The Kinship of Perikles and Alkibiades Thompson, Wesley E Greek, Roman and Byzantine Studies; Spring 1970; 11, 1; ProQuest pg. 27 The Kinship of Perikles and Alkibiades Wesley E. Thompson N EPOS SAYS that Alkibiades was Perikles' stepson (privignus) , but according to Diodoros they were uncle (()eioc) and nephew (&8eArpt8ovc).1 Since our other evidence seems to indicate that neither statement is correct, these testimonia are generally simply ignored. The purpose of this paper is to establish that Diodoros and Nepos are actually wrong and to seek the origin of their errors in a misunderstanding of an uncommon Greek kinship term. To refute Diodoros it will be convenient to set forth the family tree of Alkibiades, from which it will be seen that Perikles could not be the brother of his father, Kleinias, or of his mother, Deinomache. MEGAKLES I I I I AruPHRON HIPPOKRATES KLEISTHENES KLEINIAS ALKIBIADES I I XANTIDPPOS = AGARISTE MEGAKLES = daughter ALKIBIADES = daughter I I I PERIKLES DEINOMACHE ====== KLEINIAS AXIOCHOS ALKIBIADES This stemma depends largely upon three basic testimonia. Herod­ otus (6.131) mentions some of the descendants of Megakles and 1 Nep. Ale. 2: educatus est in domo Pericli (privignus enim eius fuisse dicitur); Diod. 12.38.3. 27 28 THE KINSHIP OF PERIKLES AND ALKIBIADES Agariste, the daughter of Kleisthenes of Sikyon: OVTOC (sc. Kleisthenes f A h ) ~ \' lI6 \ I.. "1 ' ,~\ '1 1 o tens TE 07j YWETaL lYleyaKI\€£ Kat 7T7TOKpaT7jC, EK DE 7T7TOKpaTEOC lI6 \ 1 "'\ \ , 'A 1 "\ \ '\ ~ 17\ a 1 'A 1 lYlEyaKI\E'Y]C T€ al\l\oc Kat yaptCT7j al\l\7j, a7TO T7jC fil\€£CUEV€OC yapLCT'YJC " '" c\ ",...., a' - 'A '..J.. ' " EXovca TO ovvoJLa, 'Y] CVVOLK'Y]caca T€ ,!;;!,aVvt7T7TlfJ TlfJ pt'f'pOVOC Kat EyKVOC ,~ ,.~ ".1. ' ~" ,~ , ~ \ \ 1 ~" '\ 1 €ovca €tUE O'f'tV €V TlfJ V7TVlfJ, EOOK€€ DE I\Eovra TEK€tV. Kat JLET ol\Lyac ~JL'pac 'rLKT€L IIEpLKMa Eav8L7T7TlfJ. This gives us Perikles' ancestry. To reconstruct Alkibiades' lineage we rely on speeches from cases concerning his son. In a speech prepared by Isokrates (16.26-7) the son 'A\ f3 ,~ '17\ a' • \ , ,. ~" , says, I\Kt tCX07JC Kat LVl€tCVEV7]C, 0 JLEV 7TpOC 7TaTpoc, 0 DE 7TpOC JL'Y]TpOC WV 7TP07TCX7T7TOC TOV 7TCXTPOC TOVJLOV, CTpaT'YJy~cavTEC TijC cpvyijc KaT~yayov , ~ ~ " , 'l; 'Q \ ' 1 " , TOV O'Y]JLOV Kat, TOVC Tvpawovc €':,E/"aI\OV, Kat KCXT€CT7jcav €KELV7]V T'YJV 87jJLoKpaTLav. This gives us the names of two of Alkibiades' great­ grandfathers; we get evidence for the names of the other two from a speech written by Lysias (14.39) against his son: 'AAKLf3L<~8'Y]v JLEV \ I , ""' \ \ , 'i'llA' \ , 't J , TOV 7Tp07Ta7T7TOV aVTOV Kat TOV 7TpOC JL7jTpOC lYlEYCXKI\Ea OL VJLETEpOL 7TPOYOVOL ~, ',J.. 1 'l; 1 ~ ~, " ~. f3 1 otC aJL'f'OTEpOVC €sWCTpaKtCaV, TOV DE 7TaTpoc aVTOV Ot 7TP€C VT€PO£ VJLWV 8avaTOV KaTEyvwcav. Although this passage is usually considered corrupt, there is nothing wrong with it if aVTOV refers to the same per­ son, the son of Alkibiades, in both instances. When we compare it with what the writer of [Andokides] 4.34 says about Alkibiades' ancestors, J TijC JL7jTPOC 7TaT~p MEyaKAijc Ka~ J 7Ta7T7Toc 'AAKLf3La87jc lswcTpaKLc8'Y]cav aJLcpoTEpOL, we see that Megakles and Alkibiades are the grandfathers of the famous Alkibiades and great-grandfathers of his son.2 Of course, if 7TPOC JL7jTPOC in Lysias refers to the son's mother, something is amiss. Thus I suggest that here the phrase is simply equivalent to 7TPOC yvvaLKwv. In Athenian law Alkibiades' son