowell (1962), p. 1, 206; Davies (1971), p. 27-28; Th (...)

29In this version, the focus is on Andokides’ ancestors Leogoras and Charias who are described as generals and defeat the tyrants at Pallenion.98 This military action results in their return and, by implication, that of the exiled demos. They are, consequently, responsible for bringing democracy to Athens, but, since the demos is presented as having been in exile under the tyrants, we must understand Leogoras’ and Charias’ actions as bringing back the rule of the demos. In this version, there is no mention either of the Spartans’ actions in defeating Hippias or of the establishment of the Kleisthenic democracy. Since that system is assumed to be already in existence, this family is unlikely to have been among Thucydides’ or Herodotos’ sources. In the slightly earlier speech on his return, Andokides again has recourse to the family history. On this occasion, he tells us that his father’s great-grandfather Leogoras engaged in stasis against the tyrants on behalf of the demos; although he could have reconciled with the tyrants, married into their family, and joined them in ruling the city, he chose to go into exile with the demos rather than be a traitor.99 Again, the focus is on Andokides’ ancestor who is depicted as a good democrat who has nothing to do with tyrants and the demos is described as being in exile. These two versions differ primarily on the relationship of Leogoras to Andokides.100 They both focus on the actions of Andokides’ ancestors who are constructed as good democrats who prefer exile to tyranny. The rendition in the speech on the Mysteries stresses the social standing of the family by describing both named men as generals and it implies that they were instrumental in making Athens democratic (again). This family history provides a model for subsequent members and it indicates how they should act.

• 101  Lysias, 26, 22. I see no reason to dismiss this tradition at wishful thinking, as suggested by Tho (...)

30In both of these family traditions, the ancestors are imagined as being actively opposed to the tyrants and in exile during their rule. A similar history must lie behind the words of the speaker of Lysias’ speech against Euandros. In a pre-emptive attempt to fend off charges of hating the demos, the speaker describes his and his family’s relationships to non-democratic regimes. His ancestors, he claims, were not subject to the tyrants because they were in a state of stasis against them the whole time.101 In a fuller version, the demos was probably in exile and the ancestors may have played a significant role in bringing the people back to Athens, as we see in the traditions of the Alkmeonidai and of Andokides’ family. Since the speaker describes his ancestors in the plural (πρόγονοι), the family’s history probably resembled Andokides’ with several relations participating in the stasis against the tyrants. As with our other examples, this tradition will have constructed the family members as democrats and it will have indicated how they should behave.

• 102  Isaios, 5, 46-47.

31In all three of these cases, the families’ versions differ significantly from the story promulgated by the cult of Harmodios and Aristogeiton and the role of the Tyrannicides is ignored. In the family of Harmodios, however, the situation was quite different, as Isaios’ speech on the estate of Dikaiogenes shows. The speaker imagines that his opponent, a descendant of Harmodios, will appeal to family history to bolster his case: that his ancestors slew the tyrant; the speaker also emphasises that the Tyrannicides were honoured because of their bravery rather than their birth.102 In contrast to the other family traditions, this version involves a single tyrant, a detail which suggests that it was probably very similar to the account promulgated by the cult. Harmodios’ family, consequently, had a very distinctive tradition which was quite different from the other family histories and it will have focused on their unique contributions to the city. It will have been reinforced both by the version of the cult and by the Tyrannicides’ benefits. These external factors, however, will also have made it impossible for Harmodios and Aristogeiton to be imported into the traditions of other families because they would have detracted from the actions of those families’ own ancestors.

• 103 Connerton (1989), p. 39-40; Cubitt (2007), p. 166-167, 180-181; Shear (2011), p. 10.
• 104  Above note 3.
• 105  Compare the Labyadai, a gentilicial group at Delphi, whose regulations focus on all of these occas (...)
• 106  On these rituals, see Garland (1985), p. 39-40, 146 and Parker (2005), p. 27-29, both with further (...)
• 107 Parker (2005), p. 27-28 with further references.
• 108 Parker (2005), p. 42.

33These family traditions were put to use in legal speeches where they were certainly helpful for speakers burnishing their credentials in the law courts. Legal cases, however, can not have been the primary setting for telling and disseminating these histories. Handing on such traditions from one generation to the next requires a setting in which family members communicate with each other in an orderly way, that is through ritual, so that they remember their past together and pass it on to future members.103 Telling stories of a shared past in this way articulated the group and gave it identity, as Sourvinou-Inwood has noted.104 The primary contexts in which families gathered to celebrate their continued existence were in religious rituals marking birth, coming of age, marriage, and death.105 Such rites, consequently, are the most likely occasions when these family histories were repeated and transmitted. Although no inscribed list of sacrifices for any Athenian family is known, we can identify a number of occasions when such stories would have been appropriate. The most obvious are the rituals connected with the dead: burial and the subsequent feast, later commemorative ceremonies, and annual rites, all of which offered scope for repeating the family’s history.106 From the late fifth century, the existence of family tomb periboloi containing inscribed stelai which commemorated the family’s dead over a number of generations may have encouraged further repetition of these traditions whenever the graves were visited for post-funeral rituals. So also did the Genesia on 5 Boedromion, when families probably gathered at their graves.107 Since, as Robert Parker notes, the Apatouria ‘was by definition a festival at which kinsmen came together’,108 it provided a further potential occasion for rehearsing these family traditions.

