K.H. Kinzl home page TLG code for Greek text
A Study of the Early History of the Term1
Gymnasium. Zeitschrift für Kultur der
Antike und humanistische Bildung,
© Carl Winter Universitätsverlag Heidelberg
[Best results with Netscape 4.7]
Recent scholarship has attempted in a growing number of investigations of Athenian Democracy either direct or indirect to probe its problems to the limits of human thought; scholars from a variety of disciplines and of varying political orientation have added their voices to the discussion.2 If there is any justification for yet another study on this theme,
it is my attempt bring the debate back to the basis of the sources. My fundamental position is that of a minimalist sceptic; my aim is, as it were, of a negative nature. It is my desire in this way to be able to present a more rational reality. The impetus came from two investigations which were characterised by contrasting methods. On the one hand a historical study of the political developments in Athens from the early twenties of the sixth century to the late sixties of the [fifth] century; on the other hand a philological reexamination of democratic terminology in Herodotos. The main conclusions of the historical study which appeared elsewhere3 will here be taken for granted.
The present investigation takes as its starting point three problems (or, rather, clusters of problems), i.e., Herodotos usage of terms such as DHMOKRATI/H; how to interpret the text of Hdt. 6,131,1; and [above all] the plain and incontrovertible fact that the compound noun DHMO-KRATI/A represents a linguistic anomaly.
In the course of our investigation we shall attempt to prove that the noun DHMOKRATI/A was coined some two generations before its first attested occurrence, and that it did so not in the narrowly delimited and often derogatory sense of rule by the people [(cf. the German compound Volksherrschaft)] which begins to emerge comparatively late, in the context of the intensifying political debate of the last decades of the fifth century. Finally I shall put forward [a hypothesis] according to which the etymology and origin of this compound noun may differ from the one traditionally assumed.
No one will of course seriously suggest that Herodotos says that Mardonios established democracies mirroring Athenian de-
mocracy of his own age (nor that Otanes had a vision of this at the time non-existent form of government); the purpose, however, of using the stylistic device of interconnecting components4 in his narrative is equally obvious as the demonstrative juxtaposing on several occasions of being under the yoke versus being free. We shall return to the only other occurrence of the term, at 6,131,1 below (part II). The derivative verb DHMOKRATE/ESQAI serves the same purpose (see below, p. 122).
Two of the four occurrences belong to the famous constitutional debate which we can treat here only as briefly as possible. Let me state that in my judgement justice is not done to Herodotos artistic and intellectual talent if one assumes that Herodotos only recreated this whole artfully composed scene which he found in some unknown source (Quelle). I am also not prepared to believe that Herodotos reports history as an objective recorder (even though one must admit that in the context of the authentic historical conspiracy there will have been a good deal of debate amongst the conspirators). I regard it as totally unbelievable that the reason why we encounter the term I)SONOMI/H, rather than DHMOKRATI/H, is that Herodotos had not known the latter term at the time he wrote this passage in book three. Thus the interpretations by Vlastos and Ostwald are likely correct in noting the qualitative difference of the terms and in observing that I)SONOMI/H does not represent a label for a form of government.6
It is particularly remarkable, however, that the great speech of Otanes
(3,80,2-6) does not take the form of a plea in favour of the
constitution preferred by him but, rather, a passionate
condemnation of MOUNARXI/H; his few positive words by contrast read like
dry text book wisdom:
We are thus dealing with a concept of a form of government which is grounded in the principle of [TO\] I)=SON and to which therefore an OU)/NOMA KA/LLISTON attaches something clearly ruled out in the case of MOUNARXI/H. But this does by no means suggest that this is a plea [for] DHMOKRATI/H. Rather, we are dealing with a comparatively vaguely defined concept of some kind of majority government, perhaps more along the lines of the Spartan kind: a majority based administration and a council are stressed, with a people s assembly mentioned only as a last resort (PLH=QOS A)/RXON, BOULEU/MATA, TO\ KOINO/N). We may therefore fairly safely conclude that Herodotos could not, and certainly would not, have used the term DHMOKRATI/H in this context.
It is a revealing exercise, however, to track the terms pertaining to constitutional law which [Herodotos] puts in the mouths of the various speakers. As we have seen, Otanes uses the three terms MOUNARXI/H, PLH=QOS, I)SONOMI/H (except in 3,80,4, were we encounter an A)NH\R TU/RANNOS). The next amongst the conspirators to speak favours oligarchy: O)LIGARXIH| E)KE/LEUE E)PITRA/PEIN (3,81,1). Megabyxos chooses his words more pointedly; his three terms consist of TURANNI/S, DH=MOS, A)/RISTOI. The more moderate aspects emphasised by Otanes are presented by his opponent in more extreme terms: government by one (MOUNARXI/H)
(it must not be ignored that it is precisely this form of government which will ultimately emerge victorious) is condemned as TURANNI/S; the ruling body, a moderate majority (PLH=QOS) in Otanes speech, is transformed by Megabyxos into an all powerful peoples assembly (presumably to convey a negative meaning), i.e., DH=MOS; the principle of [the] I)=SON in government (in Otanes choice of words an OU)/NOMA KA/LLISTON, i.e., I)SONOMI/H) is b rushed aside with the argument that A)RI/STWN DE\ A)NDRW=N OI)KO\S A)/RISTA BO ULEU/MATA GI/NESQAI (3,81,3).8
Because Megabyxos favourite oligarchy apparently cannot be apostrophised using the OU)/NOMA KA/LLISTON, i.e., I)SONOMI/H, Herodotos is able skilfully to turn to the speech of Dareios and the dismissal of the proposals of those who had preceded him in the debate; on the other hand the use of I)SONOMI/H for contrast becomes evident. Herodotos is not in the first instance intent on having Otanes making a well argued and substantive constitutional proposal but, rather, to create a sharply contrasting background against which to view any form of government which does not accord with the principle of equality be it MOUNARXI/H, TURANNI/S, or non-isonomic oligarchy , i.e., to underscore the contrast between freedom and arbitrary rule.
The same somewhat free usage may also be observed in the two remaining passages where Herodotos uses the word I)SONOMI/H. Both Maiandrios Maiandriou of Samos and Aristagoras of Miletos create I)SONOMI/H for their citizens after the removal of their respective tyrant predecessors.
Neither of these last two passages justifies the conclusion that Herodotos meant a form of government which could be defined as democratic. The least forced interpretation of all passages in which I)SONOMI/H occurs is that the focus is on the stark contrast between a condition of freedom versus one of subservience to a despots régime of terror.
This rather uncommon term is found in the speech of Soklees of Korinthos before Sparta s assembled Peloponnesian allies in which he condemns the attempt to return Hippias to Athens by the use of military force (5,92A,1).
