Cleisthenes and Attica

Author(s): D. M. Lewis
Reviewed work(s):
Source: Historia: Zeitschrift fr Alte Geschichte, Bd. 12, H. 1 (Jan., 1963), pp. 22-40
Published by: Franz Steiner Verlag
Stable URL: http://www.jstor.org/stable/4434773 .
Accessed: 19/02/2012 04:04
Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at .
http://www.jstor.org/page/info/about/policies/terms.jsp
JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of
content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms
of scholarship. For more information about JSTOR, please contact support@jstor.org.
Franz Steiner Verlag is collaborating with JSTOR to digitize, preserve and extend access to Historia:
Zeitschrift fr Alte Geschichte.
http://www.jstor.org
CLEISTHENES AND ATTICA.
There has been much recent work on Cleisthenes.' The justification for yet
another article lies, I hope, in its different approach, from the land of Attica
itself and the framework which Cleisthenes gave it. I hope also that this
difference of approach will excuse what may seem a somewhat cavalier attitude
to my immediate predecessors, whose arguments will seldom appear, although
I have read them with attention and profit. The argument of this article
proceeds largely from survivals, and leads to the paradox that we can under-
stand Cleisthenes' work best in the places where he failed. These failures,
however, give us clearer light not only on his aims, but on the woven texture
of tradition and innovation which is formed by the life of classical Athens.
I. Regional Parties in Sixth-Century Athens.
Let us begin by sketching our evidence for regional divisions in Athenian
politics before Cleisthenes' reforms. The parties of the 560's have regional
names, acquired, we are told,2 from the places where they farmed. We should
emphasise that parties may include members from outside the original area of
the nucleus. When Peisistratus waited at Marathon in 546, supporters came to
him from Athens itself as well as from the country-villages.3 This should also
act as a warning against any easy assignment of single economic or even con-
stitutional motives to a party. The party may be a complex of driving-forces
held together by its leader; such a complex is in fact sketched for Peisistratus'
party by Aristotle.4 However, even with these qualifications in mind, it is
legitimate to look for the nuclei.
The pedieis are relatively straight forward. None will doubt that their
nucleus is in the plain of Athens itself, not indeed the whole of it, for we shall
1
The two important items of the last fifteen years which do not appear in my notes are
Larsen, Representative Government ch. I, and Bradeen, T.A.P.A. lxxxvi, I955, 22-30. The
second starts from presuppositions which are here rejected, the first I cannot accept,
largely because I believe that the bouleutic oath is more likely to have formalised than
to have altered the situation which existed before it. This paper has been read by
A. Andrewes, W. G. Forrest and A. E. Raubitschek, and much improved by them. They
are not responsible for the blemishes which remain, and would, I think, differ from me on
several points. (See also Hopper, BSA LVI, I89-219.) 2 A th. Pol. 13. 5. 3 Hdt. 1 62. X
? Ath. Pol. 13. 4-5. My idea of the 'Parties' does not differ substantially from that
formulated by Sealey, Historia IX, I63-5.
Cleisthenes and Attica 23
see reason to exclude the area south and south-east of the city, but at least the
more substantial portion to the north and west. Support for this comes from
the name of their leader, Lycurgus.5 The chances are high that he, like the
later prominent men of the name, came from Boutadai, a little to the west of
the city.6 No other family can be confidently associated with this party, but
one may be tempted to go back to Myron of Phlya, prosecutor of the Alcmeonids
for their action against the Cylonian conspiracy.7 He will have come from the
north-east of the pedion.
The paralioi may best be located by looking for Alcmeonid land. Alcmeonids
are found in three demes, Alopeke,8 Agryle,9 and Xypete,'0 all to the south
and south-east of the city.I' They farm the land which goes down towards the
coast, and will thus have got their name. Two other families are to be associ-
ated. The principal family of the Kerykes, headed in this period by Kallias
son of Phainippos, is attested as notably anti-tyrant by Herodotus.'2 After
many wanderings in modern scholarship, it is now firmly settled in Alopeke.13
Another Alopeke family, later allied with this one by marriage,14 is the house
which produced Aristides,'5 who, we are told,'6 was in early life a hetairos of
Cleisthenes. Alopeke is reasonably to be identified as the headquarters of the
party, and this finds an echo in a partisan drinking-song which will go back to
this period'7 oIj )atcv &7s X CLV, oV8' &p.ro'rpoLa yLyveat L pLov, where
a'&xrCX v refers as much to the political shifts of the men of Alopeke and
their leader Megakles as to the cunning of the fox.
Both of these parties will have had their supporters in the city. Alcibiades
of Skambonidai and Leogoras of Kydathenaion, head of the other main branch
of the Kerykes, could claim opposition to Peisistratus later.'18
The Hyperakrioi are, on the face of it, the men beyond the hills, outside the
Parnes-Pentelikon-Hymettos ring.'9 Diakrioi, the main alternative form of
their name, will come from the Diakria, which is said, on the most plausible
5
Hdt.
I
59. 3.
6
Prosopographia Attica
924"9251.
7 Plut. Sol. 12. 4.
8
M?yox?,q 'Itr7roxp&-Touq (Ath.
Pot. 22.
5
and numerous
ostraka)
and others.
9
Acoprq
'A?x,tkcavoq
(Plut. Them. 23. i); cf. also 'A)
X,eo)v(8&n 'Epex'&-r(oq
(Tod,
GHI 26. 135). 10
KocXM?evoq
'ApaTco.v4u (Hesperia XIX 376ff.).
11
Sealey, op. cit. I63 n. 4r, adds Leukonoe from IG I2 368. 3-4. This must remain
uncertain, since no certain Alcmeonid is called Megakleides. If he were accepted, we would
have Alcmeonid land north of the city.
12
VI 12I.
18 Hesperia V
410,
BSA L
13-14,
JHS LXXXI
120.
14
Plut. Arist. 25. 4. If Aristides had been himself a Keryx, we would have been told.
15 I bid . I . I.
16
Ibid. 2. i. His father Lysimachos may be the tamias of IG I2 393, which presumably
dates from a period of Peisistratid exile, since it has an Andocides (see n. i8). But J. K.
Davies points out to me that this Lysimachos could also be a Boutad, ancestor of the
family analysed in BSA L 7, which would make as good political sense.
17
Ar. Wasps 1240-I241. 1' Isocr. XVI 26; And. I io6, II 26.
19
So Wade-Gery, Essays in Greek History, I67 n. 2.
24 D. M. LEWIs
emendation of Hesychius,20 to run from Parnes to Brauron. It is at Brauron
itself that we can safely place their leader Peisistratus.20a Marathon, further
north, is well within their area, and it is tempting to think that the tyrant's
party landed there in 546 and 490 because they could be sure of local support.
