Dionysius of Halicarnassus On the Style of Demosthenes 10: 'but the orator aims at the sufficient and gauges his opportunities accordingly, adapting his style not to a memorial and possession only, as the historian does, but also to advantage. So that he has never exceeded the bounds of clearness, the first requirement in competitive forensic speeches, and he has obtained the goal of seeming to be powerfully eloquent, for which above all he is zealous'. [1]
No single passage in ancient historiography has been more discussed by modern scholars than Thucydides' exposition of his historical method in 1.22; no single statement in ancient historiography is more famous than that exposition's culminating claim: 'it is set down as a possession for always rather than as a competition piece for present hearing' (1.22.4).[2]
This paper has six, interlocking, aims:
These aims require some comment and contextualisation within existing scholarship.
As for aim (1a), the case that Hecataeus, Herodotus and Thucydides (and indeed other early Greek authors, historiographical and other) represent their works as monumental inscriptions has been variously propounded and goes back to the very beginnings of modern study of Greek historiography (bibliographical references in note [32]). Nevertheless, not only has this case failed to invoke Thucydides' use of the Athenian inscriptional formula (which is actually the single strongest piece of evidence), but it has made little impact upon mainstream Thucydidean scholarship; its fullest expositions are very recent; and the arguments advanced in its support, diverse as they are, require careful assessment; furthermore, there has been even less consideration of the consequences for the interpretation of Thucydides, particularly as regards his relationship with his predecessors; finally, the first historians' use of the image of the monumental inscription must have implications for three important questions about ancient historiography generally: first, the question of origins, self-definition and genre; second, the question of the historians' attitude to inscriptions as a form of evidence; third, the question whether they are projecting their works primarily as reading texts.
Though aim 3 reflects my general conviction that ancient writers see far more in one another's texts than do modern critics, it had its immediate origin in the discussion at the end of the conference, held in the University of Leeds on 16 April, 1999, on the theme of narrative issues in classical historiography. One of the questions raised by the Chairman, Simon Hornblower, was why ancient literary critics seem to say so little that is genuinely useful even about passages whose greatness they rightly proclaim (in this case Thucydides' account of the sea battle at Syracuse). To which (with characteristic subtlety) your editor replied that, seeing that ancient literary critics were 'no good' in general (pained expressions appeared on the faces of some of the participants), it was not at all surprising that they should be 'no good' on ancient historiography, and that just as the best ancient critics of Homer, Apollonius and Virgil, were their poet successors, so the best ancient criticism of historiography was to be found in ancient historians themselves. Both parties to the debate were using the term 'criticism' in its neutral sense. For although most of the historiographical 'criticism' in the works of the ancient historians takes the form of negative criticism and this negative criticism is often, indeed characteristically, captious and unfair, the criticism is not always or wholly negative: it may bear on issues of real substance or involve real insight into the procedures and thinking of a predecessor, and this even when its main purpose is to demonstrate the superiority of the historian writing. Even more revelatory, however, is the implicit criticism involved in ancient historians' implicit allusions to, or intertextual debate with, their predecessors.
To such ancient readings modern criticism of ancient historiography needs to be particularly responsive. For, while it is true that modern criticism of classical literature in general is greatly concerned with the revaluation or recuperation of ancient authors, to such an extent that it is nowadays extremely rare to find an article arguing that such-and-such is a bad poem or author, this humane and generally productive (if sometimes highly implausible) principle has not been absorbed to nearly the same degree in the criticism of ancient historiography. The main reason for this is that many of the modern critics of ancient historiography are primarily historians and fail to recognise the need for proper literary criticism of historiographical texts, or at least, while formally recognising that need, fail properly to implement it in their practical criticism. If the present paper can show (inter alia) that Polybius, Livy and Arrian, and, still more alarmingly, Dionysius of Halicarnassus and Thucydides' biographer, Marcellinus (n. [17]), were superior readers of Thucydides 1.22.4 to (seemingly) all modern scholars concerned with Thucydides, this ought to constitute a salutary and humbling lesson. It is true that existing scholarship has shown some awareness of the interpretative value of ancient authors' responses to 1.22.4 (see e.g. section 4 below and n. [72]), but I shall try to show that the three great historians mentioned display far greater acumen than has yet been recognised.
Of course, such an appeal to the 'authority' of ancient historians' readings of other ancient historians is but another aspect of the single greatest advance in the interpretation of ancient historiography made in recent years: the insistence (à la school of Wiseman, Woodman and Marincola) on reading ancient historiography in the first instance by ancient criteria, not by inappropriate (and frequently unsubtle) modern ones.
Although the six aims of this paper are necessarily interwoven, the difficult problems that arise necessitate a good deal of weaving back and forwards and stopping and starting, but I hope that the main threads of the argument can readily be followed.
The paper is organised into the following sections (with some interlinking passages):
In items 11-13, 'superior' means 'superior to any offered by modern scholarship'.
[1.22.1] As for all the things that each side said in speech [logos], either when they were going to war or when they were already in it, it was difficult both for me in the case of the things that I heard myself and for those who reported to me from various different places to remember completely the accurate content of the things that were said. But as it seemed to me, keeping as closely as possible to the general drift of what truly was said, that each speaker would most say what was necessary concerning the always presents, so I have rendered the speeches. [22.2] But as regards the deeds of the things that were done in the war, I did not think it a worthy procedure to write by asking for information from anyone who chanced to be present, nor just in accordance with what seemed to me to be so, but both in the case of things at which I myself was present and of things which I learnt about from others, by going through them in each case with accuracy as far as possible. [22.3] They were discovered with much labour, because those who were present at each particular deed did not say the same things about the same things, but in accordance with the individual's sympathy for one side or the other or his memory. [22.4] And perhaps the lack of the muthos element in my history will appear rather unpleasing to an audience, but if those who wish to look at [hosoi de boulesontai skopein] the clearness [to saphes] both of the things that happened and of those which, in accordance with the human thing, are going to happen again some time like this and near the present ones, should judge it helpful, that will be sufficient. It is set down as a possession for always rather than as a competition piece for present hearing.The extremely literal translation aspires (in the interests of maximum comprehensibility but at the cost of considerable inelegance) to convey as many as possible of the implications and of the verbal interrelationships of the original. [4] A wholly English exposition, however, is impossible, owing to the extreme verbal density of Thucydides' Greek, and the analysis must sometimes have recourse to transliterated words. Naturally, readers with Greek and Latin should read this paper with the originals before them.
(b) The claim that Thucydides' wording 'it is set down as a possession' (ktema ... xugkeitai) recalls a Homeric phrase, specifically one used of ancestral wealth, domois en ktemata keitai = 'possessions are laid down in the house' (Iliad 9.382; Od. 4.127), is much less self-evidently true than (a) and requires fairly protracted argument and demonstration.[7]
The next question for the moment: how can Thucydides claim that his work is a 'possession set down for always'? Some scholars seem to think that Thucydides means merely: 'my treatment of the Peloponnesian War will be the standard treatment for a very long time'. [10] But this anyway banal reading neglects the implicit temporal logic of the context. Thucydides' work is a 'possession set down for always' because it transcends its own time frame, providing coverage both of 'the things that happened' (the events of the Peloponnesian War) and of 'those which, in accordance with the human thing, are going to happen again some time like this and near the present ones'. [11] Is the immortality of Thucydides' work also promoted by its being a written text to be received primarily by readers? The matter of reception requires separate treatment (section 7) and, moreover, Thucydides' relationship with Hecataeus and Herodotus (sections 8 and 9) must also be taken into account before a final answer to this question can be given.
There is another, slightly more oblique, implication. Thucydides describes his work as a 'possession set down for always rather than a competition piece for present hearing'. Now, although his primary concern is with his work as received by his audience/readers, i.e. with what it will be to them, he is obviously also claiming immortality for it on his own behalf. [12] And since the Greek word ktema (here translated 'possession') can also be used of 'prizes' won in war or in any agonistic context, [13] there is another terminological and conceptual transference in play: [14] the 'prize' that Thucydides wants is not the 'prize' that might accrue from the success of a 'competition piece' but the 'prize' of everlasting usefulness. In this way Thucydides absorbs and transcends the oral/aural agonistic culture which he formally consigns to second place. This association with 'prizes' and 'agonistic' contexts naturally facilitates the Homeric allusion for which I shall argue positively in section 5.
Gentili and Cerri, comparing the Latin use of monumentum, see both terms as referring to the 'materialità della parola scritta' ('the materiality of the written word').[15] Although this interpretation points, I believe, in the (or a) right direction, it is, as stated, insufficiently grounded and excessively abstract.
More interestingly, Svenbro claims, that when Thucydides describes his work as a ktema ... es aiei, 'the idea of a mnema "(funerary) monument" is not far off'.[16] The bases of this claim are, firstly, the prior claim (resting on Thuc. 1.1.1 and its Hecataean and Herodotean parallels) that Thucydides presents his History as a monumental inscription; secondly, Svenbro's belief that there is (always?) a strong felt association between ktema and mnema. To the validity of the first claim we shall return (see section 8, which includes discussion of Svenbro's arguments on this point). The second claim seems to me speculative and certainly cannot be resolved here, though it might seem a point in its favour that Thucydides is undoubtedly indulging in some rhyming verbal play in this sentence (ktema ... xugkeitai, ktema ... (para)chrema [n. [9]). On the other hand, the specific implication 'funerary monument' would seem to make Thucydides' project inertly commemorative, rather than timelessly relevant. Finally, the parallel with the Athenian inscriptional formula already suggests that Thucydides here has a different sort of 'inscription' in mind.