was re­ lated to Megakles 7TPOC yvvaLKwv, not 7TPOC av8pwv.3 The evidence of Lysias and [AndokidesJ, taken together, shows that Alkibiades' grandfathers were named Megakles and Alkibiades. Plutarch provides confirmation of this conclusion, for he names Megakles as the father of the statesman's mother, Deinomache, and says that Alkibiades' father, Kleinias, excelled at the battle of Arte­ mision (Alk. 1). According to Herodotus (8.17) the hero of Artemision 2 Up to this point I have followed the interpretation of A. E. Raubitschek, TAPA 79 (1948) 203-04. 3 Under the inheritance law the descendants of any sons of Megakles would take pref­ erence, as l~ appl.vwv, over the descendants of any daughters, such as Alkibiades and his son. For an analogous situation cf Oem. 44.13 and 19, where the contrast is between the descendants of the deceased's brother (1TpOC avSpWv) and those of the deceased's sister (1TpOC yvvruKWV). WESLEY E. THOMPSON 29 was Kleinias, the son of Alkibiades. Even if one rejects Plutarch's identification of the hero of the battle with the father of Alkibiades, we have yet another way of establishing the name of his second grand­ father: Alkibiades' uncle, according to plato (Euthyd. 275A) was Axiochos, the son of Alkibiades.4 Now that we have established the names of Alkibiades' grand­ fathers, who were ostracized, we can determine those of his other great-grandfathers from ostraca which name Megakles, the son of Hippokrates, and Alkibiades, the son of Kleinias.5 Aristotle (Ath.Pol. 22.5) confirms the fact that Megakles, the son of Hippokrates, actually was ostracized. Perikles' parents, then, were Xanthippos and Agariste, Deino­ mache's were Megakles and a daughter of Kleisthenes,6 and those of Kleinias were the Alkibiades who was ostracized and a daughter of the Alkibiades who helped to overthrow the tyrants and establish the democracy. This, at any rate, is what the evidence from the fifth and fourth centuries implies. To this reconstruction, which is essenti­ ally the one proposed by Raubitschek,7 Vanderpool raises the objec­ tion that Hit makes the rather rare name Alcibiades occur not only in the family of the elder Alcibiades but also in the family of the woman he married, a strange coincidence unless perhaps the woman he married was a cousin."8 Since, however, the marriage of close rela­ tives was a common occurrence in the Athenian aristocracy,9 we should not hesitate to follow the evidence of the orators. As far as consanguineal relationship is concerned, therefore, our stemma shows that Perikles and Deinomache were first cousins. In Attic kinship terminology Perikles was Alkibiades' aVEtfU)C, while 4 An Axiochos, son of Alkibiades, appears among those condemned in connection with the sacrileges of 415; cf IG 12 328. I> For the latest list of known ostraca cf Russell Meiggs and David Lewis, A Selection of Greek Historical Inscriptions (Oxford 1969) 45-47. 6 For the daughters of Kleisthenes cf Cic. Leg. 2.41: et Atheniensis Clisthenes Iunoni Samiae civis egregius, quom rebus timeret suis,jiliarum dotis credidit. 7 TAPA 79 (1948) 203--04; RhM 98 (1955) 260 n.4; cf also T. Leslie Shear Jr, Phoenix 17 (1963) 99-112. The usual reconstruction makes the Megakles who was ostracized the son of Kleisthenes instead of the son of Hippokrates; cf Wilhelm Petersen, Quaestiones de historia gentium Atticarum (Kiel 1880) 90--91; Johannes Toepffer, RE 1.2 (1894) 1557-58; Johannes Kirchner, Prosopographia Attica II (Berlin 1903) no. 9688. 