The direct arguments for Peisistratid supporters in this area are not however
very strong. Wilamowitz21 thought that the Aristion who moved the decree
for Peisistratus' bodyguard22 was the Aristion commemorated in the well-
known relief from Velidaneza, who would thus have come from half-way
between Brauron and Marathon, but the date of the relief is too late23 and the
name not uncommon. More hopeful supporters might be found in the two
Marathonians found on ostraka of the 480's. Of these, Boutalion is a serious
possibility for the missing name of the pro-tyrant ostracised in 485.24 The
other, Habron son of Patrokles,25 I was once inclined to think identical with a
Habron, archon under Hippias in
5I8-7,26
possibly son of the Patrokles who
dedicated the mid-sixth-century altar of Athena Nike,27 but Lobel28 has now
made it highly unlikely that there was such an archon. The case for Peisistratid
supporters in Marathon must remain largely one of probability.
Hyperakrioi is, I take it, a wider term than Diakrioi, and will accommodate,
besides these north-eastern hills, the East Attic plain which we now call the
Mesogeia. The population of this plain must already have been substantial in
the time of Cleisthenes, for the descendants of the demesmen of 507 provided
over an eighth of the fourth-century boule. It is, however, extremely short of
early families of importance. The one sixth-century figure is Socrates of
Paiania, father of the Phye who played a part in Peisistratus' first return and
married his son Hipparchus.29 Paiania lies directly on the only route from
Brauron to Athens. The Mesogeia has been described as "the plain dominated
by Brauron and Marathon."30 This is slightly misleading. Marathon is very
much tucked away behind Pentelikon, and a sharp uphill walk is needed to
take one out of the marshy valley of Brauron itself to the 120-metre level of
the main Mesogeia. Nevertheless it is true that, in the absence of powerful
20
S.V.
A6oxpeL5
.... xoct
'
X'pot
ALaxptoc, i &br6 llpvroo eLS
)oXuSX6voq.
Ure, Origin
of Tyranny, 312, suggested gc,g
AiSXcovoq
which has reached Latte's apparatus, but Aulon
is of no importance and forms no kind of landmark to define a boundary. Clear evidence
for Brauron in this context comes from Bekker, Anec. 242 ALdXpLMC t60roq 'AvrtLxi
(ukr6
Bpxupuvo 'EXeua(vLov A' actpo x
ep
sepM'T Lep6v. I would put a full stop after
Bpxuprovm, and regard the last five words as a totally different gloss, wrongly incorporated.
20a
Plato Hipparchus 228b, Plut. Sol. 10. 3.
21
Aristoteles und Athen I 26I; cf. Burn, Lyric Age of Greece, 305. 22Ath. Pot. I4.
I.
23
See most recently Richter, The Archaic Gravestones of Attica, 47. She puts it c. 520.
24
Ath. Pol. 22. 6. Hesperia XXI 8 with n. I5.
25
Hesperia VI
I55-6,
Suppl. VIII 409.
26
Cadoux, JHS LXVIII I12.
27
Raubitschek, Dedications from the Athenian Akropolis no. 329.
28 On P. Oxy. XXVI
2438.
29 Kleidemos 323 F
I5.
30 Sealey, op. cit. i65.
Cleisthenes and Attica 25
families in the plain itself, Marathon and Brauron, both with important cults
to serve as a focus, could exercise influence on the plain, which would tend to
look east to them before it looked west to Athens.
Two more families of importance remain for consideration. Since Wilamo-
witz31 it has been normal to associate the Philaids with Brauron and the
Hyperakrioi.32 The case for this has been substantially political, and the only
topographical argument is that Philaios is said to have settled at Brauron,
which was called Philaidai in the Cleisthenic deme-system.33 However, it can
hardly be true that Philaios bound his descendants to the site forever. On this
argument, it would be surprising that Themistocles came from Phrearrhioi
and not from the Lykomid centre at Phlya, and in fact the ancestors of
Epicurus, the only man described as a Philaid by an ancient source, lived at
Gargettos, well away from Brauron, in 507.34 There has never been any doubt
that the Kimonids, who succeeded to the position of Miltiades son of Kypselos,
lived elsewhere. Their family estates were at Lakiadai,35 west of Athens on
the Sacred Way, and Kimon and doubtless his father bore that demotic.36
Their family tombs were at Koile, just outside the city to the south-west.37
They may indeed have been a different family, although it has recently been
observed38 that it would make better sense of Miltiades' adoption of Stesagoras
II if Stesagoras I had also been a Philaid. A neglected Pindaric scholion even
positively asserts the descent of Kimon from Aias.39 A stronger positive reason
for detaching the Philaids from Brauron may be found in the story of Miltiades'
meeting with the Dolonkoi.40 The Dolonkoi leave Delphi, go through Phokis
and Boeotia, receiving no invitations, xo' SxrpS7:va'r0c VIC' 'Ahv6&v, that is,
they turn along the Sacred Way through Eleutherai and Eleusis. Miltiades,
sitting by his front door, sees them passing and shouts his invitation, certainly
not from Brauron, but rather, we may confidently assert, from Lakiadai.
Territorially, at any rate, the case for supposing the Philaids pedieis seems
strongest.
Cleisthenes' opponent, Isagoras, presents a more difficult problem. Two
recent scholars41 have made him a Philaid, largely on the strength of his
father's name, Teisandros. The name appears in other families too, and the
3'
Op.
cit. II 73-74.
32
Wade-Gery, Hignett, Hammond, Sealey, Burn, to name only a few.
33 Plut. Sol. 10. 3. 34 Diogenes Laertius X i. as Ath. Pol. 27. 3.
36
Plut. Cim. 20. 2; cf. IG 12 295. 8.
37
Hdt. VI I03. 3, Marcellinus Vit. Thuc. 55. 38 Burn, op. cit., 3TI.
39
On Nemean II I9, from
Didymus, but I have an
uneasy feeling
that
Didymus may
have got muddled and thought that the Pherecydes genealogy quoted from him
by
Marcellinus, Vit. Thuc. 3 proved the Aiantic descent of the
younger branch,
which it
does not. Doxopater ad Aphthon. p. 439, 3 Walz also goes back to this (cf. Schmidt,
Didymi Fragmenta, 324 ff., who misses the point about the relationship of the two
branches.)
40
Hdt. VI 34-35 41 Sealey, op. cit. 172; Hammond, C.Q. n.s. VI
127-8.
26 D. M. LEWIS
case is hardly strengthened by a mysterious brother of Miltiades rr, with a
name ending in -agoras, who appears in 489.42 Herodotus43 says firmly that he
does not know Isagoras' family, but that his auyyevese sacrifice to
Zeuq
K&pLo~.
It is hard to believe that he discovered the family cults and not the Philaid
blood, and even harder to believe him a deliberate liar. The best approach is
probably through the family cult. Zu6~ KocpLo4 has been thought to be a sneer
at oriental origins by many from Plutarch" onwards. But he seems to be at
home in Ikaria in the Pentelikon area, where as simple Karios he seems to be
the basic deity, beside whom Dionysos is an importation.4 Here too we may
place the headquarters of Isagoras' genos, though he himself may have moved
elsewhere.