• 109  Herodotos, V, 57, 1; Parker (1996), p. 288-289.
• 110  RO 37, 85-94. The most obvious occasions for rehearsing the history of the Salaminioi are the sacr (...)
• 111  Herodotos, V, 61, 2.
• 112  Salaminioi: RO 37, 88-89.
• 113  Andokides: Davies (1971), p. 27; Alkmeonidai: Parker (1996), p. 318-319.

34Harmodios and his family were also members of the Gephyraoi, who seem to have been a genos.109 The extant list of sacrifices of the Salaminioi indicates how ritually active a genos might be, although this document provides us with no indication of the content of the accompanying rites.110 According to Herodotos, the Gephyraoi had shrines and rituals which were restricted to members of the group;111 they would have provided an ideal setting for rehearsing the group’s history and for stressing that the killing of the tyrant made its members different from other Athenians. Perhaps, like the Salaminioi, the Gephyraoi also made their own additional sacrifice at the Panathenaia;112 certainly such an offering would have been extremely appropriate and it would have brought out their unique history. Andokides and the Alkmeonidai probably also belonged to gene, but we do not know their identities.113 Their rites would have provided additional occasions for telling the two families’ traditions. Such religious rituals at the level of both the family and the genos, consequently, must have provided one of the primary settings for repeating the histories of these groups and (re)creating the members’ identity.

• 114  Demosthenes: above note 96; orators: above note 54. The account in Aristotle, Athenaion Politeia, (...)
• 115  Above note 6.
• 116 Shear (2011), p. 96-106, 109-111, 119-121, 126-127, 139-140, 151, 162-163, 247-250, 259-262, 272, a (...)

36The competing traditions of these sub-groups, however, did not all continue beyond the early fourth century. The Alkmeonid story appears with genealogical errors and simplification in Demosthenes’ speech against Meidias, but otherwise, the fourth-century orators repeat versions reflecting the Tyrannicides’ cult.114 Those rituals annually presented Harmodios and Aristogeiton as the killers of the tyrant and the bringers of democracy to Athens so that they made this moment ever present for the Athenians; when the libations were added to sacrifices in all the sanctuaries, they imported the cult’s focus on democracy and democratic identity to other rituals. Demosthenes does not specify the occasion for this development, but he also does not say that the authorising law was new.115 The first half of the fourth century does not provide obvious circumstances for this legislation. In the late fifth century, however, there seems to have been renewed interest in the Tyrannicides, particularly around 415-410. The opposition between tyranny and democracy would especially fit into the Athenians’ public responses to the events of 411, but not to the Thirty, when democracy was juxtaposed with oligarchy.116 This evidence suggests that the introduction of the libations to all the sacrifices probably took place in the years immediately after 411 when the new rituals will have constituted part of the Athenians’ reactions to oligarchy.

• 117  Pausanias, I, 29, 15.
• 118  Andokides 1 dates to 400, Isokrates 16 to soon after 397, Isaios 5 to ca. 389, and Lysias 26 to 38 (...)

37At this time, the version promulgated by the rituals for the Tyrannicides became all pervasive and inescapable. It would also have been reinforced by public monuments, particularly the statues in the Agora, but also the men’s grave in the Demosion Sema.117 This repetition together with the authorisation of the polis made the cult’s version far more powerful than the stories of any individual family. These alternative versions would have been told in various settings, including family rituals, but these occasions in the fourth century would not have been as numerous as the sacrifices in all the sanctuaries and these histories may well not have been reinforced by physical monuments. Faced with such stiff competition from the polis in the years after 411, these alternative versions could not compete indefinitely and they quickly died out: only the Alkmeonid tradition is attested after 380 and only then in a speech given by Demosthenes, a man who was not a member of the family, in 347/6.118 The cult, in contrast, flourished and it continued to articulate what it meant to be a citizen in democratic Athens and how that citizen should act. The end of tyranny and the institution of democracy was now the story of the whole city and not the possession of individual families. After the early fourth century, these families would have to (re)create their histories and identities in other ways so that they did not compete with the city and its religious rituals.