8 Hesperia 37 (1968) 398. 9 Phoenix 21 (1967) 273-82. To the instances of marriage between first cousins collected there, add the marriage of Themistokles' daughter to her first cousin (Plut. Them. 32) and that of the Platonic Phaidros to his first cousin (Lys. 19.15). 30 THE KINSHIP OF PERIKLES AND ALKIBIADES Alkibiades was an aV€!fuxOOVC to Perikles. To call them O€'ioc and aO€AcptOoVc is incorrect.1o If Perikles was not Alkibiades' uncle by blood, perhaps Diodoros meant to indicate that he married an aunt of Alkibiades. We know from Plutarch (Per. 24) that Perikles married a kinswoman. Some German scholars in the nineteenth century thought that she might be Deinomache's sister.ll Whatever the truth of that suggestion, Dio­ doros cannot be used as evidence for such a relationship. Whenever the context makes the relationship clear, O€'ioc in Attic prose always indicates an uncle by blood, not by marriage, and likewise aOEAcptOOVC always indicates a blood relationship. This is, of course, not proof that the words cannot indicate a marriage tie, for the bulk of the instances come from oratory and in particular from inheritance cases, in which affinal relatives play little or no part. Much more significant is the usage of Polybius and Plutarch. In about twenty instances where the context or historical situation determines the meaning of O€'ioc and &8€Acpt8ove these words always indicate a blood relationship.12 In addition to this negative evidence we have three instances which indi­ cate quite clearly that neither author would use aO€AcptOOVc to mean nephew by marriage. In Plutarch (Mar. 6) Marius marries Julia, ~e -ryv a8€AcptOove Kaieap.13 He also calls Apollokrates the nephew of Dion's wife, not of Dion himself: a8€Acpt8ovv fJ-EV oJI'Ta -rfje Eav-rov yvvatK(k (Dion 56), and Polybius describes the relationship of Flamininus to a certain Quintus Fabius, DC -ryv afJ-r0 -rfje yvVatK6e a8€Acpt8ovc (18.10.8). Likewise, if Perikles married Deinomache's sister, Diodoros should call Alkibiades the nephew of Perikles' wife. As for Nepos, we can rule out the possibility that Perikles actually married Alkibiades' mother. According to Plutarch, Perikles' "wife was near of kin to him and had been wedded first to Hipponicus, to whom she bore Callias, surnamed the Rich; she bore also, as the wife of Pericles, Xanthippus and Paralus. Afterwards, since their married 10 It is true that we have only one reference to a parent's first cousin ([Dem.] 43.41 and 49, where &v€l/J,6c is used), but &v€l/J'«oovc occurs frequently enough to show that it is the proper term for the son of one's first cousin; cf Isaios 9.2; Dem. 44.26, 57.67 and 57.68 (a sure emendation based on 57.38). 11 Cf Wilhelm Vischer, Kleine Schriften I (Leipzig 1877) 98 n.l. 12 Polyb. 10.7.7 and 12.13.4; Plut. Ant. 19, Brut. 2 and 22, Caes. 62, Cat.Min. 1,2,3, and 36, Cic. 46, Demetr. and Ant. 5, Dion 6, 7, and 14, Lue. 1, Marc. 30, Otho 16, Pel. 29, Popl. 3 and 4, Mor.492c. 13 Cf also Pluto eaes. 5. WESLEY E. THOMPSON 31 life was not agreeable, he legally bestowed her upon another man, with her own consent, and himself took Aspasia, and loved her ex­ ceedingly."14 Had this woman also been the mother of Alkibiades, Plutarch must have known and must have said SO.15 Given, then, that both Diodoros and Nepos are wrong in their state­ ments about the kinship of Perik1es and A1kibiades, how can we account for their errors? It is