II. Cleisthenes' Reforms.
A. The Deme-Names.
Obviously, there were centres of population before Cleisthenes. Cleisthenes
gave them corporate existence by making demotai and demarchoi.46 Aristotle47
notes that not all places had names and Cleisthenes gave them names. The
interesting names are those Ocro T&v wttmovTcov, the patronymic deme-names.
Some of these, for example, Boutadai and Paionidai, are certainly also the
names of gene. It is not stated and there is no probability that membership
of such demes was confined to members of these gene. It is obvious that it
would be distinctly weakening to a kinship-organisation if the same name
were given to a state unit of more open composition. One clear instance shows
that this blow was felt. By the fourth century48 the genos of the Boutadai,
which held two of the most important Athenian priesthoods, has renamed
itself the Eteoboutadai, the real descendants of Boutes.4" If the blow was felt,
it may well have been intended; the assumption that it was is
strengthened,
when we recall the case for supposing that the Boutadai had headed the
Pedieis. This may not be the only case, and observation seems to show that
there is a concentration of patronymic names in the likely
area of the Pedieis.
The game could perhaps be played both ways. That Brauron was a place
of importance is clear, but its name was a word of power, the epithet of Artemis
42
Nepos Milt. 7. 5. 43 V 66. x. U Moralia 86oE.
45
IG 12 i86. This is at least the present state of our evidence, but Forrest rightly warns
me that it is only in 1. 6 that the deity is unequivocally
K&pLoq
and not
'Ix&pLoq,
and that
there may be an error in the reading here. If there is such an error, all reason to connect
Isagoras with Pentelikon of course vanishes, and I would withdraw what is said below
about the political motive for Cleisthenic reorganisation on Pentelikon, but not the
reorganisation itself. 46 Ath. PoI. 21. 4-5. 47 Ibid.
48
Dem. XXI I82 is perhaps the earliest reference.
49
Cf. IG II2 3474. 3 Bou'rr.&v k 06cv t
g[atLro;]
and Toepffer, Attische Genealogie,
I17. There is an interesting parallel in the genos of the Salaminioi, which renamed itself
Archaiosalaminioi ca. 300 B.C. (Daux, R.E.G. LIV 220-222).
Cleisthenes and Attica 27
Brauronia, and power to the wrong party. Brauron is not the name of a Cleis-
thenic deme. Instead the deme is called Philaidai, an annoyance to the Philaids
in itself and perhaps an attempt to foster such cult of Philaios as there was
away from the Philaid centre at Lakiadai.
Territorially the country-demes would have been easy to arrange. If there
was gerrym andering at this level, we cannot recover its details. The real problem
will have com e in the city, and here we are still almost without evidence to
determine the principles on which the deme-system was arranged.50
B. The Trittyes.
It will be helpful at this point to consider the present state of our evidence
on the names of the trittyes and their distribution on the map of Attica. To
save space and to avoid
parti-pris,
I shall almost always follow the views of
Kirsten,51 with the single qualification that I cannot follow his belief that it is
nearly always possible to draw a neat line on the map round the territory of
any individual trittys.
Erechtheis (I). The names of the trittyes are unknown. The city-trittys
included no true city-deme, and was mostly taken from the territory of the
old Paralia, without going down to the coast. Its principal demes were Euony-
mon and Agryle, the latter having Alcmeonid land. The coast-trittys was in
South Attica, south-east of Hymettos, formed mainly of the demes of Lamptra
and Anagyrous. The inland-trittys was in the far north-east of the Attic plain,
with Kephisia its most important element. Whether by accident or design,
the Boutadai, who held the hereditary priesthood of Erechtheus, were not
included in this tribe, which had a priest of Erechtheus appointed by lot.52
Aigeis (II). We have the name of one trittys, that of the inland, Epakreis.53
The word has some history. Strabo54 gives Epakria as one of the twelve cities
founded by Kekrops. Plutarch55 once uses Epakreis as the name of Peisis-
tratus' party, probably by simple confusion. An Epakria distinct from the
Cleisthenic trittys is also attested by Philochorus.56 The city-trittys included
50
Young (Hesperia XX 140-I43) produces some evidence to show that main roads
acted as boundaries.
61
The best map is that in Westermanns Atlas zur Weltgeschichte, p. 13, the best survey
of the position in Atti del terzo congresso internazionale di
epigrafia
greca e latina 15I-I71,
with Tav. XXVI. The evidence is scattered through Philippson-Kirsten, Die griechischen
Landschaften I 3, and must be traced through the index of sites there, pp.1065-1068.
52
IG I2 II 146. 4. 53 IG II2 2490. 8, II72. 30.
M
IX i. 20. 55 Moralia 763D.
56
328 F 206. This passage which puts the deme Semachidai, which belongs to the tribe
Antiochis, in the Epakria raises an unsolved problem. The only direct evidence bearing on
the site of Semachidai is the mention of a Semacheion near Laureion in a mining-lease
(IG II2 1582. 54). If this is a shrine and not a mine, does it place the deme in South Attica?
Trittyes of Antiochis lie both near Laureion and on Pentelikon, and are no help. This is
the only piece of evidence for the Southern Diakria envisaged by Ure, op. cit, and I would
doubt whether it was ever really legitimate to argue from Epakria to Diakria.
28 D. M. LEWIS
Kollytos, which is actually in the city, south-west of the Agora, but it also
ran far enough to the north-west to take in Kolonos Hippios and to the north-
east to take in Ankyle. It seems unlikely that it was fully contiguous.57 The
coast-trittys was on the east coast, from Brauron north to Myrrhinoutte. The
inland-trittys was oddly shaped, lying mostly on Pentelikon, with a long
tongue south into the Mesogeia plain to include Erchia. These two trittyes
touch, and it is perhaps even doubtful to which trittys Ionidai and Teithras
should be assigned.
Pandionis (111). Here we have three trittys-names, known and straight-
forward. The city was Kydathenaion,58 the coast Myrrhinous,59 the inland
Paiania.6? The city-trittys included the Akropolis and therefore the shrine of
Pandion, and was formed only of the deme Kydathenaion. The coast-trittys
was a powerful block south of Brauron, taking in part of the south-east of the
Mesogeia plain. It seems to have touched the inland-trittys at an angle. This
trittys was composed of the north of the Mesogeia, with the exception of
Erchia. The strange feature of this tribe, to which we shall return, is its in-
clusion of Probalinthos, which seems likely to have been at Xylokerisa, on the
coast immediately south of Marathon and separated from the rest of the
coast-trittys by the whole length of the coast-trittys of Aigeis.
Leontis (IV). We have the names of the city-trittys Skambonidai and the
coast-trittys Phrearrhioi; the inland-trittys, six letters in the genitive, evades
identification.6' The city-trittys has Skambonidai in the city itself and some
of the plain to the north. Part of the northern section of the city-trittys of
Aigeis seems to intervene. More oddly still, Halimous, well down the coast to
the south-east, is completely separate from it. The coast-trittys has the
southern end of the east coast down to Sounion, but the territory of Thorikos is a
bite taken out of it, which effectively divided it into two. The inland-trittys
has the eastern slopes of Parnes and a portion of the north-west Attic plain.