The term is clearly not used as a synonym for DHMOKRATI/H: the plural would in this case make no sense. Even Herodotos surely would not have suggested that Soklees could have held the view that there existed democracies throughout Greece (this in the last decade of the sixth century). Ostwald9 offers the attractive suggestion that I)SOKRATI/H is put in Soklees mouth because it represents the common denominator of the régimes of Corinth, Sparta and Athens. This would place Ostwalds interpretation in the company of our reading of 3,80,6. The antithesis of repressive despotism and the condition of living in freedom thus finds its most compelling expression.
The pseudo-Xenophontean Athenaion politeia for instance speaks of I)SHGORI/A of citizens and metics. It is therefore impossible to conclude that Herodotos alludes with this word, which occurs only at 5,78 (on this cf. below[, 123]), to the constitutional right of every citizen to speak in the assembly, as a term used in constitutional language (there would be no doubt in this regard, however, had he used PARRHSI/A).
(e) [Related] Verbs [and verb like expressions]
Verbs or phrases with the function of a verb10 occur in several instances but only DHMOKRATE/ESQAI derives from one of the nouns so far discussed: 6,43,3 (see above[, 118]) and 4,137,1 (presentation of the reasons why Histiaios thought that Miltiades counsel to destroy the bridge across the Danube was ill conceived) are both passages where DHMOKRATE/ESQAI (along with DHMOKRATI/H at 6,43,3, and E)LEUQERI/H at 4,137,1) sounds like a counter point to TURANNEU/ESQAI. Further expressions of this kind may be found in the shape of, e.g., (KATA-)TI/QESTAI E)S ME/SON (TW|= DH/MW|) TH\N A)RXH/N (uel TA\ PRH/GMATA, TA\ A)/LLA PA/NTA, cf. 3,80,2; 142,3; 4,161,3; 7,164,1). It is always the contrast to a condition of previous oppression which emerges as the common denominator.
Fundamental to Herodotos work is the theme of freedom from oppression, which he sees as the root cause for the rising fortunes of the people of Athens, as an expression of their determination in pursuing it as a political goal, and arrived at by their assembly by majority decision. Herodotos builds a bridge, as it were, from the expulsion of the despot to Athens role as the leader in the fight for Greeces
The theme is first introduced in but few yet all the more resonating
The second stage is reached with [the Athenians] determination in
their opposition to Isagoras and his Spartan ally:
It is that same Athenian peoples assembly which responds to the
Ionian plea for assistance by dispatching those notorious 20 ships, which
constitute the A)RXH\ KAKW=N (5,97,3); accordingly it is the Athenians who
are forced to fight against the Persians for freedom on Greek soil, in
Attika, at Marathon. There, in the Athenian citizen soldiers
encampment, with the Persian invasion force in plain view, it is Miltiades
who puts forward these ideas once again in his endeavour to win over the
archon polemarchos to his strategy of forcing the decision in the
Marathonian plain: not only would Kallimachos win for himself greater
glory than the tyrannicides Harmodios and Aristogeiton (6,109,3); Athens
would emerge the leading polis amongst the Greeks:
And so Herodotos, turning up the volume step by step, as it were, and
putting his ideas in the mouths of a variety of persons,[10a] builds the tension up
to the climax of the famous and often cited passage in book seven where he
expresses these same ideas emphatically as his own:
It is not surprising that Herodotos terminology fails fully to satisfy both the constitutional expert or the philosopher (or to cause them to construct artful edifices of complex ideas). Herodotos aims are those of the artist and writer. As such he endeavours to convey a sense of his more general concept of the superiority of [the rule of] law and of rule based on law needless to say, there exist governments which fulfill this expectation and there are those which do not. When he uses terms such as I)SONOMI/H, I)SHGORI/H, and even DHMOKRATI/H, and phrases such as E)S ME/SON KATATI/QESQAI TA\ PRH/GMATA as well as DHMOKRATE/ESQAI in all these instances it is the antithetical concept, the contrast between despotism and the condition of freedom from it, enjoying an constitution based on the rule law and equality, which Herodotos strives to place centre stage by employing all the skills of the writer[, rather than those of the jurist or philosopher.] In no instance can it be said that he attempts to impress on his audience the Attic type of democracy as he observed it [as the one and only possibility of achieving this condition of freedom from despotism].
The adjective I)SO/NOMOS which the linguistic laws of the Greek language
require as the precursor for the formation of the noun [I)SONOMI/A] can be
traced to the first years of the fifth century or even to the late sixth
century: to two of the Harmodios-skolia which (together with the
other two skolia which do not, however, contain this word) were even
elevated to the status of a national anthem [by Ehrenberg].
Surely no one would have sung praises of the tyrannicides (plate I)11 as the ones who
if their deed did not fundamentally affect the history of Athens that much is certain, regardless of whether we follow Herodotos and Thukydides less than positive assessment or not. And if this is so, the new condition in Athens needed to be characterised most suitably by
an adjective, since the skolia only run to four brief lines each: *A)QH/NAS I)SONO/MOUS. In order to have a meaning, the adjective also ought to express some kind of contrast. We are fortunate in knowing from Thukydides (6,53,3ff.) that the man on the street (who owes his historical knowledge to the equivalent of the [boulevard press]) did indeed believe that it was Harmodios and Aristogeiton who deserved the credit for liberating Athens.
Now Vlastos[11a] pronounces that to the best of his knowledge no one would hold as improbable a view as to suggest that I)SO/NOMOS is a plain little adjective, meaning no more than liberation from despotism, without emphatically identifying by name the form of government which followed. Herodotos, however as we have seen , uses the noun derived from [I)SO/NOMOS (i.e., I)SONOMI/H)] in precisely such a general meaning: the newly found condition of freedom from despotism, rather than some specialised and obscure constitutional concept, is what Herodotos aims to express. We must also remind ourselves that not every word fits a given metre. And, even more important, we must ask ourselves what form of government these little poems alluded to. Even though drunken men (and children) speak the truth[, to paraphrase a German proverb], they are not necessarily prophets as well. It is far less probable that these merry participants in the symposia at which they sang or extemporised these skolia were capable of experiencing highs of political vision and philosophical insight which enabled them to condense political conception of the later fifth century into a national anthem, than that which Vlastos regards as perverse: the simple expression of the contrast of tyranny and freedom thereof.
If my interpretation is judged too simplistic and a step backwards, one may contemplate the message of a scene on a red figure krater attributed to the Copenhagen Painter, of ca. 470 (see plate II).12 If viewed in this perspective it becomes also easier to understand how Aristophanes could have his fun with this theme.[12a] The Greeks evidently were able to bring the exalted back to almost banal human proportions just as
the tragedians on the other hand were able to present human every-day situations in superhuman dimensions. Our own psychology admittedly does not permit such, which renders it more difficult for us to interpret the Harmodios-skolia. Who would even think about representing martyrs of the Nazis peoples court [such as those of 20 July 1944] in the way the Copenhagen Painter drew the tyrannicides?