Without himself endorsing the claimed association between ktema and 'monument', Nicolai has adduced some passages from ancient authors which make this move. The two most striking passages are as follows:
Dionyius of Halicarnassus, On the Style of Demosthenes 10; this clever passage is quoted as the superscription to this paper. Dionysius, clearly alluding to Thucydides 1.22.4, equally clearly understands ktema as involving anathema, that is, in context, a literary 'memorial'.
Marcellinus, Life of Thucydides 35; after stating that Thucydides wrote for the very clever/wise, Marcellinus generalises: 'for the man who is praised by the best men and has won an adjudged reputation wins honour inscribed for future time, not honour that risks being obliterated by subsequent judges'. This generalisation clearly derives from an inscriptional reading of 1.22.4. [17]
Nicolai also contextualises such an understanding of Thucydides' 'possession' within the general application of imagery based on statues, inscriptions and monuments to be found in ancient poetry (including Ennius and Horace) and in ancient discussions of historiography, biography and poetry.
Nicolai's citation of these ancient interpretations is valuable (although, as we shall see, there are even more useful ancient examples [sections 12 and 13]). But it is of course open to us not to accept this ancient interpretation, and, given the inadequacy of the modern arguments of Gentili and Cerri and Svenbro, the case for the interpretation cannot properly be assessed outside the whole question of the prooemial form of Thucydides' work (and Hecataeus' and Herodotus'), which will be treated later, in section 8. For the moment, therefore, I revert to the Homeric quality of Thucydides' 'possession'.
At this juncture it becomes necessary to consider the force of the xug- prefix in the verb xugkeitai, the last word of ch. 22. A minimalist interpretation of xugkeitai would appeal to the fact that xugkeimai is a standard word for literary composition [20] and naturally this is part of the meaning at 22.4, but Thucydides' deployment of the word at this particular moment in his text is also strategic, helping to articulate the architecture of the whole argument of the extended preface.
Thucydides begins his History with the words: 'Thucydides the Athenian wrote up [xunegrapse] the war of the Peloponnesians and Athenians' (1.1.1). He then claims that this war was the greatest kinesis of all (1.1.2). He then attempts to prove this claim, firstly by a negative ('earlier periods were not great': 1.1.3-21.1), secondly by a positive ('this war was the greatest' [1.21.2; 23.1-3]). The first, negative, part of the proof ends with a justification of Thucydides' historical method regarding the reconstruction of the periods before the Peloponnesian War (20.1-21.1). The second, positive, part of the proof (1.21.2), begins with a justification of Thucydides' historical method regarding the writing up of the war (1.22.1-3); both these methodological sections provide the necessary underpinnings for his factual claims.
Although it is a formal error to invoke the contents of chs. 20 and 21 as direct evidence for Thucydides' reconstruction of the war itself, [21] it is nevertheless true and important that the methodology of ch. 22 is given increased definition by a series of parallels and contrasts between chs. 20-1 and 22. Then 23.1-3 makes the positive case; then Thucydides goes back in time to 'pre-write' (prographein) the causes of the war (23.5 'I pre-wrote' picks up, and contrasts with, 1.1.1 'Thucydides wrote up the war').
Thus one point of 22.4 xugkeitai is to respond to 1.1.1 xunegrapse: xunegrapse makes the initial statement, xugkeitai describes the end of the process, incorporating the methodological principles of 22.1-3 (Thucydides' handling of the speeches and the deeds of the war). But since these methodological principles involve implicit comparison and contrast with Thucydides' various rivals in the sort of exercise he is engaged in, xugkeitai also includes this process of comparison and contrast. Moreover, when contrasting his reconstruction of events before the Peloponnesian War with those of rivals in that field, Thucydides stated (1.21.1): 'But anyone who considered from the aforesaid indications that things were more or less what I have described would not go wrong, neither trusting what the poets have eulogised about them, nor what the prose writers have put together (xunethesan) for the purpose of enticement to the audience rather than the truth, things that cannot be checked and the majority of them having prevailed owing to time untrustworthily to the point of myth, but believing that they have been discovered from the clearest signs sufficiently in so far as they are ancient things'. Hence, because of the implicit parallels and contrasts between chs. 20-1 and 22, xugkeitai 'trumps' xunethesan, of the logographers' inferior efforts: Thucydides is involved in a superior exercise of 'putting things together'. In sum, Thucydides represents his History as the work in his field, which 'synthesises' (xunethesan) and 'compounds' (xugkeitai) all other relevant works.
Given also the already discussed transference of 'possession'/'prize' terminology to Thucydides' History (end of section 3), a literary allusion to the Homeric 'ancestral wealth' phrase is thoroughly apposite, clinching the case for the overwhelming superiority of Thucydides' work. Within the whole tradition of literature concerned with war and politics, Homer must be regarded as Thucydides' greatest single rival - but Thucydides comprehensively surpasses him. In Homer the expression 'possessions laid down in the house' describes the literal ancestral wealth of a favoured few (the accumulation of all the individual 'possessions' or 'prizes' won by members of those families). In Thucydides, by contrast, the 'possession' is Thucydides' work itself; it has greater value than any mere physical possession or prize; its value, while like the ancestral possessions described by Homer in passing down the generations, is immeasurably greater in that it will last 'for always'; and it can be acquired by/it is (as it were) 'free' to 'anybody who wishes to look at it'; finally, the shift from the Homeric plural ktemata to the singular ktema neatly reinforces the idea that Thucydides' is the single work, which renders all others (relevant to the field) redundant.
Of course, our hypothetical minimalising critic might again respond to all this by saying that the hyperbaton involved in reading ktema ... xugkeitai as a gloss on the Homeric ktemata keitai is too great to render the reading plausible. Yet not only does the reading make a thoroughly fitting culmination to Thucydides' dense literary allusions and historiographical claims in this passage, but the hyperbaton itself can be given a double value. In the first place, the hyperbaton does indeed pose an interpretative problem: one has to see it (and, as we have seen, it has seemingly been seen by no modern critic hitherto); by its very nature hyperbaton involves perceiving and then filling in a gap, and it can be argued that such filling of gaps is one of the great interpretative challenges that Thucydides imposes upon us.[22] In the second place, the enclosing word order actually enacts the all-embracing quality of Thucydides' 'possession' and so itself contributes to the meaning.
Interesting, though not initially straightforward, further confirmation of the Homeric allusion at 1.22.4 is provided by the language of the (alleged) letter of the Persian king Xerxes to the Spartan Pausanias, of which Thucydides gives an allegedly verbatim version (1.129.3):[23]
'King Xerxes speaks as follows (hode legei): "for the men whom you saved for me beyond the sea from Byzantium your good service will be laid up (keisetai) for you in our house for ever inscribed (anagraptos) ...'On this letter Crane [24] has written: 'the inscribed monument can formally place a written utterance in the public domain, citing its authors and presenting its contents for all to see. The solemnity and authority of such monumental writing for fifth-century Hellas is clear in Thucydides. He quotes a letter in which Xerxes acknowledges his indebtedness to the Spartan Pausanias: "for the men whom you have saved for me across the sea from Byzantium, your good service (euergesia) will be deposited (keisetai) for you in our house, written down (anagraptos) forever" ... A "good service," euergesia, is typical of the language of traditional ritualized friendship and points backwards to the bonds which link Glaukos with Diomedes in Iliad 6, and Mentes with Odyssey in Odyssey 1. The verb keisetai belongs to the language of wealth: heirlooms or precious metals "are stored away." The adjective anagraptos, "written down," points, however, towards the recent (for the Greeks, anyway) language of financial accounts. The Great King not only acknowledges a debt, but emphasizes the value which he places on it through the metaphor of writing. Pausanias' euergesia is converted metaphorically to a monetary sum, but this conversion enhances rather than reduces its value. Written down in the accounts of the Great King, Pausanias' euergesia will never be forgotten'.
This analysis is penetrating and perceptive, but, from my point of view, it fails to make the requisite connection with 1.22.4, and this despite the use of significantly similar wording ('for all to see'). Also, the analysis rather overplays the Greekness of the concepts and downplays the Persian ones: Persians could also acknowledge 'good service', and 'our house' (vel sim.) is a stock phrase of the Persian king.[25] As regards both these matters and Crane's comment on keisetai, account must be taken of the much-discussed Letter of Darius to the satrap Gadatas (preserved in a Greek inscription of the second century A.D.), lines 15-17 of which read: 'because of these things there will be laid up (keisetai) for you great gratitude in the King's house'. But this letter does not make against a Homeric reading of keimai. On the contrary, the natural inference is that translations of Persian kings' letters into Greek readily resorted to the Homeric notion of 'goods laid up', the more so because of the convenient switch from the Greek conception of the family 'house' to the Persian King's and the piquant transference of context thus entailed: instead of the beneficiary's having riches laid up in his own house, they are laid up in the Great King's. In sum, Xerxes' letter of Thucydides 1.129.3 supports a Homeric interpretation of ktema ... xugkeitai, especially given the further parallels es aiei and (as we shall see) anagraptos. It is indeed difficult not to feel that Thucydides' mind is working with the same sorts of concepts in 1.22.4 as in 1.129.3. [26]
We may now go back to that other aspect of the notion of 'possession' anticipated at the end of the first paragraph in section 3. It is simply this: among all the competing claims of different authors and different types of work and all the sorts of 'possession' an individual may reasonably hope to have, Thucydides' 'possession' is the one thing you've got to have.