As will be seen, it is difficult to make Hekale, on Pentelikon, join it.
Akamantis (V). Here we begin to run into difficulties, for we have four
trittys-names. Cholargos was the name of the city-trittys ca.
450,62 but it is
difficult to see how IG 12 883 (ca. 420) can be restored as anything but
[Kep]a,cI&,v.63 The coast-trittys was Thorikos," the inland-trittys Sphettos.""
On the face of it, the city-trittys was renamed between 450 and 420.66 These
two names reflect its mixed composition. Kerameis was certainly at least
partly in the city; the rest of the trittys was a part of the plain including
Cholargos. Unless one draws one's lines in a very arbitrary manner, the trittys
was certainly split. The coast-trittys only contained the deme of Thorikos. The
inland-trittys had the whole of the south Mesogeia, except Myrrhinous.
57
See BSA L i6-17. 58 IG
112
I748. I4.
59 IG
12
898.
60 IG
12
898,
112
1748.
61
Hesperia IX 54, XXX 265.
62
IG
12
goo. 6 See Meritt, Hesperia IX 53-54.
64 SEG X 371. 65 SEG X 370. 66f So Raubitschek, AJA LX 28I.
Cleisthenes and Attica 29
Oineis (VI). Here we have two certain trittys-names, Lakiadai67andTlriasioi,68
presumably the city and coast trittyes. A third appears on two stones which
appear to confirm each other in the restorations
1L[8]tov
and [H?
e G[OV].69
This was hield by Wade-Gery to be the inland-trittys, but he felt the difficulty
of this view, since the inland-trittys was composed entirely of the deme of
Acharnai and Thucydides70 seems to draw a distinction between Acharnai and
the pedion. One cannot exclude the possibility suggested to me by Mlr. Forrest,
that this is a case analogous to Plhilaidai, a stroke of malice on Cleisthenes'
part, putting the trittys Pedicis wlhere there were manifestly no pedieis in the
party-sense. However, the discovery of a name-change in Akamantis opens
another possibility here, for of the two stones which give the evidence for
Pedieis, one is certainly,7" the other possibly, earlier than the stone which gives
Lakiadai. A solution which makes Pedieis an earlier name for Lakiadai seems
more attractive. For this city-trittys will be the heart of pedieis country, the
plain west of Athens towards Korydallos, reaching Athens along the Sacred
WVay through Lakiadai and Boutadai. The coast-trittys is only partly coast.
It includes the eastern part of the Thriasian plain, but reaches right up on to
Parnes to include Phyle. The inland-trittys, as we have seen, is only composed
of the deme of Acharnai.
Kekropis (VII). The trittys-names are unknown. The city-trittys was
composed of Xypete and Melite, which cannot be made to join.72 The coast-
trittys was south of Hymettos, and mostly the territory of Aixone. The inland-
trittys is geographically uneven, with parts of the north-east Attic plain and
south-west Pentelikon.
Hippothontis (VIII). One inscription73 gives us two trittys-names, Piraeus
for the city, Eleusis for the coast. Another,74 much earlier, gives us a third
name, a long one, beginning with Ts or Ze. Wade-Gery, to whom we owe the
restoration,75 assigned this automatically to the inland-trittys. But no suitable
restoration is available, and the possibility of a change of name allows the
chance that this may be another name for the city or coast trittys. A new
suggestion is possible, and the argument may lead us to one. The city-trittys
was Piraeus with the land west along the coast and Koile in the city itself. It
may be possible to join these elements, but it hardly seems necessary. The
coast-trittys is Eleusis and the west of the Tlhriasian plain, but it goes up on
to Parnes to include Oinoe. The inland-trittys is the extreme east of Parnes,
Decelea, Sphendale and Oion, very widely spread in an area of thin population.
67
IG. J2
884. 68 IG 12 899.
69
Wade-Gery, Melanges Glotz II 884-6; Meritt, Hesperia IX
55, XXX
265.
70
II 20. I 71 IG 12
899 has three-bar sigma.
72 BSA L 17. My observations on the location of Melite are, I am told, confirmed by an un-
published inscription from the shrine of Artemis Aristoboule.
73
IG 12 897. 74 IG 12 90I. 75 Mllanges Glotz II 886-887.
30 D. M. LEWIS
The tribal centre, the Hippotlhontion, was, abnormally, not in Athens, but in
Eleusis.76
Aiantis (IX). Here we have one name, for the coast-trittys, Tetrapoleis.77
The city-trittys was composed of Phaleron alone. The coast-trittys will be
discussed later. The inland-trittys was in the hills east of Parnes, mostly the
territory of Aphidna. The tribal centre was not in the tribe at all, but at the
Eurysakeion in Melite of Kekropis.78
Antiochis (X). Here we have two trittys-names, Alopeke for the city,79
Pallene for the inland;80 Anaplilystos hias been plausibly suggested for the
coast.79 The city-trittys was coimposed of Alopeke alone. The coast-trittys lay
in south Attica, running inland from the west coast of Cape Sounion. The
oddly-shaped inland-trittys covers the Pentelikon-Hymettos gap; its centre
was Pallene, which blocks that gap.
In this survey we have on the whole assumed that our evidence, which is
mainly fourth-century, is valid for Cleisthenes' reform. We should note that
not much of our evidence for the tribal affiliation of demes is earlier than 450,
although some of our evidence for the affiliation of trittyes is earlier. We have
seen some reason to believe that some trittyes were renamed between 450 and
420, but I do not think that this should deter us from employing the only
available working hypothesis, that there was no major reorganisation. Such
a reorganisation should not in any case be inferred from the manuscript
reading of Hdt. V 69. 2, which seems to assert that there were only one hundred
demes in the time of Cleisthenes. It is indeed theoretically possible that
splitting will have increased the number of demes, but the reverse process wil
have had to apply to Aiantis, which never had more t han six demes.8'
C. The Anomalies.
It should now be clear that Cleisthenes did not just draw lines on the map.
Such lines are difficult to draw, and break down notably in the city. Why the
city is organised as it is, we cannot tell, and our questions will produce few
profitable answers. Our best line of approach is through the anomalies outside
the city, demes wlhich appear to be enclaves, detached from their trittyes. If
we can gain some insight into the reasons for these anomalies, we may find
that we have some light on the reform as a whole. There seem to be three
notable enclaves worth discussion, Probalinthos, Hekale, and Halimous.
For Probalinthos the explanation seems clear, and Kirsten82 has already in
76
IG II2 1149, 1153.
77 IG
12
goo.
78 Hesperia VII i8.
'9
Hesperia XXX 264. 'I SEG X 374-
81
I should however draw attention to Raubitschek's
arguments (AJA
LX
28I)
for a
change in or regularisation of the official order of the tribes in the middle of the fifth
century (I believe IG 12 943 to be ca. 447 rather than ca. 440).
82
Atti (see n. 51) i62.