I)SONOMI/A, the noun which derives from I)SO/NOMOS, first occurs in the sixth century. It is, in any event, attested for Alkmaion of Kroton (24 B 4, Diels-Kranz 1,215,12). The context is non-political, and it signifies a medical ideal situation and the lack of I)SONOMI/A brings with it disease. We should, however, remind ourselves that, although the terms used (I)SONOMI/A and MONARXI/A) do not have the ring of a popularising paraphrase, we do not have the original words of Alkmaion.
We have already suggested that the noun I)SONOMI/A may under certain
circumstances even be applicable to the Spartan constitution (see above, 120). Thukydides usage corroborates this
suggestion; he does not use I)SO/NOMOS, I)SONOMI/A, AND I)SONOMEI=SQAI in
as narrowly defined a way as we might expect from Thukydides. We even
encounter an O)LIGARCHI/A I)SO/NOMOS: in the speech before the Spartans in
which the Thebans defend themselves against the accusations of the
Plataians they point out (3,61,2ff.) that the injustice against Plataiai
was perpetrated at a time when a DUNASTEI/A O)LI/GWN A)NDRW=N EI)=XE TA\
PRA/GMATA (3,62,3); this régime is compared to a tyrannis and
described with the words (3,62,3) NO/MOIS .... KAI\ TW|= SWFRONESTA/TW|
E)NANTIW/TATON. We are told (3,62,4) that H( CU/MPASA PO/LIS OU)K
AU)TOKRA/TWR OU)=SA E(AUTH=S because she was, as the Thebans explained
(3,62,3), OU)/TE KAT' O)LIGARXI/AN I)SO/NOMON POLITEU/OUSA OU)/TE KATA\
DHMOKRATI/AN; the same note is struck in 3,62,4
This contrast enhancing employment of I)SO/NOMOS and its derivations can
be found in a number of passages in Thukydides, e.g. 4,78,2f.
It may also be noted that in a philosophical fragment (Demokritos 68 B 251, Diels-Kranz 2,195-9-11) it is also the contrasting quality [of I)SONOMI/A] which prevails (without in itself implying a clearly defined positive alternative).
III. Herodotos 6,131,1
In that notorious chapter which concludes with Agariste dream and
[the birth of Perikles] Herodotos writes:
To begin with, DHMOKRATI/H cannot in this instance represent the mere contrast word as which we have encountered it elsewhere, since that would in effect mean Kleisthenes who established for the Athenians both the phylai and freedom from tyranny something that makes no sense chronologically13 and leaves the connexion of phylai14 and [freedom from tyranny] unexplained. Yet that democracy which Herodotos could observe as an adult in Athens was neither created nor even envisioned by Kleisthenes.15 The mythology of Kleisthenes, like the standard tradition of the achievements of Solon, Drakon and Theseus, belongs to the fourth century.16 In the passages
in which Herodotos traces the growing power and confidence of the Athenians there is no lack of hints at freedom from tyranny, but there is regarding DHMOKRATI/H, not to mention Kleisthenes and his phylai. Herodotos evidently shows little interest in his figure. This is all the more surprising in view of his frequently demonstrated art of composing portraits of various individuals, both Greek and barbarian. The most obvious explanation of this apparent contradiction is in my opinion that Herodotos neither knew nor could have known more about Kleisthenes than the little which he communicates.
The name of Kleisthenes had impressed itself on the collective memory of the Athenians primarily in connexion with the new administrative order of Attika based on the the new demoi which make up the ten new phylai. He was also remembered as one of the main players in the attempted usurpation by Isagoras with the aid of Kleomenes and his Spartiatai but mainly as a passive participant who on command disappears and reappears. He finally leaves the stage without a trace. Despite all ingenuity of modern scholars the Kleisthenes of the years after the expulsion of Hippias continues to escape our attempts to flesh him out, so to speak.
In this we receive no significant help from our ancient sources. The
Aristotelian Athenaion politeia devotes more words to Kleisthenes
than Herodotos but on closer inspection we discover that the account of the
Athenaion politeia has nothing of substance to add except that which
could be conjectured by observing the functioning democracy of the fourth
century and which was consequently added. The account in 21,2-6 is
summarised in 22,1 with these words:
makers had already established itself. In addition certain NO/MOI (22,1) of Kleisthenes are cited but only one (the law on ostrakismos) is identified.
We therefore depend on Herodotos, and more specifically the two passages
[in] 5,66-69 (67f., incidentally, is a digression [within the digression])
to which we are implicitly referred back in 6,131,1, i.e., [5,66,1 and
5,69,2]. After the departure of the Peisistratidai Kleisthenes and
Isagoras were the top men (E)DUNA/STEUON, 5,66,1) in Athens:
I have attempted to show elsewhere[16a] that the verb PROSETAIRI/ZESQAI has nothing to do with the [Athenian] HETAIREI/AI, but, rather, that it means something like to win over [to ones own side or cause]. The comparison with 3,70,2-3 shows how in these quasi-feudal circumstances the Persian aristocratic conspirators sought additional conspirators of an equal rank, no doubt in order to obtain the support of their followers, whilst at 5,66,2 and 5,69,2 Herodotos mentions only the followers but implicitly includes their aristocratic leaders, so that [in books three and five] we are presented with respectively one of the two sides of one and the same coin, as it were.
Here I present the second part of my interpretation of the meaning of all this. If Herodotos at 5,66,2 and 5,69,2 uses the word DH=MOS we do not need to jump to the conclusion that he meant it in a derogatory sense.
All difficulties of interpretation vanish once we take DH=MOS in its constitutional meaning, as a legal entity in which all constitutional authority resides, regardless of which term happens to be used, such as DH=MOS or DA=MOS, PLH=QOS or KOINO/N, etc. This legal entity manifests itself by its people assembly in session (whilst it is virtually non-existent if it is not in session).
Both the measures of Isagoras and, at least in some respect, those of the driving forces behind the reversal of his revolution point in the same direction: after the expulsion of the Peisistratidai the leading aristocrats went about their business as usual, and in this they had no use for the peoples assembly (not on a regular basis in any event), especially since its legislative functions had been lawfully delegated to the council of four hundred of Solon and because there was in addition the venerable council of the Areopagos. Kleisthenes and the aristocrats in his alliance transferred the political weight to the peoples assembly with such success that the only counter action left for Isagoras was to appeal to Kleomenes. He was thwarted by the aristocratic majority in the council of four hundred with their by now mobilised followers who rushed from the country-side to their aid and went on to humiliate the Spartan king and his Athenian protégée. It is only a logical consequence that the new council of five hundred would have their authority reined in and that the peoples assembly was strengthened.