But of course, in the same way as Thucydides' work transcends its more general physical equivalent, the 'possession', so it enormously transcends any conceivable physical inscription. Thus Thucydides is arguing for the superiority of engagement with a great written text to oral/aural media and this argument is itself an important element of his overall argument for the superiority of his work to all other possible rivals. I shall return to this aspect of 1.22.4 below in section 10 and there try to import some further refinements in the light of the question of Thucydides' relationship with Hecataeus and Herodotus.
Now the above interpretation may seem to run into two, related, difficulties.
First, if any of Thucydides' predecessors and rivals likened their works to inscriptions, where does this leave Thucydides' claim to superiority on the basis of his work's being like a written inscription? The problem here is one both of logic and of literary originality and creativity.
Secondly, if, as has been traditionally supposed, Thucydides is targeting Herodotus, whether primarily or exclusively, in his critical allusion to 'competition pieces for present hearing', how can his argument sustain itself, given that Herodotus had certainly produced a 'monumental' written text before Thucydides? (I here use the term 'monumental' in the sense of 'composite', as G.S. Kirk, though an oralist, refers to 'the monumental composer' of the Iliad.) This second difficulty would seem to be compounded, if it can be shown that Herodotus also had likened his written text to a monumental inscription. In my opinion, however, these two factors are not so much difficulties as additional complexities which need to be factored into a more nuanced interpretation of Thucydides' argument.
The respective prefaces begin as follows: [31]
(1) 'Hecataeus the Milesian speaks thus: I write these things as they seem to me to be true. For the accounts of the Greeks are many and ludicrous, as they appear to me.'(2) 'Of Herodotus the Halicarnassian this [is - the verb 'to be' is understood] the demonstration of his enquiry, in order that neither should the things that came to be by human beings become faded away (exitela) through time nor should great and wonderful deeds, some demonstrated by Greeks, some by barbarians, come to be without fame, and especially for what reason they made war against each other.'
The most difficult element here is the understanding of exitela. The rendering 'faded away' is intended to keep open two possibilities, which will be assessed below. The rendering 'came to be'/'come to be' is intended to highlight a paradox in the use of gignomai (used both of 'birth' and of 'death').
(3) 'Thucydides the Athenian wrote up the war between the Peloponnesians and Athenians, how they waged war on each other ...'.
As their prefaces develop, both Herodotus and Thucydides make the same move as does Hecataeus in his opening words: that is, they switch from referring to themselves in the third person to first-person references (Herodotus 1.5.3-4; Thucydides 1.1.3, etc.).
The scholarship on this topic is not easy to summarise. Some of it is excessively brief, some of it excessively long-winded. Of the scholars who read (either one, or two, or all, of) these prefaces as likening the works to monumental inscriptions, some do little more than state the interpretation but others provide arguments. The arguments vary.
I first present a mélange of the arguments in list form (adding one or two of my own):
the mere act of commemoration (all three authors); the avowed purpose of commemoration, that is, a statement of the function of the text (Hecataeus/Herodotus); the use of writing (all three authors); the stated use of writing (Hecataeus, Thucydides); the fact that suggrapho, Thucydides' word for his 'writing up' his History, can be used of decrees etc., as can prographo, Thucydides' word for his 'pre-writing' of the causes of the war (1.23.5); the naming of the maker of the artefact; the combination of named third person (outside the main text), announcement of the presence of the text (Hecataeus: 'I write these things'; Herodotus: 'this is the demonstration ... '), the use of the deictic 'this' (Hecataeus/Herodotus); the use of the deictic 'this' as the subject of the sentence with the verb 'to be' unexpressed (Herodotus); then enunciation of the text itself with a switch from the 'external' third person to the first person of the actual speaker/writer and assertion of its truth content; in this switch some have seen the influence of official letters (perhaps particularly letters from the Persian king), which could be memorialised in inscriptional form, and/or of Persian historiography (two influences which might have been particularly strong on Hecataeus of Miletus, especially given the space he creates between himself and 'the Greeks'); Herodotus' use of the term exitelos; Thucydides' use of the past tense, his phrase 'for always' and his use of the verb keimai; the evident familiarity of both Herodotus and Thucydides (and undoubtedly also of Hecataeus) with inscriptions of every kind, including the official letter. [32]
It will be noticed that this mélange of arguments seems to canvass two different types of inscription: the funeral memorial or inscription and the governmental inscription. This is not in itself a problem: one might align a particular historian more with one of the two (Hecataeus with the governmental one, Herodotus with the funeral memorial), or one might allow for the possibility of a sort of composite conception of an inscription (this could be true of Thucydides, in his opening words rather like Hecataeus, but at 1.22.4 hosoi ... boulesontai skopein invoking a specifically Athenian inscriptional formula).
In the evaluation of such readings, several other factors need to be taken into account.
First, it is widely agreed (and is clearly true) that Herodotus is imitating Hecataeus' preface and Thucydides in turn is imitating Herodotus'.[33] That Thucydides is also imitating, or reacting against, Hecataeus is indicated not only by the general similarity of the initial prefatory statements but by the common use of the verb 'write' and also by Thucydides' dismissive wording at 22.2 'nor just with what seemed to me to be so', which looks like a confutation of Hecataeus' 'as they seem to be so' and 'as they appear to me'. [34] Thus a successful case for a 'monumental inscription' reading of the preface of any one of these authors might in itself support the case for such a reading of the other two.
Second, as a counter-balance to the official letter or to any Persian parallel, the precedent of Hesiod in the Theogony could be urged: there, as in the three historians, the author moves from third-person self-naming, to mention of his literary product, to first-person reference, to the claim of his own truth as opposed to others' falsehoods (Theogony 22-35). And Hesiod was certainly generally influential upon both Hecataeus and Herodotus and also (I believe) influential upon Thucydides both in his prefatory material and in the main body of Book I. [35]
Third, the preface of Herodotus has of course stimulated other interpretations, most of which focus particularly on Herodotus' use of the term apodexis ('demonstration'). These interpretations may be divided into three main categories:
(a) those which take Herodotus to be stressing the oral nature of his work: his work as an oral 'demonstration', a force apodexis can certainly have (shading into epideixis);
(b) those which see apodexis as dramatising the transition from oral (as in the acquisition of the bulk of his material, the oral performances preceding the construction of the unified text, or the fiction of Herodotus as the voice of the actual text) to the written text as it is read (hence 'this is the demonstration');
(c) those which appeal to alleged parallels in other 'wise men' or sophists for the claim, even the obligation, to dispense superior knowledge.[36]
Of these categories, (a) would be incompatible with a monumental inscription reading; (b) would be quite compatible (since the writer of an inscription is characteristically represented as speaking), and (c) would be indifferent. Although the question is much debated by very distinguished scholars of all persuasions, an investigation such as the present one cannot take refuge in endless deferrals, and I shall here simply state my own opinion that: (a) is quite perverse (in what oral context would the preface make any sense?); the parallels adduced for (c) are (with one immaterial exception, that of Gorgias' Helen - see n. [36]) unimpressive; and (b), which takes proper account both of Herodotean usage within the text, of the numerous oral encounters which provided most of the raw material of the text, of the fiction of 'the speaking text', and of the indubitable fact that the History in its final form is a text constructed by Herodotus himself, the interpretation, in short, which has the widest explanatory power, is clearly right. And (b) is, as I have said, quite compatible with a monumental inscription reading of Herodotus' Preface.
As for the second factor to be taken into account, the parallel with Hesiod, there are several interpretative possibilities. One could say that Hesiod is not an active influence upon any of the historians' prefaces but merely illustrates a common process of validation. I personally think that Hesiod is an active influence, certainly on Hecataeus (with his distinction between truth and [relative] untruth) and on Thucydides. If one says that Hesiod is an active influence, one can then, if one is happy with the notion of Hesiod's using writing (as I am), talk of Hesiod's 'inscription' in a broader sense of that term. If one rejects Hesiod's having used writing, one can say that the historians have transmuted Hesiod's oral technique of validation into inscriptional validation. A final point: acceptance of Hesiod as an active influence upon the historians' prefaces in no way entails his having been the only influence, and indeed in all three historians plurality of influences is readily demonstrable.
The way is now cleared for return to the evaluation of the arguments for some sort of inscriptional reading of the various prefaces. I shall consider what seem to me to be the three most crucial separate questions: (1) the case as based on the overall format of the three prefaces; (2) the argument from Herodotus' use of exitela; (3) possible differences between the three prefaces.