Cleisthenes and Attica 31
part given it. One of the cities of Kekrops83 was the Tetrapolis, wllich con-
tinued to exist as a separate territorial cult-organisation, sending independent
embassies to Delphi and Delos, down to the first century B.C.84 It was composed
of the four units, Marathon, Oinoe, Trikorynthos and Probalinthos. These
demes will have been fairly close together, and a suitable site for Probalinthos
has been found at Xylokerisa, confirmed by two gravestones85 and a dedication.86
The antiquity and deep roots of this organisation are confirmed by the list of
its sacrifices.87 A body which sacrifices to 'Awvocto 'E?RoL)d4 Zsu 'AvAa,
Xxo.i nocpc To: Meta?xou, raxLoq, tlle Neasvtm and a succession of name-
less lheroes and heroines is not an artificial construction. But the unit
formed by these four demes will not only have been territorially compact,
but also Peisistratid territory. Efficiency will demand that it has to be some
kind of administrative unit. It can even be called the trittys of the Tetrapoleis.
What could be done, however, and was done, was to detach from it the prosper-
ous deme of Probalintlios (providing five bouleutai in the fourth century88) and
to attach Rhamnous, which had totally different cults and doubtless different
traditions. Probalinthos is detaclhcd, but it is attaclhed, not to the contiguous
coast-trittys of Aigeis, but to the more distant coast-trittys of Pandionis. The
only explanation can be that there is little point in detaclhing it from Marathon
in order to attach it to the trittys wlhich contains Brauron. In its detachment
from both, it becomes an enclave, separating the two centres of Peisistratid
influence.
Here, on a larger scale, we see the same mind at work as we saw with the
Boutadai. The traditional unit remains and keeps its name, but by its side we
lhave a new unit, imposed from outside, with the same name, but of different
composition. Even this may not have been felt to be enough, and there is an
indication that steps may have been taken to improve the political situation.
For, by shortly after 490, it is possible to appoint at Marathon to supervise
a festival thirty men ?X ti6V C'MLMn.ov, representing each of the ten tribes.89
The ambiguity of the word allows the possibilities either that these arevisitors
for the festival or that they are new settlers, perhaps on confiscated property,
who have come to Marathon since the Cleisthenic reforms.
The second case is Hekale. Its site is a problem. It is somewhere near the
Tetrapolis on the road from Athens to Marathon,90 which probably puts it
somewhere on Pentelikon, and certainly well away from any part of Leontis.
It has been proposed"1 to put it at Kukunari on the north side of Pentelikon,
where in fact the sacrificial calendar of the Tetrapolis was found. There was
83
Strabo IX i. 20. 84 Boethius,
Die
Pythais, 38, 107.
85 IG 112
7292, 7304.
86 IG II2 7296. 87
IG
112 1358 (some revisions in Ath. Mitt. 1942, I2-I3).
88
e.g. IG 112
1700, 175I.
89 SEG X 2. I9-22. 90 Plut. Thes. 14, Callimachus F
230.
91
Milchhoefer, Untersuchungeni
uber die
Demenordnung
des
Kleisthenes, 21-22, Richardson,
AJA X 219.
32 D. M. LEwIs
a Hekalesia at Hekale, celebrated by oc irepI . 8NuLOL,92 and since the Tetrapoleis
are in any case so far inland, scholars have wished to equate the groups. But
Hekale is not part of the Tetrapolis, and although Theseus stayed there on
his way to fight the Marathonian bull, it is not stated that the Tetrapoleis
took part in the Hekalesia. Nor do Hekale or Ze'u 'Exoc)lato- appear on the
Tetrapolis-calendar, which is admittedly incomplete. It is probably better to
see in Kukunari merely a cultspot on a hill, and I see no real reason to deny
that Hekale is at or near the present village of the name, soutlh-west of Kuku-
nari, but still in the north-north-west of Pentelikon. This is still far from
Leontis, and the reason for detaching it from the nearby inland-trittyes of
Aigeis, Kekropis and Antiochis must be its importance as a cultcentre. If we
knew which oL Tcp'i 8n,uoL were who celebrated the Hekalesia, we would find
them in these trittyes. The suspicion that deliberate interference has been at
work in drawing the trittys-boundaries in this part of the world is strengthened
by a consideration of the shapes of these trittyes, all distinctly malformed.
The political motive may well lie in the position of Isagoras. We have already
seen grounds for placing him in Ikaria. There, as at Boutadai, a non-Cleisthenic
organisation survived side by side with the deme.93 The Cleisthenic trittys to
which it belonged retained the name Epakreis. We do not know the home or
domain of Ze'u
'E7c&xpLoq,14
but here again we may suspect that territory is
removed from it"- and that non-Epakria territory, Erchia, is brought in from
the plain below.
The third anomaly, Halimous, has been explained by Kirsten96 as belonging
not to the city-trittys of Leontis, but to the inland-trittys. The deme, it is
said, is composed of Alcmeonid supporters from near Leipsydrion,97 who
moved after 507 to Alcmeonid land, retaining their old trittys-affiliation.
However, the arrangement of a prytany-list suggests fairly strongly that the
deme belongs to the city-trittys,98 and there is no evidence that Alcmeonid
land came down to the coast, for the territory of Euonymon is in the way.
No clear answer can be given. This is an area of old settlement," with a very
92
Plut. Thes. I4. 2.
93
IG 12 i86 B 3, II2 1178. 4-5. If I had to make a firm guess, I would say that the
'IxocpteZq
were a phratry, standing to 6
ftLOq
'rv 'Ixaptfcwv, as the AexeXeetq stood in
relation to the Cleisthenic deme of that name (Wade-Gery, Essays in Greek History,
133-134). But there are perhaps other possibilities.
94 Et. Magn. 352. 49; perhaps IG 112 1294.
95
If Semachidai is in this area, this would be a certain case. See n. 56.
96
In Philippson-Kirsten I 984 and Atti i6i.
97
The case for supposing that Leipsydrion was in Alcmeonid land is in itself not cogent.
It rests on Herodotus' statement (V 62. 2) that Leipsydrion was ukrsp IIMtov[7 and the
story in Pausanias II i8. 8 that the Alcmeonidai and Paionidai came from Pylos together.
98 IG II2
I742,
as
analysed by Loeper,
Ath. Mitt. I892, 388.
99
Mylonas, Haghios Kosmas.
Cleisthenes and Attica 33
primitive Demeter cult on Cape Kolias.'00 Perhaps here too there is a desire
to detach a cult-site from its neighbourhood.
D. Other Religious Organisations.
The anomalies have indeed provided us with a clue, which suggests that
it may be profitable to investigate the effect of the reforms on other religious
organisations of a territorial character.