All this, however, must not be equated with the Athenian [assembly democracy] which emerged in the latter decades of the fifth century. A glance at Herodotos constitutional debate in book three (see above, 119) makes this perfectly obvious.
We may conclude that all that Herodotos meant to say is that it was the ten phylai order (together with the deme order) and some kind of majority based government or administration by a transfer of political weight to the peoples assembly17 which was remembered by the Athenians in conjunction with the name of Kleisthenes. This represents a more natural interpretation of Herodotos 6,131,1 and it fits seamless into the political thinking of Herodotos as it expresses itself in a variety of passages; it does, however, confront us with the challenge to demonstrate that Kleisthenes and DHMOKRATI/A can indeed be chronologically connected, and that DHMOKRATI/A can indeed assume the meaning which I postulate for Herodotos 6,131,1 or,
in more general terms, that DHMOKRATI/A did not enter the Greek language with a meaning that would be the equivalent of rule by the people (cf. the German compound noun Volksherrschaft). This we shall attempt to show in the following section, IV.
About one half of the occurrences of DHMOKRATI/A in Thukydides are from
book eight, and of these two thirds occur bunched together in 28 chapters
(8,47,2, bis; 48,4; 48,5; 53,2; 63,3; 75,2; 89,3, bis; 90,1).
In these passage, in the context of the revolution of 411, there can be no
doubt regarding the only possible meaning: DHMOKRATI/A is part of the
anti-democratic vocabulary with a distinctly negative connotation and a
partitive meaning (corresponding to the way in which DH=MOS was used in
aristocratic circles), which comprises everybody but the
aristocrats. In strictly logical term DH=MOS cannot of course be partitive
but only all inclusive. Thukydides himself expresses this in the speech of
Athenagoras of Syrakousai (6,39,1):
The other half of the uses of DHMOKRATI/A is spread fairly evenly over the first six books (1,115,3; 2,37,1; 2,65,9; 3,37,1; 3,62,3; 4,76,2; 5,31,6; 6,39,1; 6,89,6). It is from these passages that we can excellently deduce the use of the word by Thukydides: he employs DHMOKRATI/A in much less restricted and rigid way than we might expect. Thukydides uses expressions such as DH=MOS, PLH=QOS, TO\ CU/MPAN, PO/LIS comparatively freely interchangeable, in order to mark the counter position to conditions or concepts such as DUNA/STEIA, bad oligarchy (cf. 3,62) etc. This may partly be explained by the fact that a significant number of our examples comes from speeches (2,37,1, epitaphios; 3,37,1, Kleon; 3,62,3, the Thebans; 6,39,1, Athenagoras; 6,89,6, bis, Alkibiades) but it is not a complete explanation.
We observe how Thukydides aims at delineating the contrast of
DHMOKRATI/A and A)RXH/. If, however, A)RXH/ (as one can hardly doubt) is
supposed to mean rule, and indeed any kind of rule, then by
implication DHMOKRATI/A is denied the meaning of rule. Nowhere is this
more given a more pointed expression than in the epitaphios18 (2,37,1):
In this usage DHMOKRATI/A conceptually approaches the meaning of the only other two -KRATIA compounds which are firmly attested for the fifth century, i.e., I)SOKRATI/A and A)RISTOKRATI/A. These two terms of course do not signify a form of government but, rather, a condition which characterises a certain form of government: a condition which is brought about by the application of the principle of equal-ness or best-ness20 The equation introduced by Thukydides/Perikles is solved cleanly: DHMO-KRAT/IA = E)S PLEI/ONAS OI)KEI=N, i.e., the condition and the characteristics of a form of government and administration (but not rule and power), not the form itself.
By using this interpretation it far easier to capture the meaning of
most of the passages outside book eight than by applying the restricted
meaning which is most appropriate in book eight. In particular for
instance if applied to the famous words of Thukydides according to which in
the lifetime of Perikles
Kleon is completely right when he complains that DHMOKRATI/A is
incapable of ruling (A)/RXEIN, 3,37,1)21 over others. And it marks the completion of the genesis of
the term if the detractors (in Athens and elsewhere) of DHMOKRATI/A feel
emboldened to raise as their objection that DHMOKRATI/A represents rule by
the DH=MOS (in its partitive-negative connotation). Now DHMOKRATI/A does
indeed rank as a form of government and rule, equal in its exercise of
power to other forms of rule, such as oligarchy or even monarchy. It is
only consistent that we read in book eight:
It is my hope that I have shown with these considerations that even for Thukydides the term DHMOKRATI/A had not yet reached the stage at which it had assumed a rigid and narrowly defined meaning. For its opponents the term meant of course the exercise of rule and power by the DH=MOS (part of which they did not consider themselves), and Thukydides uses DHMOKRATI/A in precisely this meaning wherever this is called for (throughout book eight). But because the meaning of DHMOKRATI/A had achieved its final definition only with the fulfilment of Athenian democracy (both in its positive and its negative aspects) and with the political rallying together of its opponents in the waning years of the fifth century, we can still encounter DHMOKRATI/A in Thukydides in varying meanings. We may take it for granted that Thukydides was conscious of the political as well as the semantic changes
which enabled him to use the term with due caution in such a way that he avoided anachronistic connotations.
Contrary to what might be reasonably expected there is (as far as I can see) only one examination of DHMOKRATI/A that was ever undertaken by a [specialist in] Indo-European linguistics, using the modern methods of comparative linguistics. A. Debrunner23 incontrovertibly (in my judgement) demonstrated that the compound noun DHMO-KRATI/A is an anomalous formation. It is not a natural growth but a Kunstwort, an artificial creation, without [a direct] linguistic parent. Further work must be undertaken on the basis of this discovery; an investigation which fails to take as its starting point the premiss that the word is an artefact will not produce any worth while results.
Let me quote Debrunner:[23a] Daß DHMOKRATI/A Volksherrschaft bedeutet, ist so selbstverstädlich, daß sich, wie es scheint, nie jemand Gedanken über die Art der Bildung dieses Wortes gemacht hat. Da Volksherrschaft tatsächlich gleichbedeutend mit DHMOKRATI/A ist, empfand man offenbar auch die formale Bildung der beiden Wörter als gleichartig. Aber diese formale Gleichsetzung ist, wie eine einfache Überlegung zeigt, falsch: Volks-herrschaft als Herrschaft des Volks entspricht einer ganz gewöhnlichen deutschen Zusammensetzungsart (Tages-arbeit, Gottes-haus, Feuers-glut usw.); aber, da es kein * KRATIA gibt, ist DHMO-KRATI/A nicht eine Zusammensetzung aus DH=MO(S) und KRATI/A, sondern es erinnert an den im Griechischen alltäglichen Typus von PARASU/NQETA: FILOLOGI/A Ableitung aus der Zusammensetzung FILO/-LOGOS. Volksherrschaft ist also eine sachlich richtige Übersetzung von DHMOKRATI/A, aber formal eine andere Bildung. Versuchen wir nun aber, DHMOKRATI/A nach dem Muster von FILOLOGI/A zu erklären, so stoßen wir auf ein unübersteigbares Hindernis: so geläufig auch seit dem 5. Jh. das Wort DHMOKRATI/A ist, so gibt es doch in der ganzen griechischen Literatur keinen einzigen Beleg für das zu fordernde Grundwort DHMO-KRATHS!