(1) Svenbro makes a gratifyingly clear formal statement: [37] 'the works of all these historians thus bear monumental inscriptions in the sense that, seen "from outside" (this is the first phrase that permits the reader to enter into the work), they refer to their authors in the third person, as if they are absent. The authors are no longer there. Only through a fiction - a fiction that for us has become a rule - do they refer to themselves in the first person inside their works, just as if they were present in the text. After the introductory sentence, their works thus assume the form of a transcription of a living voice, as if they had first existed orally and only later were faithfully transcribed. In contrast, where Hesiod uses the form of a transcription, referring to himself as ego, or even hode ego, thereby setting himself up as an interlocutor who is present, it probably was a real transcription, whether made by Hesiod himself or by someone else'. But this statement falls considerably short of proving the case that such 'monumental inscriptions' are the literary equivalent of actual monumental inscriptions. The case can be strengthened as regards Hecataeus by the clear parallel between his format and those of official letters, [38] but is perhaps in itself as yet not decisive, especially given the clear formal parallel between Hecataeus and Hesiod.
(2) Herodotus' use of exitela can be argued to be 'inscriptional' in two ways: (a) if exitela is a recognisably genealogical term (= 'extinct'), this might align Herodotus' work with funerary inscriptions; (b) if exitela conveys, rather, 'fading', this might suggest (in some sense) the 'fading' of an inscription.
(a) The case for exitela's being a recognisably genealogical term [39] rests on several factors: (i) Herodotus' only other use of the term at 5.39.2 (genos to Eurustheneos genesthai exitelon = 'Eurysthenes' family became extinct'); (ii) the possibility that the wording of the Preface is itself to be understood 'genealogically' or 'biologically', with ta genomena ex anthropon suggesting 'things born from men' and with exitela suggesting telos ('end'/'death');[40] (iii) the (possible) broader parallel between the Preface and 5.39.2: hos mete ta genomena ex anthropon toi chronoi exitela genetai ~ genos to Eurustheneos genesthai exitelon (where the similarity consists not only in the underlined wording but also in the paradoxical use of genesthai/genos, as applied both to 'birth' and 'death'); (ii) and (iii) can be regarded as being mutually reinforcing; (iv) the fact that later writers seem to have understood Herodotus' use of exitela in the Preface 'genealogically'. [41] This seems to me a good cumulative case.
(b) Luce writes: 'the metaphor in the first clause suggests a stone inscription whose letters fade with weathering', thereby apparently suggesting: 'Herodotus represents his History as a superior sort of inscription, for a literary text which can be copied and circulated widely can achieve the permanence denied to stone inscriptions located in one place and anyway subject to physical decay' (as indeed, to date Herodotus' History has achieved permance).[42] It seems that in ancient usage generally the predominant application [43] of the word exitelos is to 'fading' (of colours etc.). [44] But what sort of 'fading' would apply to a stone inscription? Later examples talk of the 'fading' of the paint inside the letters of inscriptions. They also apply the notion of 'fading' to non-inscriptional written texts [45] and to paintings themselves.[46] The former category is obviously excluded as Herodotus' analogy here, but might he be contrasting his work with pictorial representations of the Persian Wars such as those on the Stoa Poikile? In that case, exitela would give no clue to the nature of his own metaphorical representation of the past. To our logic, it would seem to make little sense if Herodotus were referring to the 'fading' of letters on a stone inscription, since (a) letters themselves do not 'fade' in the same way as colour does, and (b) if the colour of letters fades, the letters themselves remain (at least for a very long time) readable. This difficulty, however, can be overcome by the interesting case of Thucydides 6.54.6, who refers to the 'faint' (amudrois) letters of an inscription of which the letters are still clear today and who must therefore be referring to the disappearance of the paint.[47] It is true that amudros rather than exitelos seems to be the mot juste for such 'fading' of inscriptional letters, but exitelos is also attested in this sense. [48]
There are, then, things to be said for both interpretations and no decisive arguments against either. One might even attempt to combine the two, for on the usual etymology, which seems correct, exitelos simply derives from exeimi = 'go out', [49] and both letters (or their colours) and families/'things born' can reasonably be said to 'fade away'/'fade out'. Indeed, Aeschylus' allusion to the 'fading' of blood (fr. 162.4) may perhaps even bestraddle the colour and the genealogical usages.
My own feeling, however, is that (a) is the stronger case, because it seems to me more organically integrated both into the wording of the Preface and into (if one may so put it) Herodotus' general philosophy of life, and because in the Greek it is difficult not to take ex anthropon as affecting not only ta genomena but also exitela, which is rather easier if exitela refers to genealogical 'fading out'. Nevertheless, (b) can hardly be excluded (particularly if there are other reasons for regarding the History as an 'inscription'). Both interpretations, therefore, provide some support for an 'inscriptional reading' of the Preface.
(3) Hecataeus is certainly evoking an epistolary form. But the fact that he follows up the epistolary formula 'so-and-so says' with 'I write' emphasises the written nature of his work. The combination of verbs of saying with those of writing is of course commonplace in inscriptions. It would be easy to read Hecataeus' work as an 'inscription' of an official governmental letter, especially given Hecataeus' extremely lordly tone, and even as a specifically Persian letter, given his dismissive allusion to 'the Greeks', though allowance must also be made for Hesiodic influence. If Hecataeus is considered in isolation from the rest of the historiographical tradition, the case for an inscriptional reading, though pretty plausible, is perhaps not quite made out.
Despite some obvious differences from Hecataeus, the case for an inscriptional reading of Herodotus' Preface (again considered in isolation from the rest of the historiographical tradition) is of approximately the same weight, based as it is on general format (the move from 'external' third person to 'internal' first person etc.), explicit commemorative purpose, deictic 'this' as subject without expressed verb 'to be' and use of exitelos. There are no strong negatives: as we have seen, strongly non-inscriptional readings of apodexis fail and the oral residue implicit in apodexis may always be part of an inscription, nor does Herodotus' general narrative tone preclude an 'inscriptional' claim.[50]
At 1.22.4 hosoi ... boulesontai skopein Thucydides is clearly imaging his History as an inscription (section 2).
We may now essay a general conclusion on this question. Given that 1.22.4 is decisive for Thucydides, given that all three works actually are texts and that the formats of their prefaces are formally the same as those of inscriptions as well as of other media (non-inscriptional letters; Hesiodic poetry), and given the close patterns of imitation which exist among the three historians, there seems to me a persuasive cumulative case alike for the initial prefaces of Hecataeus, Herodotus and Thucydides. One might perhaps water this down by saying merely that Thucydides could have chosen to read his forerunners' prefaces as inscriptional, but I think the case for their actually being so is objectively strong.
We may also at this point take account of two further factors: that, as we have already seen (section 4), some ancient authors understood Thucydides' ktema as a monument, and that, as we shall see (sections 12 and 13), Livy too understood the ktema as a monument and Arrian understood the whole business of beginning historiographical prefaces with one's name and particulars as inscriptional. Hence the case for an inscriptional reading of Hecataeus', Herodotus' and Thucydides' prefaces as inscriptional seems effectively decisive. As already suggested, however, this conclusion does not entail their being all 'inscriptions' of the same type. Herodotus' 'inscription' is somewhat different from Hecataeus', an honorific commemoration of great deeds, a funerary monument set up by Herodotus, rather than a peremptory epistolary inscription asserting its writer's superior authority to deluded fools, an inscription, moreover, whose racial even-handedness implies a generous contrast both with the narrow nationalism characteristic of physical inscriptions and with Hecataean racially-defined arrogance (cf. also n. [80]). Thucydides' 'inscription' is more for information and the promotion of understanding 'for anyone who wishes to look'. Such apparent diversity of conception does not invalidate the general reading: indeed, from the point of view of the competitive literary creativity which is so marked a feature of ancient historiography, it is positively better if these historiographical 'inscriptions' are seen to differ in their nature. I shall return to these matters in section 14.
We may therefore now revisit (temporarily!) the two related problems arising from Thucydides' use of the monumental inscription image: (1) in what does his 'inscriptional' superiority to his historiographical predecessors consist, when they too had presented their works as inscriptions? (2) what sense can be made of his claim that his work is a 'possession set down for always rather than a competition piece for present hearing', when Herodotus had already imaged his History as an 'inscription', especially if Thucydides is targeting Herodotus, whether primarily or exclusively, in his allusion to 'a competition piece for present hearing'?
(a) Provided relevance and logic in the immediate context of 1.22.4 can be satisfied, such targeting of Herodotus would fit Thucydides' general procedure in his extended Preface (1.1-23), built as it is on an extended parallel and contrast between himself and Herodotus, and containing detailed allusions and criticisms in chs. 20, 21, 22 (our chapter) and 23.[52] I shall return to the appropriateness of an allusion to Herodotus in 1.22.4 in more detail below in connection with items (b), (d) and (g).
(b) There is an ancient tradition that Herodotus delivered oral lectures throughout Greece but especially in Athens; this tradition is admittedly 'late' and therefore distrusted by some scholars, [53] yet it harmonises well enough both with general reconstructions of the cultural milieu of the period and with more specific hypotheses about how Herodotus projected himself before he finally produced his complete text, e.g. that he was the last of the old Ionian story-tellers (Murray), or that he gave numerous 'pre-publication' lectures or readings (Evans), or that his public deportment would not have been unlike that of a sophist (Thomas).[54] And, like the sophists, Herodotus, according to the literary tradition, received money for his performances. Of course all of these hypotheses depend largely on evidence from within the text, some of which will be mentioned below.