clear that in one of his sources Diodoros found a reference to the relationship of the two men. The question is, whether he misinterpreted what he found or whether his source was inaccurate in the first place. Although we cannot be certain, I would discount the second possibility in view of the fact that as late as Plutarch's time there was abundant material on the family of Perikles available to an historian engaged in independent research. On the other hand, a compiler like Diodoros, finding an ambiguous kinship term in his source, might easily misunderstand it. For in­ stance, the word K7JOECT~C, which can mean father-in-law, son-in-law, or brother-in-law, is frequently mistranslated in various volumes of the Loeb Library, when the translator is lax in investigating the historical context of the passage.16 In modern Greek 8ELOC can mean not only uncle but also one's parent's first cousin17 and thus can describe the relationship of Perikles to Alkibiades. If Diodoros' source used 8ELOC in this sense, it would be an easy error for him to interpret the word as uncle and assume that Alkibiades was Perikles' nephew. Unfortunately, I have been unable to find any instance in classical Greek where 8ELOC does have this meaning. In the only place in Attic literature where a parent's first cousin is mentioned ([Demos­ thenes] 43.41 and 49) the term aV€VJLoc is used.18 14 Per. 24, translated by Perrin (LeL). 10 Isok. (16.28) and Plato (Alk. 104B; Prot. 320A) certainly do not suggest that Perikles married Alkibiades' mother. 16 Some examples are Aischin. 1.115, Oem. 45.8, Isai. 2.29, Isok. 18.52, and Lys. 13.1. In each instance the translator renders KT)S€C-n]C as brother-in-law, but there is no evidence to show what the actual relationship is. To take an example from an ancient historian, Livy (37.45.5 and 37.55.3) expands Polybius' term &S£Acp'SOVC (21.16.4) to frarris filius. 17 John Andromedas, American Anthropologist 59 (1957) 1087; J. K. Campbell, Honour, Family, and Patronage (Oxford 1964) 42 n.6. 18 Aristophanes of Byzantium (fr.9 Nauck) and Pollux (3.22) give only 'uncle' as the meaning of Oli.oc. Kritias was a first cousin of Plato's mother (cf Hans Raeder, Hermes 72 [1937] 404-10), and Aristotle was a first cousin of Kallisthenes' mother (Plut. Alex. 55; Suidas, s.v. KaX>"cO'VYJc), but no ancient source applies a kinship term to Kritias or Aristotle in this connection. 32 THE KINSHIP OF PERIKLES AND ALKIBIADES The very fact that such a remote relationship is so seldom men­ tioned suggests that the sources simply described Perikles as a relative of Alkibiades and did not specify the degree, just as Plutarch (Alk. 1) calls Perikles and his brother, Ariphron, 7TpOC~KOVT€C KaTa '}'EVOC of Alkibiades. It would be especially appropriate in this case to add the fact that Perikles was a maternal relative, since the connection was through the family of the Alkmeonidai. Of course, no one could reasonably extract the meaning 'uncle' from such a phrase as 7TpOC~KWV 7TPOC p-7JTp6c, but there is a term outside the Attic dialect which could easily lead to such a mistake. According to the scholia Pindar uses the word p,eXTPWC to mean ~ J TijC P-7JTpoc &8€)..cP0c ~ J KaTa P-7JTEpa cvyy€~c,19 and the Gortynian Code uses it in this wider sense.20 The range of meaning of the Ionic p-~TPWC cannot be deter­ mined for lack of evidence,21 but it was probably the same. Conse­ quently, if Diodoros found Perikles described as a maternal