Debrunner even demonstrated that there would be no cause for surprise if the word meant rule over the people, given the way in which it is formed a point which further underlines the artificiality of the compound noun. We do not need to discuss Debrunners own attempt at explaining how DHMOKRATI/A came about, nor correct some imprecision in his argumentation
(it is not particularly consistent, for instance, to claim that Volksherrschaft is a factually accurate translation of DHMOKRATI/A, if there is no * KRATI/A, the meaning of which we would have to [(but obviously cannot)] determine first; furthermore Debrunner bases his explanation on the traditional idealised version of the creation of Athenian democracy by Kleisthenes which is now being called into question, if not tossed out altogether, by [at least some] historians).
The artificiality of the word is at the centre of the problem. If we wish to grasp it, we must ask ourselves without prejudice: by whom, when, for what purpose was the word invented? a question which would be irrelevant if the word had had a natural development. Anyone who chooses to avoid the question may be able to make sense of all passages containing DHMOKRATI/A from the fourth century BCE onwards and some (e.g., [Xenophon] Ath. pol.[; cf. above, 316ff.;] etc.) from the fifth century, but one will not be able to gain an understanding of the origin of the word in its historical context.
In the following pages we shall take it as a given that DHMOKRATI/A was invented in the political sphere, as a political catch word, with the purpose of describing and promoting some new political concept. It is also highly probable that no negative connotations were employed in coining this word.24 Most negative political catch words of our times were originally created in order to promote [and not to denounce (e.g., fascism, communism, etc.)]. It is on the other hand a well known phenomenon that an ideal may be turned into its opposite in the mouth of the political opponent [if not in the practice of its own proponent].
We must track down always bearing in mind the artificiality of the word DHMOKRATI/A the concrete political occasion with which the creation of the word can be reasonably connected. However, even if one clings to the conventional concept of a gradual evolution of Athenian democracy, founded by Kleisthenes, progressing towards a clear goal, enhanced by incremental reforms (a view I do not share, see above, 312), it is not easy to come up with an answer. To which particular step in the direction of the radical democracy could one attach this term? I cannot find any measure between the
destruction of the Areopagos council and the earliest25 attested occurrences of DHMOKRATI/A with which one might easily connect DHMOKRATI/A as a propaganda slogan. Even for Aristotle bad democracy only began to start on its course when Ephialtes (462) stripped the Areopagos of its powers, and Perikles introduced pay for jury service (pol. 1247a8). I cannot deny that the reforms of Ephialtes represents a plausible occasion if only because Ephialtes was murdered and we are therefore unable to tell if he had more far reaching goals26 (nor what these might have been). It would appear, however, that Ephialtes decrees which assigned the most important functions of the Areopagos to the peoples council and the peoples courts could not easily have been promoted as DHMOKRATI/A: these institutions belonging to the people had by now been active for almost two generations.
In order to answer these question we might best follow the example of Debrunner and scrutinise the problem from the etymological perspective. The reverse dictionary by Kretschmer and Locker lists seventeen compounds of the type -KRATI/A. Of these, three belong to the fifth century:27 A)RISTOKRATI/A, DHMOKRATI/A, and I)SOKRATI/A; another six occur first in philosophical or political writings of the fourth century: GUNAIKOKRATI/A, QEATROKRATI/A, PONHROKRATI/A and TIMOKRATI/A in Platon and Aristoteles, and I(PPOKRATI/A (as a military term) and PLOUTOKRATI/A in Xenophon. Polybios offers O)XLOKRATI/A28 and XEIROKRATI/A, the remainder
is even later.29
If conversely we group the -KRATIA compounds according to their meaning we find that there are two classes which are defined by the meaning of the second (the -KRATIA) component. In one class there are all those compounds by which a situation is described in which one group rules under exclusion of those who do not belong to this group. This meaning clearly derives from the anti-democratic application of the term democracy in which DH=MOS is not taken to include everybody; in these cases (about one third of the total) the first element of the compound implies a plural or a plural like meaning (the use of DHMOKRATI/A in a derogatory sense obviously also belongs here): GUNAIKOKRATI/A, DOULOKRATI/A, QEATROKRATI/A, O)XLOKRATI/A, PONHROKRATI/A. In the other compounds we observe a less clearly defined and more variable meaning of the -KRATIA component. At the same time the first element cannot be regarded as instrumental in the grammatical sense since it represents an abstract concept used in the function of an adverb or adjective which also cannot be derived from a noun or an adjective used as a noun: A)RISTOKRATI/A, QALASSOKRATI/A, I)SOKRATI/A, PLOUTOKRATI/A, TIMOKRATI/A, XEIROKRATI/A (another two words from the military vocabulary are of only indirect concern in this context: I(PPOKRATI/A and NAUKRATI/A).
It is significant that it is the second class in which we find the fifth century terms ARISTOKRATI/A and I)SOKRATI/A and also DHMOKRATI/A29a in the meaning suggested above (see esp., sections III and IV), as the condition of [living under] a form of government and administration based on the majority concept. DHMOKRATI/A thus represents a point at which the two classes of -KRATIA compounds intersect: it assumed a negative and partitive meaning in the mouths of its detractors and thereby became the model for further coinages of this type. It may be worth adding that ARISTOKRATI/A underwent the same kind of transformation from a term which signified a condition which ideally should attach to a particular from of government into one which itself represents
a specific form of rule (rule by the A)/RISTOI).30
In order to bring our enquiry to a conclusion we must with due emphasis pose the question: when was the term DHMOKRATI/A coined, and what did the creator of this term intend when he coined it? This seemingly basic question has to my best knowledge never been pointedly put. There is, however, no skating around this problem if we wish to comprehend the history of Athenian democracy and of its name.31
The what part of the question has been largely answered by our attempt to demonstrate that there existed an early meaning of DHMOKRATI/A as the condition of government by the majority. In reply to the when, i.e. at which point in the history of the Athenian state it would have been appropriate to characterise and promote a political programme by labelling it by means of a specially created word as the condition of majority government or administration, the answer which offers itself is obvious: at the time that such a government or administration was introduced in Attika and which as such constituted a contrast to a previous condition (the despotic reign of Hippias which did not function in accordance with the principle of majority rule), and most probably in the political environment of Kleisthenes.
How better could the Kleisthen-
ic reforms have been more aptly characterised than by a term which attests them that they brought about a condition of being governed and administered according to the principle of majority [government and administration]?