(c) If Antigone 904-20 is authentic (there is scholarly debate),[55] the passage provides external evidence that Hdt. 3.119 was in some form available in Athens in ca. 442, and a form that must have been distinct from its expression in the complete text.
(d) As we have seen in section 8, apodexis in the Preface may very well allude (in some way) to oral performance by Herodotus.
(e) The most natural explanation of Herodotus' comment concerning the historicity of the Persian Constitutional Debate (3.80.1 'speeches were said which are unbelievable to some of the Greeks but nevertheless they were said') is that he had already tried his version out orally, encountered some disbelief and felt bound to reassert its historicity in his written version.[56]
(f) The History contains several wandering wise men figures - notably Arion (1.23-24), Solon (1.29-33), and Anacharsis (4.76.2) - who dispense their wisdom orally and look very much like alter egos of Herodotus himself.[57] Of course to say this is no doubt to some extent circular - these figures look like one's preconceptions of Herodotus, but it is far from totally circular (in all cases there are other textual indications). It should be needless by now to say that to interpret the figures in this way is not to restrict them to this interpretation.
(g) Herodotus' description of the arrival at Sardis of the most important of these figures, Solon the Athenian (1.29 'there arrived at Sardis at the peak of her wealth all the "sophists" from Greece ... and above all Solon the Athenian'), can be argued (among many other things) to dramatise Herodotus' arrival in Athens and his own ambiguous relations with the sophists.[58] This passage seems to me to provide solid evidence from Herodotus himself for Thomas' hypothesis (item (b) above), although such a judgement is of course very much a 'literary' one.
(h) The ultimate mythical ancestor of all such figures is Odysseus, and Odysseus provides the single most important paradigm for Herodotus' role, both with regard to the construction of the text and to the various activities preparatory to that construction.[59] The specific role of the wandering oral performer could hardly be more fittingly represented than by Odysseus.
As a cumulative case for Herodotus' having been celebrated in his lifetime as an oral performer and hence being alluded to in 1.22.4, this case seems to me a strong one. Of course, there are objections over and above the two central problems which have yet to be faced. One objection might be that 'competition piece for present hearing' looks more like an allusion to sophistic performances,[60] and venal ones at that,[61] but this objection is well met by three counters: (i) as usual, the case for an allusion to one author does not entail the exclusion of allusion to others as well; (ii) items (b), (d) and (g) actually support a contemporary association between Herodotus and sophists; (iii) analysis of the thought patterns created by Thucydides' arrangement of his material in chs. 21-2 positively reinforces this association, for in ch. 22 'a competition piece for present hearing' picks up the idea of the relative unpleasurableness of Thucydides' History due to the absence of 'the muthos-element', an element which explicitly conduces to pleasure, and, whatever the muthos-element is, it is surely something that Herodotus has, and the allusion to the muthos-element in turn clearly echoes 21.1's allusion to 'what the prose-writers have put together for the purpose of enticement to the audience rather than the truth, things that cannot be checked and the majority of them having won out owing to time untrustworthily to the point of to muthodes', an allusion which must include the 'prose-writer' Herodotus.
But of course the first difficulty cannot really be divorced from the second, since, as we have seen, the claim to be producing an inscription is necessarily bound up with, indeed reinforces, the claim for the use, and superiority, of writing. And the second difficulty seems less superable:[63] not only had Herodotus unified his multifaceted 'enquiry' into his great monumental text, but Thucydides' own detailed and intense engagement with Herodotus in 1.1-23 is clearly at least mainly with that text, rather than with Herodotus' oral performances or even his readings (if he gave readings). But the fact that Herodotus was celebrated as an oral performer and that oral performance was at least the primary means of his presentation and of his reception at least before he finally presented his complete text in c. 426, [64] presumably not long before his death, makes it not unreasonable for Thucydides to make a broad distinction between a written text designed primarily for reading and oral productions, and implicitly to locate Herodotus on the oral side of the distinction. The same procedure would be not unreasonable in the case of the sophists, primarily associated with oral performances yet in many cases also having produced written works.
The business of reception (Thucydides' main concern in 1.22.4) gives the distinction still more validity. As we have seen, even Thucydides does not wholly exclude oral performance and aural reception of his own work and some scholars have maintained, reasonably enough, that there are segments of Thucydides perfectly well suited to recitation (though a considerate host might jib at a performance of 'The Corcyraean Stasis' at his symposium) (cf. n. [28]). Of course audiences are capable of responding to oral productions with considerable insight and subtlety and fifth-century Greek audiences were (no doubt?) better than we are today. Nevertheless, we may still suppose that the reading experience would have offered more. Pleasant and instructive as it is to listen to 'A Book at Bedtime', one gets far less out of it than by reading the book oneself.
Moreover, the notion of 'segments' of a work introduces a crucial qualification. It is one thing to hear a single performance or even a string of performances, quite another to have the opportunity (which only reading can provide) for engagement with a great work of literature in its complex totality. Modern analyses of the texts of both Herodotus and Thucydides have uncovered numerous subtleties which simply could not be detected in oral performance and whose existence cannot be denied by claims that the conception and reception of these texts were predominantly oral. For these claims are simply circular and in any case fail adequately to explain the very existence of the complete text.
To take two simple and surely indisputable [65] examples of such subtleties. The links between the beginning and end of Herodotus' History or between Pericles' speeches and Cleon's are only detectable through the medium of reading (and careful 'close reading') of the monumental text.[66] Thucydides' distinction in 1.22.4, then, is not an absolute one and obviously never can be: someone who owns even a complete work may choose to read it himself in excerpted fashion or to read it to, or with, friends in excerpts; moreover, Thucydides' History never achieved completion and must have been issued piece-meal. Nevertheless, the distinction remains a real and vital one. Nor is the distinction exactly a statement of the superiority of 'a written culture' (as some scholars have put it): rather, it says that ultimately the reading experience offers far more. In the age of the Internet, Thucydides remains absolutely and importantly right.[67]
In one of our many recent phone conversations about Thucydidean matters Gordon Howie suggested an agreeable scenario (with notable 19th- and 20th-century parallels) whereby oral performance or recitation by the author, reception by an audience, reception by a readership, acknowledgement by the author of the possibility of oral performance and aural reception, and his assertion of the superiority of the reading experience might all be plausibly combined: a recitation or reading after which the audience was invited to acquire the book. Why not?
I conclude, then, that Thucydides is indeed targeting Herodotus in the phrase 'a competition piece for present hearing', though the targeting includes sophists and analogous figures with whom Herodotus might not altogether unreasonably be associated; that the distinction between Thucydides' 'possession for always' and 'competition pieces' involves a distinction between writing/reading and oral/aural; and that his use of the image of the inscription reinforces both his use of writing and his claim for superiority over Hecataeus and Herodotus, because his inscription records past, present, future and all time. But even as the image of the inscription registers respect for inscriptions as a medium, it also conveys the supremacy of Thucydides' inscription over mere physical inscriptions, in just the same way as Thucydides' 'possession' transcends any conceivable materialist possession. Some modern historians and archaeologists affect to believe that in the study of the past physical evidence is more valuable than literary texts, but no inscription, no collection of inscriptions, however large, will ever remotely approach the value of Thucydides' work, which appropriates such value as inscriptions possess and then multiplies it to an infinite degree.
We are now also in a position to answer the final question left pending in section 3, that is, whether the immortality of Thucydides' work is promoted by its being a written text, to be received primarily by readers. The answer, it should now be obvious, is 'yes'. Indeed, on the Howie scenario mentioned above, prospective buyers are given the opportunity to make their own contribution to the work's immortality (and the more who do so, the 'more immortal' that work will be!). Now it is true that Herodotus' preface must be understood as making the same sort of claim: cf. my gloss in section 8 on Luce's interpretation of exitela: 'Herodotus represents his History as a superior sort of inscription, for a literary text which can be copied and circulated widely can achieve the permanence denied to stone inscriptions located in one place and anyway subject to physical decay'. But Thucydides' claim for his work's immortality does not rest exlusively on either its intrinsic quality (its coverage of past, present and future) or on its being a written text designed primarily for reading: it rests on the combination of the two aspects. That is, the vehicle of writing and reading (with all the advantages of close, interactive, reading) makes possible the work's timeless explorations. So Thucydides can still say: 'my work is the only work you need to understand political life and to engage in it competently, no matter what your own context or time, hence it will be immortal', especially since he is able (with both justice and injustice) to represent Herodotus as primarily an 'oralist'/'auralist'.