Let us begin with the Tetrakomoi. It is possible that they, like the Tetra-
poleis, figured in Strabo's list of the twelve cities of Kekrops,'01 a list which
certainly reflects some truth about early Attica. One name is missing, and the
conjecture that their name has dropped out after that of the Tetrapoleis102
is a plausible one. However this may be, the antiquity of their cult of Herakles
is not in doubt. The four components are given by Pollux'03 as Piraeus,
Phaleron, Xypete and Thymaitadai. The most likely interpretation of a fourth-
century inscription shows Phaleron still possessing a
xw6,.)-organisation,104
and in 330-29 they still joined in a festival.'05 They are therefore a real unit
with something in common. It is therefore interesting to find that Cleisthenes'
reforms leave them in three different trittyes, Xypete going to Kekropis,
Phaleron to Aiantis, Peiraeus and Thymaitadai to Hippothontis. It is reason-
able to suppose that the splitting was intentional. We may go a stage further,
and recall that Hippothontis has a trittys with an unrestored long name
beginning with Te or Ze. It seems not unlikely that this should be
Te[-rp0Cx6tov],
an earlier name for the Peiraeus trittys, which contains two of the old Tetra-
komon. The parallel with the Tetrapolis would not be exact, since the Cleis-
thenic Tetrapolis really contained four demes and the Cleisthenic Tetrakomon
should have at least five, Peiraeus, Thymaitadai, Koile, Keiriadai and Kory-
dallos. Either the word Tetrakomoi would not be felt to have numerical
significance, or perhaps we should regard Keiriadai or Korydallos as a splinter-
deme of later origin.
Pallene is the other main candidate for the missing city of Kekrops,106 but
the League of Athena Pallenis presents greater problems since its composition
cannot be precisely determined. The dedication quoted by Athenaeus'l07 gives
us three member-demes with relative certainty, Pallene itself, Gargettos and
loo Plut. Sol. 8. 4, Paus. I 3I. i; Nilsson, Griechische Feste, 317. 101 IX i. 20.
102
See e.g. Solders, Die ausserstadtischen Kulte und die Einigung Attikas, I07-1O8,
following Loeper. 103 IV
IO5.
104
IG
112
I598. gff. See Roussel, Revue Archiologique XVIII (I941) 226-231, though
he thinks that the two x(t)Lu which both have komarchs from Phaleron are sub-divisions
of Phaleron, and I incline to think that one is Phaleron and one one of the other three
components of the old Tetrakomon, which happens to be headed by a man or men on the
Phaleron deme-register. 105 IG II2 3I03.
106
It will have dropped out at the end before 7I&XV in the next sentence. See Solders,
op. cit. I I I. 107 234 F.
3 Historia XII,z
34 D. M. LEwis
Pithos, and the quotation from Themison which follows'?8 assures us that
there were a number of member-demes. Schlaifer'?9 adds Acliarnai to these
from the quotation from the UL?c4)lco
vo,lot
which lies between these passages,
but this seems an unnecessary addition, of no great topographical probability.
It seems more likely that Athenaeus is simply quoting from Polemon a
number of passages bearing on parasites."10 Some of these certainly have
nothing to do with Athena Pallenis, and I would add the passage on Acharnai
to their number. Even so, we are left with a cult-league centred at Pallene,
with the names of at least three demes which belonged to it. Under the Cleis-
thenic system there was a trittys named Pallene of elongated shape, from
which Gargettos and Pithos were excluded. This seems to conform well to
the pattern which we have established.
The Trikomoi,"' on the other hand, present a different picture. Their
three members, Eupyridai, Kropidai and Pelekes, are all in the same trittys.
We know virtually nothing of the unit, but may surmise from its situation and
the lack of references that it was of little importance. Nothing of importance
emerges from an examination of the Mesogeioil12 of the Paraloi,13 attractive
though their names are in this context.
Some minor points may be briefly noted. Kirsten 14 may be right to see
significance in the isolation of the ancient city of Thorikos. There is some
tendency to detach old cults of the tribal eponymoi from those tribes. The
case of Erechtheus is clear. An ancient cult of Leos is attested at Hagnous,1"5
which is nevertheless excluded from Leontis' 6. Similarly Erechtheis, Kekropis
and Aiantis have to go outside their tribes to worship their tribal hero,"7
although Pandionis, centred on the Acropolis, and Hippothontis do not."8
E. The Purpose of the Trittyes.
Our entire argument so far has tended to show that demes were allotted to
trittyes with some care, and that the trittyes were constructed in a deliberate
attempt to create units which would be sufficiently distinct from existing local
108
235 A.
109
Harvard Studies in Classical Philology LIV 35-67, a thorough discussion of the
evidence for the League. (But see Additional Note.)
10
See Jacoby's analysis of the passage, F. Gr. Hist. IIIb, i 147-150.
111 Steph. Byz. s.v. Ei7rup'L8ou; perhaps IG I12
12I3.
112
Schlaifer, Classical Philology XXXIX 22-27 has all the evidence and establishes a
prima facie case for considering them a genos. I am not very happy about this view. The
likely demes represented are Bate, Diomeia, Kydathenaion, and Kerameis, all city demes,
falling into three trittyes.
118
A very puzzling organisation. IG II2 1254 shows them with epimeletai sacrificing to
Paralos in the Paralion, but it is tied to the state-trireme by the tamias.
114
Atti I62. 115
&i6ve4
in Steph. Byz. s.v. "Ayvouv.
116
Contrast the cult of Leos in the Cleisthenic deme of Skambonidai (IG
12
I88).
117
IG
112
I146, II43; Hesperia VII i8. 118 IG II2 II38, I149.
Cleisthenes and Attica 35
units to compete with them and to destroy the influence wlhich they gained
from possessing a common cult in a common locality. Modern scholarship has
tended to neglect the trittyes, seeing in them mere devices to ensure mixed
composition of the tribes. This is an understandable tendency, since in the
event the tribes won a success which. can only be described as surprising,
while the trittyes totally failed to compete with the older local organisations.
There was a parallel failure in Cleistlhenes' attempt to redefine the
qualifi-
cations for citizenship. He put the deme beside the phratry, but could not
persuade the Athenians that phratry-membership was of no importance.
Pericles had to recognise this basic unwillingness to accept the substitute.119
However, the fact that the trittyes failed to establish themselves as local
units does not mean that they were not intended to have importance. The
mass of fifth-century horoi from the Agora and from the Peiraeus indicates
that they were intended to serve real purposes. The purposes are indeed
mysterious. Raubitschek'20 has recently suggested uses for the Agora markers.