I personally think that it is possible to take this interpretation a step further by the hypothesis laid out in this Appendix. By and large the powers of the DH=MOS (as represented by E)KKLHSI/A and BOU/LH) were not significantly broadened by the Kleisthenic reforms of the last decade of the sixth century (except that the size of the council grew by a quarter) but were essentially left as Solon had conceived them. The Attic nobility retained the reins of power. Kleisthenes did as little to create democracy as Augustus did to restore the republic. Yet in the collective memory of the Athenians the name Kleisthenes is connected with the most significant reform of the late sixth century (if not of Attic history altogether), i.e. the introduction of the ten new phylai which represent the natural consequence of the deme reform.
This might find a surprisingly simple explanation by equating the first element of DHMO- KRATI/A not only with DH=MOS, the people, but also and chiefly (and in the beginning perhaps even exclusively) with DH=MOS, the deme. The original impetus for the creation of the new word DHMOKRATI/A may have been provided by the new condition of a government or administration by demes. There is ultimately no fundamental difference between DH=MOS (deme) and DH=MOS (people): the DH=MOS (people) is made up of the sum of the DH=MOI (demes). This kind of playing with terminology would not be uncharacteristic of Kleisthenic thinking: we recall that in the context of the Kleisthenic reforms the fresh term NO/MOS replaces the ancient QESMO/S (thus Ostwald, see above, n. 5). Our hypothesis would also explain why KW/MH disappears completely from the official vocabulary and is without exception replaced by DH=MOS, without any linguistic or dialectal (pace Aristot. poet. 1448a37) or any other reason. The implicit emphasis on
the rural element also fits perfectly our concept of the character of
the Kleisthenic reforms in which the rural aristocracy was assigned a
hitherto unknown significance in the structure of the state. A passage in
Isokrates (7,46) might underline this view if we could only determine its
The history of the word DHMOKRATI/A would then read approximately as follows. The term was created by the Kleisthenic reform movement with the purpose of characterising and promoting a new situation. This new situation should not, however, be understood to represent the fantasies of a visionary who like a prophet announces the democracy of the last decades of the fifth century. The development of that democracy was not what the reform act of the lat sixth century had aimed for, or could have envisaged. Rather, a word had to be coined which would serve as a label for the new condition of government by majority and administration by deme. Since DH=MOS (the entire people) and DH=MOI (the demes) are of the same substance, as it were, it comes as no surprise that there never existed a clear distinction between the two aspects in the minds of the Athenians. We equally need not be surprised that in parallel with the increasingly obvious emergence of the role of the peoples assembly the principle of majority in Athenian government completely obliterated the aspect of administration by demes which after a generation was [no longer a novelty but] simply taken for granted. It is thus logical that Aischylos in his Hiketides (see above, n. 31) could allude to DHMOKRATI/A (an allusion which only makes sense if he can expect his audience to recognise it), and that Herodotos could indeed write the words about the creation of DHMOKRATI/H [which we read in] 6,131,1 the word had been around long enough for everyone to be familiar with it; its connexion with Kleisthenes had never been completely forgotten. Nor had DHMOKRATI/A achieved its ultimate meaning to the exclusion [of all others nuances]: it shared its place in the vocabulary of poets, prose writers, of carousing noblemen and other citizens (whose state of mind was
less clearly recorded) with terms such as I)SONOMI/A, I)SOKRATI/A, I)SHGORI/A, and presumably also others which are not attested, for the purpose of describing a condition in which the principles of equality and justice, of the unchallenged rule of NO/MOS, prevailed aristocratic ideals32 by origin which could with natural ease be reinterpreted as democratic values, only again to be reinterpreted in a derogatory sense by certain aristocrats as characterising the dictatorship of the non-nobles (the non-A)/RISTOI, the non- KALOI\ KA)GAQOI/). DHMOKRATI/IA had now indeed come to denote a form of government. Because it was best suited as a utility word (and the least aristocratic one) it would emerge victorious just as its etymological father or, rather, adoptive father, the Athenian DH=MOS. But this happened in another age.
I now tend to prefer the view that DHMOKRATI/A was invented in the context of the Ephialtic revolution. Although I have not followed developments very closely (the date of the appearance of the volume cited above, n. 2 is only the result of delays caused primarily by the editor; it was originally accepted for publication in 1982), I have not found one single serious and well reasoned refutation of the hypothesis vented in the Appendix, only ex cathedra dismissal, and I still think it entirely possible that it in fact represents the correct explanation of the origin of the word. Possible, however, is not good enough in history as a scholarly discipline, as I am the first to demand. Unfortunately, we are dealing only with possibilities in this matter, and history as a scholarly discipline also requires reasoned arguments in order to dispose of a possible explanation such as the one offered in the Appendix.
I am also not aware of any thoroughly researched new investigation into the whole matter, and esp. such problems of detail as the ones touched on above, n. 7; the use of constitutional and similar terms in Thukydides (cf. above, 316ff.); terms formed similar to DHMOKRATI/A as pseudo-parasyntheta (cf. above, 319ff.); etc. If I have overlooked an article, I should be grateful to anyone who reminds me of its existence.]
1 Some of the problems dealt with here were first touched on in a paper read at the sixth annual meeting of ancient historians, May 1975, Columbus, Ohio. The present paper is a greatly expanded version of a paper read at the 39th annual meeting of the American Philological Association, December 1975, Washington, DC, which also formed the substance of papers or seminars at the universities of Berlin (FU), Freiburg, Jena, Leeds, Leipzig and Manchester, 1977.
[1a Originally published in German, *D*H*M*O*K*R*A*T*I*A: Studie zur Vor- und Frühgeschichte des Begriffs, Gymnasium 85 (1978), pp. 117-127; 312-326; plate I-II. I thank the publishers for permission to make this paper available on the Web. Square brackets normally indicate corrections, or alterations to the text beyond what might occur in fairly free translation. I have made no attempt here to bring the paper up to date although I have added a few references here and there.
Translation © K.H. Kinzl, 1999. E-mail .
2 I list here the most important contributions (with references to further specialised literature in the footnotes below)[. Some articles were republished in: K.H. Kinzl (ed.), Demokratia: der Weg zur Demokratie bei den Griechen (1995) (Wege der Forschung 657.); here abbreviated as Weg].