I turn now to aim 4 of the paper: the attempt to show that qua readers and critics of 1.22.4, some of Thucydides' historiographical successors understood Thucydides' claims better than modern scholars do. That is (of course), I have to show that these successors understood Thucydides in the same ways as this paper does. The superscription adumbrates the point: Dionysius of Halicarnassus saw that Thucydides' 'possession' was a 'memorial' (and indeed incorporated this insight into a rather clever and elegant turning of the tables upon Thucydides (n. [17])). So too did Thucydides' biographer Marcellinus (section 4). There will be dispute about the quality of such writers as literary critics, especially in the case of Marcellinus: it is at least of some interest and importance that two ancient writers not normally considered as of the first rank saw a great deal more (in this respect) than did Pohlenz, Gomme, Walbank, Andrewes, Dover, Canfora, de Romilly, Macleod, Lateiner, de Ste. Croix, Stadter, Lewis, Wiseman, Woodman, Lendle, Hornblower, Thomas, Badian, Moles, Rhodes, Pelling, Rutherford, Crane, Rood, Marincola ... (fill in any names you like). But the case will be more interestingly made through consideration of three acknowledged heavyweights in the agonistic tradition of ancient historiography.
'If one takes away from history the why, how and wherefore [69] the thing that was done was done [70] and whether the end it had was logical, what is left of it is a competition piece, but no learning accrues, and it gives pleasure for the present but does not help at all for the future. (3.32.1) Therefore one must consider those people to be ignorant who assume our work to be difficult to acquire [duskteton] and difficult to read because of the number and size of the books. (3.22.2) For how much easier is it both to acquire and read through forty books as it were woven together thread by thread and to follow along clearly the actions in Italy, Sicily and Libya from the times of Pyrrhus to the taking of Carthage, and those in the rest of the inhabited world from the flight of Cleomenes the Spartan continuously until the battle disposition of the Achaeans and Romans at the Isthmus, than to read or acquire the compositions of those who write by separate topics? (3.32.4) For apart from their being many times as long as our History, it is quite impossible for their readers to get anything securely from them, firstly because of the fact that the great majority of them do not write the same things about the same things, then because they leave out the contemporary actions, of which when reviewed and judged together by means of comparison each event receives a different estimation than it would by separate assessment, and they are altogether unable even to touch on the most essential things. (3.32.6) For crucially we assert that the most necessary parts of history are the things that come after the deeds and that the things that follow concurrently with them and above all the things to do with the causes. (3.32.7) We see that the war with Antiochus took its departure-points from the war with Philip, and the war with Philip from the war against Hannibal, and the war with Hannibal from the war about Sicily, and the things between these events had many and varied characters but all tending to the same underlying purpose. (8) It is possible to recognise and learn all these things through those who write a whole history but through those who write the wars themselves, impossible, unless indeed anyone reading the battle disposition themselves takes it that he has got a clear knowledge both of the management and character of the entire war. (10) But none of this is at all possible: rather I take it that, as much as learning differs from mere hearing, so our History differs from the compositions by separate topics.'
Interpretation of this passage has greatly benefited from the discussions of highly distinguished modern scholars. But while these discussions have said important things about Polybius' general debt to Thucydides, about his theory of causality and about his notion of 'interwoven history', [71] they seem to miss most of the intelligence and subtlety of the argumentation and all of the verbal cleverness deployed in the underpinning of that argument.
An important preliminary point (seemingly ignored hitherto) is the elegant verbal play in 31.12-13 centred on 31.12 aphelei and 31.13 ophelei. To 'take way' (aphelein) from history the complex analysis of causality deprives it of the ability to 'help' (ophelein) for the future. It is as if the apo-prefix in aphelei functions as an apha-privative, itself taking away the omega of ophelei: the result, deprivation, not increase. Moreover, the use of the term aphelein glosses the stylistic term apheleia: that simplicity of style from which unnecessary excrescences are 'taken away'; for Polybius, however, such a 'take away' style ('taking away' elaborate analysis of causality) is to 'take away' history's greatest value. This complex verbal elegance should alert us to the sophistication of Polybius' writing in this passage.
Here, as elsewhere, Polybius clearly interprets Thucydides' 'helpful'/'useful' (1.22.4) as implying 'practical usefulness based on political understanding' rather than merely 'political understanding'. That, of course, is a matter of modern scholarly debate, too intricate to broach here; I shall simply state here my opinion that Polybius is right.[72] As for the insistence on the importance of causality, thus far it looks sufficiently Thucydidean (~ 1.23.5-6), though developed in detail.
But the following justification of his historiographical programme utilises Thucydidean motifs to define both the similarity and dissimilarity of that programme to Thucydides'. The confrontation with the difficulties felt by some of his readers at 32.1 ('therefore one must consider those people to be ignorant who take our work to be difficult to acquire [duskteton] and difficult to read because of the number and size of the books') echoes Thucydides' remark at 1.1.3: 'to discover clearly the things before it [the Peloponnesian War] and the things still more ancient was impossible because of the quantity of time intervening'. But whereas Thucydides was there talking about events before the Peloponnesian War, i.e. things strictly outside his proper sphere of 'the present war', Polybius' proper concern is precisely with the past and with a very extended and complexly interconnected past at that. Polybius' project is in this sense fundamentally un-Thucydidean.
The nature of the difficulty experienced by those readers also repays consideration. The difficulty is a double one. Obviously, readers might find Polybius 'hard to read' (dusanagnoston) - though Polybius will go on to argue the contrary, but, apparently mundanely, Polybius also mentions that some have found his work 'hard to acquire' (duskteton), it being of forty books. The next sentence twice repeats this double motif. The effect is to put great emphasis on the notion of 'acquiring', the same word (whether in adjectival or verbal form) as Thucydides had used for his 'possession set down for always'. Thus Polybius creates an intertexual implication based on a sort of 'absent present': are the apparent 'difficulties' of his own History so great as to disqualify it from being a Thucydidean ktema? Do those difficulties make it a dus-ktema, a misacquisition? He here also clearly registers one important aspect of Thucydides' ktema, namely that it is a transference from, or appropriation of, materialist terminology. But for him there is a genuine materialist difficulty - the practical difficulty some readers have in getting hold of all forty books - and the concomitant fact that they regard this as a 'difficulty' in the way of their reading him. He also begins to broach the question of the relationship between a ktema and its size.
The sequel deploys some fairly obvious Thucydidean motifs. (1) There is the Thucydidean criterion of 'clearness' (32.2, 9 ~ Thuc. 1.1.3; 22.4). (2) The criticism at 32.4-5 of 'most' of the historians of single wars clearly echoes Thucydides 1.22.3 on the difficulty of finding out the truth of the deeds: 'they were discovered with much labour, because those who were present at each particular deed did not say the same things about the same things'. Polybius, however, transfers this difficulty from the historian to the readers of these separate histories as a way of boosting the credentials of his single, continuous, History. (3) Polybius maintains Thucydidean causation terminology. (4) The whole section closes, in a formally and logically satisfying ring structure, with a reformulation of the Thucydidean claim of 1.22.4.
But there is again a more subtle debate going on beneath the surface. It is Polybius' contention that, despite the two apparent difficulties posed by the length of his work (that of readers' getting hold of the books and that of their reading and understanding them), his work is thoroughly unified (32.2) and secures the complex unity of understanding of historical events which the medium of separate histories of individual wars simply cannot provide. To underpin this case Polybius consistently uses sun-compounds (the normal Greek equivalent of Thucydides' archaising xun-compounds) of his work, its historical methods, its subject-matter and its goals: 32.3 to suneches, 32.5 suntheoroumenon ... sugkrinomenon, 32.7 sunneuousas, 32.9 sumpantos. This complex is sharply contrasted with the 'partial' treatments of those who write about single wars: 32.3, 5, 10. Thus the application of the term suntaxis (32.3, 10) to their 'partial' histories acquires an ironic edge. Given Polybius' systematic allusion in this passage to Thucydides 1.22, this play on the contrast between the writer's properly sun-treatment and others' partial and inferior efforts and the ironic application of a sun-compound to those inferior efforts must recall Thucydides' claim that his work xugkeitai in contrast to the logographers' inferior xunethesan. In other words, Polybius clearly sees that another aspect of Thucydides' ktema claim is its transcendingly 'synthetic' quality. Yet his own stance is again simultaneously both Thucydidean and out-Thucydidesing Thucydides. Polybius' History is much longer than Thucydides' (40 books as against 8); this length is necessary because only so can one incorporate the developed Thucydidean analysis of causation that Polybius holds to be indispensable for true historical understanding; but this length allows a far more unified treatment and far more systematic understanding than accrues from individual treatments of wars. At this point, one cannot fail to recall that Thucydides too comes into the misguided category of 'those who write the wars themselves' (32.8).
In sum, this is a brilliant piece of writing by Polybius, whose articulation of his own historiographical claims involves a thoroughly astute reading of Thucydides 1.22 and an appropriation of his great predecessor's terminology which out-Thucydideses Thucydides himself. Not Thucydides' ktema, but Polybius' own vastly longer and more complex but nevertheless still more unified ktema, is 'the one thing you've got to have'. From the point of view of this paper, the single most significant thing is the complexity of Polybius' understanding of Thucydides' conception of 'possession', a complexity which (I categorically state) dwarfs that of any modern scholar's interpretation of that concept.
I here concentrate on one particular section (10), where Livy defines the purpose of History:
Hoc illud est praecipue in cognitione rerum salubre ac frugiferum omnis te exempli documenta in lustri posita monumento intueri; inde tibi tuaeque rei publicae quod imitere capias, inde foedum inceptu, foedum exitu, quod vites.