The Peiraeus horoi presumably served as muster-stations on the lines that
Demosthenes was trying to revive as late as 354.121 The division of the
prytaneis
into trittyes survived into the time of the Ath. Pol.,'22 and the epimeletai of
the tribes were selected by trittyes in the fourth century.'23 The trittyes
retained property'24 and cults.'25
Another substantial reason for modern depreciation of the trittyes has
been their apparent random distribution on the map of Attica. Little attention
has been paid to the enclaves, but a great deal of attention has been given to
the contrary situation, where blocks appear to have been left untouched by
the redistribution. The importance of these blocks can be overstressed. As
Ehrenberg has said, Cleisthenes was interested in people, not territory.'2 But
even the territorial arguments lose some of their force, when examined on a
contoured map. Let us work through a recent list of tlhese blocks.127 It is not
true that the inland and coast trittyes of Akamantis join. Thorikos is an
enclave in Leontis, and the territory of Phrearrhioi lies between it and Keplhale,
the nearest point of the inland-trittys. In Pandionis, the coast and inland
trittys touch at a point, not at a line, and the centre of population of the inland
trittys is at Paiania, far to the west. In Aigeis, the trittyes are contiguous, but
they are quite different in character. The coast-trittys is composed of coast-
towns and plain; the inland-trittys is mostly hill, and has been deliberately
119
This has been demonstrated by Andrewes, JHS LXXXI
I3-I4.
120
AJA LX 279-282. 121 XIV 22-23. 122 Ath. Pol. 44. i; cf. IG I12 1748.
123
So in IG
112
1151, 1152, 28I8. I retain a suspicion, however, that
[&Ir T]X(
is
the correct restoration in IG I12 2824, where all three come from the same deme. None
has ever produced an alternative which is plausible.
124 IG
II2 2490.
125 IG I2
90,
II2 I
I72.
126 Neugrunder des Staates,
go.
127
Hignett, History of the Athenian Constitution, 134.
3.
36 D. M. LEWIS
deformed. The case which is invariably quoted is that of Aiantis, where a flat
map shows a solid north-eastern block. This of course was the case which made
Beloch suspect Peisistratid influence, and a solid Peisistratid block in the
north-east would contradict the general drift of our enquiry. But the case
does not hold, either territorially or politically. The centre of population for
the inland-trittys is at Aphidna, well up in the hills, a long and tedious climb
from Marathon, while the evidence for interference with the composition of
the coast-trittys is clear. Politically, it may indeed be true that Marathon
remains a centre of Peisistratid influence. Can this really be the case with
Aphidna, where we know of two leading families, the Gephyraioi, the family
of Harmodios,128 and Kallimachos the polemarch who led the Athenians at
Marathon ? 129 If Aiantis was a Peisistratid-packed tribe, they had forgotten it by
the time they led the right wing in 490.13?
There is indeed a certain contiguity of trittyes, but this is to be expected
if the trittyes were, as we are told,'3' assigned to the tribes by lot. Provided
that the trittyes were properly constructed in themselves, any random allo-
cation would do. The only necessary qualification was that each tribe should
have, a city, a coast and an inland trittys, 6-r& Ex tr.'k6 RE &XT v 6T V
r67r&v. The trittyes cut across the most obvious local loyalties. It could be
left to Apollo to create the tribes, to produce the wider loyalties to be im-
planted when men served together on the boule and, more importantly, in the
army. In the event, the men of Phaleron fought side by side with those of
Marathon and Aphidna, and Leontis united Sounion, Skambonidai, and
eastern Parnes. Eleusis, Kerameis, and Aphidna might have been better still,
but the lot would produce adequate results, and in fact it taught the right
lessons.
III. Cleisthenes.
We have left some parts of the reform in the dark for want of evidence.
This lack comes partly from the paradox I began with. We know about local
cults, because they survived, because Cleisthenes failed with them. Other
institutions did not retain their hold, and we know so little of the old
tribes)
the old trittyes and the old naucraries that our picture may have been sub-
stantially distorted. The naucraries in particular may well have had territorial
implications'32 and Cleisthenes does not seem to have abolished them.'33 But
there are other patches of darkness. Some have recently seen in Cleisthenes'
moves a desire to end the old threefold distinction of caste.'lu Many have found
128
Plut. Moralia 628 D. 129 Hdt. VI IO9.
130
Plut. Moralia 628 E.
131
Ath. Pol. 21. 4.
132
The one name we possess, Kolias, suggests as much (Bekker Anecdota p. 275).
133
Kleidemos 323 F 8.
134 Wiist, Historia III 137-9, VI I 76-191, VIII I-I I, accepted by Oliver, Historia IX 503.
Cleisthenes and Attica 37
the evidence that Cleisthenes desired to bring in new citizens unconvincing.135
Perhaps most important of all, we have next to no evidence on the confiscation
and redistribution of land in sixth-century Athens.136 The two things which
emerge most clearly are an attack on organisations which held a locality by
religious ties, some of them in areas attached to political opponents of Cleis-
thenes, and an attempt to unify Attica by making men from different areas
work and fight together.
How far was Cleisthenes a politician, how far a statesman? The Alcmeonids
in the sixth century had been caught between the fires of the autochthonous
landed population of the plain, secure in the possession of the most important
state cults, and their fellow-Pylian Peisistratus, who in the east had built
himself a remarkable coalition of varied interests and had secured himself in
the tyranny by promoting economic prosperity and by giving Athena wider
appeal without the aid of her traditional guardians. The Alcmeonids possessed
no important local cult of their own that we know of. Their power rested
solely on land and wealth, which they buttressed by looking outside Athens
to Delphi, Sicyon and Lydia.'37 In the struggle for power before Peisistratus,
they had been squeezed out and forced to rely on their position outside
Athens. Cleisthenes had watched this process under his father Megakles and
Megakles' successor, his uncle Alkmeonides.138 The veering of tyrannical
favour which brought him home and to the archonship in 525 made no essential
difference to the lesson. He had learnt the power of local cults and the value of
a wider Athens.
When he came home finally in 510, the position had been transformed by
Peisistratus and his sons. New classes had been given economic hope, and men
had learnt to look to Athens as the undisputed centre. There were features
135
Wade-Gery, Essays in Greek History, 148-9, but more firmly and less convincingly
Oliver, op. cit., who, in order to maintain not only that there was no admission of new
citizens, but that none ever said that there was, has to offer an impossible meaning of
uX&rTeae and an extremely unlikely one of veooM'roct.
138 The only pieces of evidence worth considering are these: i) redistribution is said
to have been in the air at the time of Solon's reforms (Plut. Sot. i6. I); 2) Peisistratus'
enemies confiscated his property and auctioned it while he was in exile (Hdt. VI 121. 2);
3) Peisistratus made money-grants to farmers (Ath. Pol. i6. 2); 4) when he recalled Kimon
Koalemos from exile, Kimon xiTXf bdt 'r& &uro5 (Hdt. VI 103. 3);
5) Peisistratus'
friend Lygdamis of Naxos did indulge in confiscation (Arist. Econ. i346b 7). It is clear
that Peisistratus had the power to confiscate and it would have been surprising if he had
not used it, but the positive evidence for saying so does not exist.
137 I do not wish to be understood to say that they had mercantile interests. I know of
no evidence for that proposition, widely-held though it is.
138 I see no good reason to disturb the manuscript reading at Hdt. I 64. 3. For Alk-
meonides, see Raubitschek, Dedications from the Athenian Akropotis, pp. 338-340, where
the evidence is collected. I cannot follow him in supposing that there is any strong case
for supposing Anaxileos a member of the family.