relative (p,~TpWC), he would naturally assume that the word meant 'uncle' and that Alkibiades was his nephew. Of course, proof is out of the question, but this does seem to me the most likely way to explain Diodoros' error.22 With Nepos the problem is trying to reconstruct the Greek which he is translating or paraphrasing (at Ale. 2) with the words educatus est in domo Pericli (privignus enim eius fuisse dicitur). Isokrates, for instance, says (16.28) that Alkibiades E7T€TP07T€t)(}7J 8' V7T() II€p£KA€ovc. If Nepos found a similar statement in his source, he might simply have con­ fused the guardian/ward relationship with that of a stepfather and his stepson. On the other hand, his own method of expression is more in the manner of Diodoros (12.38.3), 'A)..K£f3£c~87Jc J &8€)..cP£So';;c. dpcPavoc wv. TP€cP6p-€VOC 7Tap' aVTctJ (sc. ll€P£K)..€i) and Plutarch (Alk. 1), TO';; 8' 19 Scholium to Nem. 5.43 (78a Drachmann). The actual usage is maternal uncle at Nem. 4.80 and 5.43 and Isthm. 6.62, maternal grandfather at 01. 9.63, and maternal ancestor or relative in general at 01. 6.77 and Nem. 10.37 and 11.37. At Isthm. 7.24 the meaning is un­ clear. 20 Col. 9, line 4 and col. 12, lines 13-14 Willetts. u Herodotus (4.80.4) uses it to mean maternal uncle and 1TtlTPWC (2.133.2, 4.76.6,6.103.4, 7.10.«.1,7.46.1, and 9.78.3) to mean paternal uncle. U For a similar mistake cf. HSA Alex.Sev. 49.5. Although Alexander and Elagabalus were cousins (Dio 78.30), the Greek historian Dexippos dicit patruum fuisse Antoninum Helio­ gabalum Alexandri. The apparent explanation here is that Dexippos correctly described Alexander as an a.v~'oc ('cousin') of Elagabalus, but the Latin writer took the word in its later sense of 'nephew' and concluded that Dexippos considered Elagabalus the uncle of Alexander; cf. A. Jarde, Etudes critiques sur la vie et Ie regne de Severe Alexandre (Paris 1925) 5. WESLEY E. THOMPSON 33 'A\ f3'~ n \ ~ "A 'J. • ,.... ()' , \ I\KL Laoov €PLKI\T]C KaL pt'f'pWV Ot .!!tav L7T7TOV. 7TpOCT]KOVT€C KaTa y~voc. ~7Tf:Tp(nr€Vov. This indicates to me that Nepos' Greek original employed some form of the verb ~7TLTp07T€VEW plus a separate kinship term, such as 'cousin', 'nephew', or simply 'relative'. Nepos uses educatus est to translate the verb, but what word underlies privignus? Its Greek equivalent is 7TPOYovoc, but it is hard to see how Alkibiades could be called the 7TPOYovoc of Perikles or anything similar. On the other hand, the reciprocal of this term is P:f]TPVLOC, 'stepfather'. As we have argued, since Perikles was in fact a maternal relative of Alkibiades, it would be natural to describe him as a f.L~TpWC. I suggest, therefore, that the errors of Nepos and Diodoros may have a common source, a state­ ment that Perikles, a maternal relative (p,~TpWC), acted as guardian to Alkibiades. Diodoros took the word to mean 'uncle', Nepos confused it with f.L7]TPVLOC, 'stepfather'. To summarize, Perikles was by blood a first cousin of Alkibiades' mother, not his uncle, as Diodoros says. Perikles married a woman closely related to him; although we do not know her identity, she was not Alkibiades' mother, as Nepos implies. The origin of these errors by the universal historian and the biographer cannot be deter­ mined with certainty, but my suggestion is that both authors were misled by finding Perikles described as a P.~TPWC (here = 'maternal relative') of Alkibiades. UNIVERSITY OF CALIFORNIA, DAVIS January, 1970 3-G·R.B.S.