No English translation can hope to convey all the implications of the Latin. The following translation tries to convey those that are relevant to the present enquiry:
'This is the thing that is most takingly wholesome and fruitful in the getting to know of history: the fact that you look upon the lessons of every kind of example placed on a conspicuous monument; from these you may take for yourself and for your own (e)state - the public state, what you should imitate, from these, disgusting in the undertaking, disgusting in the outcome, what you should avoid'.
From our point of view, this passage is significant in two respects: (1) Livy clearly sees that at 1.22.4 Thucydides images his History as an inscription; (2) Livy understands very well the main implications of Thucydides' appropriation of 'property'/'possession vocabulary (ktema) and of his application of it to his History, though Livy himself 'redefines' this vocabulary in a new way appropriate to his own historiographical purpose.
(1) Livy's 'monument' is at once Roman history itself and his own textual version of it, the Ab urbe condita. This monument contains 'lessons' of men's deeds; it is 'conspicuous/illustrious'; one has to 'look at' it. The wording therefore glosses, indeed almost translates, Thucydides' 'look at the clearness' of his History as imaged as a monumental inscription.
(2) In the 'inde'/'from these' clause Livy spells out in more detail the benefit the individual reader will gain from reading his History. He initially seems to appeal the reader's self-interest - 'for you and for your own (es)tate'. 'Tuae rei' draws on the vocabulary of personal property and advantage, but then the appended publicae (by the rhetorical figure known as correctio) sharply redefines personal advantage as indistinguishable from that of the state at large. (T)uae rei glosses Thucydides' ktema. Livy sees that it is part of Thucydides' claim for his ktema that it has generalising exemplary power and transcends the time scale of its narrative. And like Thucydides (but as it were in competition with him), Livy implicitly claims that his History is 'the one thing you've got to have', sc. for the salvation of Rome.
'As to who I am that I make this judgement on my behalf, I do not need to inscribe [anagrapsai] my name, for it is not at all unknown to men, nor what my native land is, nor my family, nor if I have held any office in my own land; but this I do inscribe [anagrapho], that my native land, family, and offices are these accounts and have been even from my youth. And for this reason I count myself not unworthy of the first place in Greek speech, just as I hold Alexander to have been in arms.' [75]
Arrian's declining to give his names, land, family and offices has aroused scholarly debate. On one reading, this declining pointedly subverts historiographical convention (no doubt with additional polemic against the excessive self-advertisement of some contemporary historians); on another, there are certain literary contexts where (à la Homer) the suppression of one's name etc. is a matter of literary modesty or decorum, properly diverting attention from the author to his subject-matter. [76]
The first thing to note is that the use of anagrapho (picking up anagrapsai) is in context very pointed. 'This can be a relatively colourless historiographical term ("record", "give an account of"), but can also mean "inscribe" on a monument, as the names and honours of prominent people regularly were "inscribed" in the ancient world. Arrian, then, does not need anagrapsai his name, country, family or offices as if on an inscription: his claim to commemoration, which he will "inscribe", lies in his emotional commitment from youth to hoide hoi logoi. Delicately, but unmistakably, Arrian is implying that hoide hoi logoi will be an imperishable artefact, just as Thucydides' history was "a possession for ever", Pindar's poetry a "treasury" and Horace's odes a "monument". This in turn implies a judgement on the relative importance of public achievement and literature in Arrian's life. The former is important in its own right, but trivial compared with the glory that comes from literary endeavour (an idea to which Arrian returns at the end of the Anabasis: vii 30.1).' [77]
This analysis, fair enough as far as it goes, is subtly developed by Marincola: [78] 'like Homer, Arrian wishes the deeds to be considered foremost, but unlike Homer his anonymity calls attention to itself, and in this sense Arrian's self-display remains well within (and plays upon) the historiographical tradition. One suspects too that something more is at stake when we look at the actual opening of the Anabasis: "Ptolemy, son of Lagus, and Aristobulus - whatever these two say in common about Alexander the son of Philip, these things I have written as entirely true". The form of the names in the nominative hearkens back to the early historiographical openings and these explicitly named sources stand in stark contrast to the narrator's anonymity. For Arrian's original audience, however, we must assume that there was, of course, not confusion, but rather admiration for the clever exploitation of an old convention'.
But there is more. The Greek word translated by Marincola as 'I have written' is anagrapho. It should be translated 'I inscribe'. For there is a clear and significant pattern: in the initial preface Arrian uses anagrapho with reference to the names of his primary sources; in the 'Second Preface' he uses anagrapho of his declining to 'inscribe' his own names and particulars. That is, Arrian is putting to the service of his own ambiguous non-self-glorification the fact that the traditional prefaces of Herodotus, Thucydides etc., in which historians named themselves and gave details about themselves, are to be read as 'inscriptions', one of whose acknowledged functions was self-glorification.
One of the ways in which Hecataeus attempted to establish his authority/author-ity within the Greek tradition of his day was to fashion his history as an inscription written under his own name (in implicit contrast of course with Homeric anonymity) and an inscription reminiscent of a governmental edict, even, it seems, a specifically Persian one, directed towards, to some extent against, 'the Greeks'. His immediate successors Herodotus and Thucydides adopted the same image of the inscription in their agonistic competitiveness both with their historiographical predecessors (respectively Hecataeus and Hecataeus and Herodotus) and with the very medium of physical commemoration itself, but they did so with creative variation.
Herodotus made his 'inscription' like that of a funeral memorial but a memorial both of Greeks and barbarians, and he combined the inscriptional quality of his work with the illusion of oral recitation (apodexis). [80] In these ways, he outdid Hecataeus. Thucydides, in turn, made his 'inscription' a double one, Hecataean in its assertive authoritativeness concerning its truth content but at the same time, in an intellectual sense, 'democratically' open to all who wished to look at it. He further combined this double image of the inscription with that of a 'possession', itself an idea of multiple implications, one of the most important of which was that of an 'inheritance'. He also outdid both Hecataeus and Herodotus in the overarching timelessness of his 'inscription'.
All three of these early Greek historians can also be seen to be registering the relatively new importance of large-scale writing as a medium of intellectual exposition and exploration and to be reacting to, and in a sense against, the relatively new phenomenon of the widespread erection of inscriptions in the Greek world.
Polybius, Livy and Arrian apprehended the implications of the prooemial images used by these first great historians and developed their implications in suggestively different ways. Polybius probed the implications of the Thucydidean concept of the 'possession' and appropriated it for his own self-presentation as a Thucydidean historian superior to Thucydides himself. Livy turned Thucydides' intellectual monument and possession into a moral one. By refusing to inscribe his own name and particulars, Arrian cleverly conveyed the impression of repudiating the self-glorification necessarily implicit in recording oneself as the 'inscriber' of a historiographical 'inscription'.
Thus from the very beginnings of classical historiography down to the 'late' Arrian, imagery of apparent permanence and fixity proved, in the hands of skilful and ambitious practitioners, capable of rich and flexible redeployment. As usual (perhaps as always), however, in the case of such literary creativity, it would be a mistake to interpret these diverse developments of the image of the inscriptional inheritance merely in terms of literary one-up-manship (important as that element certainly is): all the developments help to articulate differences of substance between the programmes of the various historians.
Within ancient historiography generally, therefore, the image of the inscriptional inheritance is one of extraordinary richness. The image of the inscription may stand for ancient historiography itself, the child (as it were) of the written inscription (among many other parents). It may also stand for the fecundity, vitality and intellectual power of ancient historiography and its developing and innovative traditions. And it may also stand for the highly superior literary criticism contained within ancient historiography: the best critics, ancient and modern, of ancient historiography are the ancient historians themselves. We should try to learn more from them.