38 D. M. LEWIS
here which he may well have disliked and which were certainly disliked by
other men of power. A
8&mkyai?,o'
cut into the tyrants' enfranchisements.'39
Other moves may have cut into their land-distributions. Cleisthenes' main
opponent came from Pentelikon, deriving strength from the close-knit religious
unions of that area, perhaps from the prestige of the new dramatic festivals
of Ikaria.'40 Cleisthenes had Delphi behind him, but so far Delphi had only
been of use as a lever to bring in Sparta, and Cleomenes of Sparta backed
Isagoras. Isageras, with Spartan backing and using the methods of the 560's,
won the upper hand.
If there was to be a place for the Alcmeonids in Athens, new methods had
to be used. One lesson could be learnt from the tyrants, and Cleisthenes turned
to the demos which the tyrants had created and favoured, which he and other
dynasts had previously rejected. It is hard to see how the complexities of the
reforms could have been used in themselves as a bargaining-counter to win
the demos, even at the lowest level, the prospect of local self-government in the
deme. Something more concrete will have been in the air-citizenship, probably
also land. Land-hunger remains strong in Athens for the next fifteen years or
so. Salamis,'4' Chalkis'42 and Lemnos,'43 all are attempts to meet a need which
Cleisthenes could only partly satisfy in Attica, since land to satisfy it could
only come from his opponents, not from his supporters.
At this point constitutional problems have been multiplied. How could the
reforms have been put through while Isagoras was archon, able, as presiding
officer, to block a vote? Was Cleisthenes an archon ? Did he hold a special
commission? How could he have obtained a majority, if his supporters had
been disfranchised? These problems seem irrelevant to a revolutionary
situation. Whether Isagoras was in the chair or not, whether the neopolitai were
legally entitled to vote or not, there was nothing to prevent them carrying clubs
and standing round or even on the Pnyx to shout "All power to the ten tribes."
Cleisthenes needed a new system. The
8tcL4rcpLa[.ot6
had shown the difficulty
of providing a legal basis in the phratry-system for doubtful citizenship, and
to intervene directly in the phratries might alienate as many voices as it won.
The creation of the deme met this difficulty, but there still remained the
problem of local vested interests. This could be met by establishing the trittyes,
which might curtail the powers of local organisations. The real problem came
in deciding how to crown the pyramid, how to ensure that the new local units
served interests wider than local ones and looked to Athens in the way Peisis-
tratus had shown. The four Ionian tribes seem now to have represented very
little."44 Changing tribal organisation was not unparalleled in the sixth century.
1'I
A th. Pol. 3. 5. 140 Pickard-Cambridge, Dithyramb, Tragedy, and
Comedy pp. 97ff.
141 Tod, GHI
iI.
142 Hdt. V 77. 2. 143 Hdt. VI I40. 2;
BCH XXXVI 329-338.
1"
I deduce this only from their lack of success in surviving Cleisthenes. One sacrifice
(Hesperia IV 5ff.) and the phylobasileis (Ath. Pol. 57, 4) exhaust the evidence.
Cleisthenes and Attica 39
Cleisthenes' own grandfather, Cleisthenes of Sicyon, had changed tribe-names,
and may have created a new tribe from the under-privileged.145 At any rate,
hehad a tribe for the previously under-privileged, and asserted the principle
that it was better than the others. At Cyrene, Demonax of Mantinea had
acted rather differently.146 Finding the original settlers organised in the three
Dorian tribes and the newcomers not organised at all, he created a new three-
tribe system on ethnic lines. These systems had given new citizens or settlers
rights, but retained the seeds of difference and dissension. Cleisthenes realised
that there was no possibility of peace and concord without mixing his dis-
parate elements. The novelty of his solution lies in its determination to make
a fresh start. All citizens, old and new, would start equal in his new demes and
new tribes.
This is his most important service to Athens. Traditionally, and perhaps
temperamentally, averse to tyranny, he found a way to create a pyramid of
power which would have at its head, at least in theory, the people and not one
person, and which would maintain the unifying force generated by the tyranny
without a tyranny's disadvantages. The way he chose had distinct advantages
for himself and his family. The details of his settlement could be and were
manipulated to the disadvantage of traditional opponents, but the nature of
the Alcmeonid position was such that it could not be affected by this settlement.
The trittys-lines were drawn to leave Alcmeonids in at least three different
tribes, a result which might well prove fatal to a family which depended for
influence on the control of a local cult, but which could be positively welcomed
by a family of land and wealth which acquired the opportunity to have a hand
in the affairs of three of the new tribes. But it cannot be denied that, politician
though he was, Cleisthenes was capable of seeing the advantages of Attic
unity. vC4LLxq would have been the key word in his thought, as it was in the
language of his most sympathetic commentator.147 Ttr E xc1 t& ToLo&Tc
145
Hdt. V 68. That the 'ApXAbooL were a new tribe seems to follow from the fact that
a new name has to be found for them when the other tribes reverted to their Dorian names.
AMytoxXet is explicitly pro-Adrastus and therefore anti-Cleisthenes.
146f Hdt. IV I6i. 3; cf. Chamoux, Cyrene sous la monarchie des Battiades, 138-142,
Jeffery, Historia X I42-144. Miss Jeffery's preferred explanation, that Demonax, like
Cleisthenes, included a cross-section inside each of his three tribes, does not seem to me
a possible interpretation of the text.
147
Twice in A th. PoI. 2 I, apart from the passage which follows, which is Politics I319 b I9.
ADDITIONAL NOTE. In discussing the League of Athena Pallenis, I forgot the fourth
century inscription Atth. Mitt. 1942 no. 26, pp. 24-29. This complicates the matter, since
it shows that parasitoi then came from a very wide area and makes the
parasitoi
previ-
ously known unsafe guides to the composition of the League, which may or may not have
been narrower originally. The four archontes of the inscription may be a safer guide; they
comc from Gargettos, Acharnai, Pallene, and Paiania, all four from different trittyes and
tribes.
40 D. M. LEWIS, Cleisthenes and Attica
XOCXaW0XeU&a1CM'rM
XplqcaL(X
7trpocJ
'
- pv
L%XpMTLV
rv
tOLOCfrjV,
ot7
K)LtaWvjq
'r 'AOeviatv
XpIaxoto
3otXA6vop
gocg
aoux
Av
8noxpax-av, xodt repl Kupyv-v
ot 'rov 8nuov x0C'Natvr0'ao . cpiXaoC 'e yap 9epoTL 7CoL'raOXL
7Tr)dOu4
xoca (po%pLaL,
XCXL TOC TCiV L8LG)V LepLV aliVCVtkiOV ?El OXLyO XOCL XOLVM, XOaL 7MV70C aOCai*OV 67rro
&CV 6TL
1DXGTa
OC V060( XaCet L 7XOTV'M X);AnXoLc, aL 8S aUvM&LOCL
&oCeUXI&(L)Lv
xL 7rpo6repov. Every increase in our knowledge shows that Cleisthenes worked
towards this aim.
Christ Church, Oxford D. M. LEWIS