J.A.O. Larsen, Cleisthenes and the development of the theory of democracy at Athens, in: Essays in political theory presented to George H. Sabine (1948) 1-16. H. Schaefer, Besonderheit und Begriff der attischen Demokratie im V. Jahrhundert, in: Synopsis, Festgabe für A. Weber (1948) 479-503 (= in: H. Schaefer, Probleme der Alten Geschichte: Gesammelte Abhandlungen und Vorträge (1963) 136-152). V. Ehrenberg, Origins of democracy, Historia 1 (1950) 515-548 (= in: V. Ehrenberg, Polis und Imperium: Beiträge zur Alten Geschichte (1965) 264-297[; = in: Weg 70-86 (includes only introduction and section IV.)]). C. Meier, Drei Bemerkungen zur Vor- und Frühgeschichte des Begriffs Demokratie, in: Discordia Concors, Festgabe für E. Bonjour (1968) 3-29[; = in: Weg 125-159]; id., Die Entstehung des Begriffs Demokratie, Polit. Vierteljahresschrift 10 (1969) 535-575; id., Die Entstehung des Begriffs Demokratie: Vier Prolegomena zu einer historischen Theorie (1970) 7-69[; id., Enstehung und Besonderheit der giechischen Demokratie, Zeitschrift für Politik NF 25 (1978)1-31 (= in: Weg 248-301)]. The article by R. Sealey, The origins of Demokratia, CSCA 6 (1973) 253-295 came to my attention only after completion of the present paper by the gift of an offprint by its author (whom I thank for it); also the collection of partially relevant articles in Klio 57 (1975) fasc. 2 (esp. that by H.-D. Zimmermann, 293-299). [K.A. Raaflaub, Einleitung und Bilanz: Kleisthenes, Ephialtes und die Begründung der Demokratie, in: Weg 1-54.]
3 K.H. Kinzl, Athens: between tyranny and democracy, in: Greece and the eastern Mediterranean in ancient history and prehistory, studies presented to Fritz Schachermeyr (1977) 199-223[; on the internet: http://www.trentu.ca/academic/ahc/schach.htm; translated into German: Athen: zwischen Tyrannis und Demokratie, in: Weg 213-247.]
4 I attempted to draw attention to this narrative device in RhM 118 (1975) 193ff.
5 On this term see esp. G. Vlastos, Isonomia, AJPh 74 (1953) 337-366; id., *I*S*O*N*O*M*I*A *P*O*L*I*T*I*K*H, in: Isonomia: Studien zur Gleichheitsvorstellung im griechischen Denken (1964; repr. Without changes 1974) 1-35 (= G. Vlastos, Platonic studies (1973) 164-203); M. Ostwald, Nomos and the beginnings of the Athenian democracy (1969); cf. B. Borecký, Die politische Isonomie, Eirene 9 (1971) 5- 24; also the articles listed above, n. 2.
6 H. Apffel, Die Verfassungsdebatte bei Herodot (3,80- 82), Diss. Erlangen (1957). On the sources cf., e.g., K. von Fritz, Die griechische Geschichtsschreibung vol. 1, Text (1967) 310ff.; Ostwald 107f.; W.R. Connor, The new politicians of fifth-century Athens (1971) 199ff.; D. Fehling, (Die Quellenangaben bei Herodot (1971) 92 [= Herodotus and his sources: citation, invention and narrative art, trans. J.G. Howie (1989) 122]; Sealey 272f.
Herodotos did not know [the term] DHMOKRATI/A: G.T. Griffith, Isegoria in the assembly at Athens, in: Ancient society and institutions: studies presented to Victor Ehrenberg (1966) 132 n. 4.
The qualitative difference between I)SONOMI/A (more of a banner than a label) and DHMOKRATI/A (a utility word) is emphasised by Vlastos 8 (= 172f.); cf., however, A.W. Gomme, HCT at Thukydides 3,82,8. Gomme already emphasised that the term denotes not a form but a characteristic of this particular kind of rule; cf. Ostwald 97.
7 Cf. the words of Theseus in Eur. Hik. 399ff. In the dialogue with the keryx from Argos, as well as 238ff. and 350ff. A. Lesky, Geschichte der griechischen Literatur (31971) 428-430; G. Zuntz, The political plays of Euripides (1955) 3-25. Cf. also the dissertations by H. Bengl, Staatstheoretische Probleme im Rahmen der attischen, vornehmlich euripideischen Tragödie (1929) and S. Michaelis, Das Ideal der attischen Demokratie in den Hiketiden des Euripides und im Epitaphios des Thukydides (Marburg 1951, typescript). This play remains largely unexplored, both from the perspective of the historian and that of the constitutional historian, both relative to Herodotos and especially to Platon (cf. H. Erbse, Gymnasium 83 (1976) 169ff.) and Aristoteles.
8 The compelling logic with which Thukydides turns this strongest of arguments into its opposite (in the speech of Athenagoras of Syrakousai, 6,39,1) makes it more than likely that Thukydides alludes specifically to the speech of Megabyxos.
9 M. Ostwald, Isokratia as a political concept (Herodotus 5,92A,1), in: Islamic philosophy and the classical tradition: essays presented to Richard Walzer (1973) 277-291.
10 Herodotos does not employ any adjectives or adverbs which derive from the nouns discussed in this section.
11 The skolia (nos 10 and 13 Diehl) at Athen. 15,695. V. Ehrenberg, Das Harmodioslied, WS 69 (1956) 57-69 (= Polis und Imperium (1965) 253-264); other references collected by J.M. Edmonds, Lyrica Graeca 3, 554f. [now superseded by D. Campbell, Greek lyric, vol. 5: The new school of poetry and anonymous songs and hymns (1993) 284ff.].
12 On this topic see S. Brunnsåker, The tyrant- slayers of Kritios and Nesiotes: a critical study of the sources and restorations (21971).
The krater published for the first time (as far as I know) in this article [see below, plates I and II] makes no appearance in either Brunnsåkers book or E. Vermeule, Five vases from the grave precinct of Dexileos, JDAI 85 (1970) 103ff. (no 3: a non-parodising red figure oinochoe, [dated to] shortly after 400), and it is also not catalogued by J.D. Beazley, ARV2 and Paralipomena.
13 Unless we resort to assuming use of the stylistic device of the hysteron - proteron, with the consequence of attesting Herodotos a remarkable degree of incompetence as a writer by suggesting that this is what he meant but did not succeed in expressing.
14 As if Athens did not have phylai before Kleisthenes which were well known to Herodotos: 5,66,2; 5,69.
16 Ath. pol. 29,3 is a comment by the author and was not lifted from the original wording of the motion of Pythodoros of 409. In Isokrates 16,29 Kleisthenes has to share the fame of the expulsion of the tyrant and the founding of democracy with the elder Alkibiades ([PAA 121620]). It is only in speeches of the 350s, i.e., Isokr. 7,16 (regarding the uncertainty of the date cf., e.g., Lesky [above, n. 7] 658) and Isokr. 15,232 (*KLEISQE/NHS .... TH\N DHMOKRATI/AN .... KATE/STHSE) (year 353) that Kleisthenes is identified as the founder of democracy but by now he has already a predecessor in Solon to whom the same achievement is ascribed (e.g., Aristot. pol. 1273b38; cf. 1274a2). Cf. E. Ruschenbusch, *P*A*T*R*I*O*S *P*O*L*I*T*E*I*A, Historia 7 (1958) 398-424 [= in: Weg 87-124] (first mention as constitutional reformer: Solon 356; Theseus 343; Drakon 328/22). There is a gap of about a century between Hdt. 6,131,1 and the passages from Isokrates of an incontestable meaning.
17 Hdt. uses PLH=QOS, DH=MOS, KOINO/N. The terminus technicus E)KKLHSI/H is not, however, used in an Athenian context but for Samos, 3,142,2.
18 Monograph length treatment (with the extensive bibliography on it): J.T, Kakridis, Der thukydideische Epitaphios: ein stilistischer Kommentar (1961); H. Flashar, Der Epitaphios des Perikles (1969) (SHAW); K. Gaiser, Das Staatsmodell des Thukydides: zur Rede des Perikles für die Gefallenen (1975); cf. Michaelis (above, n. 7). A useful bibliography was compiled by W.C. West III, in: Speeches in Thucydides: a collection of original studies with a bibliography (1973) (esp. nos 209-262).
19 Cf. Thuk. 2,15,1:
20 Cf. Gomme, above n. 6; the rather conventional exposé of H. Vretska, Perikles und die Herrschaft des Würdigsten, RhM 109 (1966) 108ff. fails to add anything of substance. Cf., however, H. Herter, Comprensione ed azione politica: a proposito del capitolo 40 dell epitaphio tucidideo, in: Studi in onore di G. Funaioli (1955) 133-140 (= Kleine Schriften (1975) 223-229 [in German]). I thank Professor Herter for a number of interesting objections to my views expressed in this article.
21 Thuk. 3,37,1:
23 A. Debrunner, *DHMOKRATI/A, in: Festschrift für Ed. Tièche (1947) 11-24 [(= in: Weg 55-69) at 11 (= 55)].
[23a Translation: [[it must be remembered that Debrunner argues as a German speaker using the German compound noun Volksherrschaft (approximately peoplerule, if the English language permitted such a formation); a French or English speaker would have introduced the reader to the linguistic problem in a different way with no effect, however, on the main thrust of the argument regarding the artificiality of the word.]] It seems so obvious that DHMOKRATI/A means rule by the people that, as it seems, no one has given thought to the way in which this word was formed. Since Volksherrschaft does indeed mean the same as DHMOKRATI/A it was apparently felt that the way in which the word was formed is identical too. But it is wrong to equate their respective formations, as can be shown by considering a few simple facts and examples: Volks-herrschaft as Herrschaft des Volkes [[both translate into standard English as rule by the people]] corresponds to a perfectly plain and common German way of forming compounds ([[D. cites German words meaning a days work; Gods house, i.e. church; the fires glow]]) But since there is no word * KRATI/A, DHMOKRATI/A is not a compound made up of DH=MO(S) and KRATI/A; rather, it reminds one of a type perfectly common in Greek called PARASU/NQETON: FILOLOGI/A is derived from FILO/-LOGOS. Volksherrschaft may be a factually correct translation of DHMOKRATI/A, but it represents formally a different type. If we were to attempt to explain how DHMOKRATI/A was formed by using the example of FILOLOGI/A we encounter an insurmountable obstacle: however common the word DHMOKRATI/A may have been from the fifth century onwards, there is not a single attested occurrence of the required precursor DHMOKRATHS!
24 W. Donlan, Changes and shifts in the meaning of demos in the literature of the archaic period, PP 25 (1970) 381-395, delineates especially the negative, partitive uses of DH=MOS, but I do not see how this would affect my argument.
25 We cannot here deal with the next to unresolvable and almost innumerable questions which are connected with dating certain passages in Herodotos or Thukydides or the pamphlet of the Old Oligarch ([Xenophon] Ath. pol.), nor would it influence our argument. It may be noted, however, that the treatment of the Kolophon decree (IG I2 14/15, 49) by R. Meiggs and D.M. Lewis, Greek historical inscriptions (1969) no 47 (pp. 121-125) is conclusive; the restoration of the text in SEG 10,17 (ATL 2, p. 68; H. Bengtson, Staatsverträge 21 no 145, p. 48) is without parallel [although sadly resurrected by Meiggs and McGregor in IG I3 37 a-c; the point may be moot if H.B. Mattingly, Historia 10 (1961) 175 (= The Athenian empire restored (1996) 37) is right in suggesting a date ca 427 instead of 447, the date of the enforcers of orthodox creed]. The literary sources already mentioned therefore remain the earliest testimonia for the word DHMOKRATI/A.
26 Cf. C. Hignett, A history of the Athenian constitution to the end of the fifth century (1952) 213.
27 Earliest occurrence according to the citations in LSJ9.
28 One might take note of the words of the messenger from Argos addressed to Theseus in Eur. Hik. 410f.: PO/LIS GA\R H(=S E)GW\ PA/REIM) A)/PO | .... OU)K O)/XLW| KRATU/NETAI. Cf. below, n. 31 (Aisch. Hik.).
29 DOULOKRATI/A (Iosephos); QALASSOKRATI/A (Strabon); NAUKRATI/A (Kassios Dion); QEOKRATI/A (Iosephos); A)KRATI/A and E)GKRATI/A are for without interest.
[29a For what it is worth I should point out that in all three cases (not only in the case of DHMOKRATI/A) there is no adjective in -KRATHS as the required ancestor attested there are, however, in all three instances proper names known which would parallel the non-existent adjectives: Aristokrates, Isokrates, Demokrates.
30 Or, more accurately, the KALOI\ KA)GAQOI/: W. Donlan, A note on aristos as a class term, Philologus 113 (1969) 268-270.
31 Excursions to other poleis, earlier centuries, or other authors, do not produce results of any significance. The Great Rhetra admittedly uses KRA/TOS [and possibly DA=MOS] or a compound of DA=MOS: the corrupted text of Plut. Lyk. 6,3 has GAMWDANGORIANHMHN KAI\ KRA/TOS); Theognis uses both terms in consecutive lines (but without melding them into a single concept), cf. 39ff.; esp. 45f.
Dealing with Aischylos, and esp. his Hiketides, is more productive. Ehrenberg (above, n. 2) 517ff. (= 266ff.) has made it highly probable that there are allusions to DHMOKRATI/A here. However, the play has since been redated to 463 (e.g., A. Lesky, Die tragische Dichtung der Hellenen (31972) 98). Cf. on the other hand Eur. Hik. 411 (above, n. 28). An observation of Sealeys (above, n. 2, 270) is also interesting: when two lines of The Suppliant Women associate words with the roots dem- [sic! Correct: demo-] and krat-, such language does not mark anything incompatible with a constitution which was called oligarchia. Cf. above, section III (312ff.), and the interpretation of I)SOKRATI/A by Ostwald, above, n. 9.