The transliterated Greek of the Dionysius passage which forms the superscription to the paper is given in n. [17], along with fairly full literary exegesis; the passage is also cited in section 4. [Return to text]
[2] The slightly non-standard translation is my own: for its theoretical rationale and practical implications see n. [4]. [Return to text]
[3] In this respect the paper may be regarded as an implicit response to two of Peter Rhodes' (in the current climate, iconoclastic) papers: 'In Defence of the Greek Historians', G&R 41 (1994) 156-71 (a much-translated paper) and ' "Epidamnus is a City": On Not Over-Interpreting Thucydides', HISTOS, July 1998; to the latter John Marincola will be making an explicit response (also on HISTOS). [Return to text]
[4] On the critical principles involved see: J. Moles, 'Cry Freedom: Tacitus Annals 4.32-33', HISTOS October 1998, n. [4]; two practical implications are: (i) aiei is always rendered by 'always', in order to convey the interplay between the different applications (cf. n. [43] below) of aiei (n. [35]); (ii) para- is always (with one impossible exception) rendered by 'present'. An immediate and decisive validation of the principles of 'literalist translation' is provided by n. [5]. The translation is, however, 'tweaked' in detail to suit the concerns of this particular paper: e.g. ophelima is translated as 'helpful' (rather than 'useful') in order to bring out the verbal parallel with Polybius 3.31.12 ophelei (section 11). [Return to text]
[5] For the formula see the discussions and examples adduced in B.D. Meritt, Epigraphica Attica (Cambridge, Mass., 1940) 90, and R. Thomas, Oral Tradition and Written Record in Classical Athens (Cambridge 1989) 60-61 and n. 151; as far as I can find out, no modern historian or literary critic has made the connection between this formula and 1.22.4, although, as we shall see (section 4 and notes [15], [16] and [17]), some ancient readers took ktema as connoting 'monument' and some modern scholars have also done so. Of course, if I have missed some crucial bibliography or some esoteric e-mail lore, I shall be grateful for correction from HISTOS readers. Note that the parallel with the inscriptional formula refutes out of hand the great majority of modern translations, which render skopein by something other than 'look at'; there are other reasons why skopein connotes in the first instance 'look at'; see also n. [29] for the implications of this visual terminology. [Return to text]
[6] The inverted commas are for the benefit of those scholars who continue to fight shy of the notion of authorial intention, although nowadays the theoretical arguments against 'intention' are looking under increasing strain (cf. e.g. S. Burke, The Death and Return of the Author: Criticism and Subjectivity in Barthes, Foucault and Derrida (2nd. ed. Edinburgh 1998) and (despite intertextuality) nowhere more so than in cases of literary allusion. [Return to text]
[7] As far as I can find out, this is a new claim (which I made public in my Leeds paper and its handout, 'Narrative Problems in Thucydides Book I [this paper will be published in a collection of papers edited by John Marincola]), though Franco Basso alerts me to G. Crane, The Blinded Eye: Thucydides and the New Written Word (Lanham-London 1996) 13, as a sort of anticipation. I cite and discuss Crane's remarks in section 5. Again, I would welcome any bibliographical correction from HISTOS readers. [Return to text]
[8] Honourable exceptions are O. Lendle, 'Ktema es aiei: Thukydides und Herodot', RhM 133 (1990) 231-42, and R. Nicolai, 'KTEMA ES AIEI: Aspetti della fortuna di Tucidide nel mondo antico', Rivista di Filologia 123 (1995) 5-26, though I disagree with much in their respective arguments; also interesting is S. Hornblower, A Commentary on Thucydides: Volume I Books I-III (Oxford 1991) 61: 'the words ktema es aiei, "everlasting possession", should be taken to mean, in a general way, "having permanent value", in the sense of being ophelimon, "useful" ... Problems arise if we take "possession for ever" too literally: Th. is not systematic about explaining technical terms ... and takes a great deal for granted'. Nevertheless, these observations of the brilliant but capricious Oxford editor seem to me highly questionable. The italicised phrases seem either question-begging (e.g. why should we regard ktema es aiei as a 'technical term' [which it clearly isn't]?) or circular (e.g. it is always erroneous to interpret 'too literally'). [Return to text]
[9] Because the proliferation of para-compounds in ch. 22 (see n. [4]) encourages one to hear the -chrema in parachrema separately. For -chrema ~ 'money' see nn. [27] and [61]. [Return to text]
[10] So, apparently (e.g.), A.W. Gomme, A Historical Commentary on Thucydides: Volume I Introduction and Commentary on Book I (Oxford 1945) 150 and Hornblower (n. [8]) 61. [Return to text]
[11] Thucydides' work explicitly provides 'the clearness' both of the things that happened and of those which, in accordance with the human thing, are going to happen again some time like this and near the present ones': on the necessary force of the Greek te ... kai see A.J. Woodman, Rhetoric in Classical Historiography (London, Sydney and Portland 1988) 24 (yet many scholars continue to mistranslate, taking the ton mellonton ... esesthai phrase as merely a gloss on ton ... genomenon); on Thucydides' profound and complex concern with time see my papers 'A False Dilemma: Thucydides and Historicism' (n. [1]) and 'Narrative Problems in Thucydides Book I' (n. [7]). [Return to text]
[12] I here use the alternatives 'audience/readers' without prejudice, though in sections 7 and 10 I shall argue primarily for 'readers'. That Thucydides was also claiming his own immortality was well understood by ancient critics, e.g. Marcellinus, Life of Thucydides 35 (cited in section 4); see also further section 6 on the necessary implications of Thucydides' 'inscribing' himself and his name within his 'monumental inscription' and section 13 on the general principle. [Return to text]
[13] LSJ s.v.; note also that keimai (or compounds) can be used of prizes, as e.g. in Pericles' words at 2.46.1 athla gar hois keitai aretes megista, tois de kai andres aristoi politeuousin (= 'the things for which the greatest prizes of excellence are laid down, for these the best men engage in politics' [? - I'm not sure]), though this usage shades into the 'permanent property' usage which I go on to discuss in the main text. [Return to text]
[14] This 'transference' is recognised (whether consciously or intuitively) by those older translators who rendered ktema by 'prize'. [Return to text]
[15] B. Gentili and G. Cerri, Storia e biografia nel pensiero antico (Rome/Bari) 11 n. 23. [Return to text]
[16] J. Svenbro, Phrasikleia: An Anthropology of Reading in Ancient Greece, tr. J. Loyd (Ithaca and London 1993), 150 n. 17. [Return to text]
[17] Nicolai (n. [8]) 21-26. Some of the other passages cited by Nicolai do not obviously support the construction he puts upon them.
Dionysius' verbal cleverness and interpretative acuity are indeed noteworthy.
The Greek goes: ho de rhetor tou te arkountos stochazetai kai tous kairous summetreitai ouk eis anathema kai ktema kataskeuazon ten lexin monon hosper ho suggrapheus, alla kai eis chresin. Hoste oute to saphes ekbebeken, hou protou tois enagoniois logois dei, to te deinos einai dokein, eph' hoi malista phainetai spoudazon, proseilephe.
For a translation see the superscription to this paper. The translations of S. Usher, Dionysius of Halicarnassus: The Critical Essays I (Loeb edition, Cambridge, Mass., and London 1974) 275-7, and G. Aujac, Denys D'Halicarnasse: Opuscules Rhétoriques Tome II (Budé edition, Paris 1988) 66, while no doubt suited to their respective purposes, convey little of that verbal cleverness and interpretative acuity. Nor does Nicolai's discussion (art. cit. 21) do more than register Dionysius' understanding of ktema as implying 'inscription/monument'.
In fact, Dionyius plays off Thucydidean vocabulary and syntax with great skill. Essentially, he inverts Thucydides' sentiment, in favour of the orator (who thus substitutes for 'the sophist' implicit in Thucydides) rather than the historian, privileging the activity of the former, while not absolutely rejecting that of the latter. The broad structure 'not only ..., as the historian ..., but ...' reworks, in positive form, Thucydides' ' "x" rather than "y" '. The superiority of the orator's activity is underpinned by systematic appropriation, and 'turning', of the vocabulary that Thucydides applied to his historiographical activity: arkountos ~ Thucydides' arkountos, but now 'the sufficient' has precise and immediate reference; ouk eis anathema kai ktema ... monon ~ Thucydides' 'inscriptional' ktema, but the orator's activity is more than a ktema; ho suggrapheus ~ Thucydides' xunegrapsen (1.1.1); chresin ~ Thucydides' parachrema (real 'advantage', not trivial 'gain'); to saphes ~ Thucydides' to saphes, but Dionysius substitutes a readily comprehensible notion of oratorical 'clarity' for Thucydides' actually rather murky notion of 'clearness'; tois enagoniois logois ~ Thucydides' agonisma, but again Dionysius gives a positive interpretation to a notion of 'competition' which in Thucydides is in part negative; dokein ~ Thucydides' 'rejection' of dokein in 1.22.2, but Dionysius redefines dokein in a positive way.
In sum, one might be tempted to say that Dionysius' response to Thucydides 1.22 is just as sophisticated and creative as those of Polybius, Livy and Arrian (in sections 11-13, cf. also n. [40] for another Dionysian aperçu), except that: (a) the point has already been made, at least potentially, by Nicolai; (b) Dionysius' quasi-'glossing' of ktema by anathema is in itself somewhat cruder than anything in the creative reinterpretations offered by those greater names; (c) many other aspects of Dionysius' response to Thucydides are, notoriously, vulnerable to rather severe criticism.
It is fair also to say that Marcellinus' (re)interpretation of Thucydides 1.22.4 has its own considerable merits (I hope the translation is right), but of course Marcellinus enjoys the considerable practical advantage of coming late in the tradition. [Return to text]
[18] John Marincola's important recent book, Authority and Tradition in Ancient Historiography (Cambridge 1997) has greatly advanced our understanding of this function of literary allusions in ancient historiography (which is of course not the only function: for a fruitful range of possibilities see e.g. T. Rood, 'Thucydides and his Predecessors', HISTOS 2 (1998), published April 1999; id., 'Thucydides' Persian Wars', in Kraus (n. [32]) 141-68; and G. Howie, 'The aristeia of Brasidas: Thucydides' presentation of events at Pylos and Amphipolis' (revised title), forthcoming on HISTOS). It is of course a matter of scholarly debate what Thucydides' 'field' is: I would myself define it very broadly as the understanding, and acquisition, of political and military competence through the case studies provided by the Peloponnesian War (cf. section 11 and n. [72]). 'Field' here is something different from genre (on which see section 14 and n. [79]). On the debate concerning the 'helpfulness'/'usefulness' of Thucydides' History see n. [72]. [Return to text]
[19] Documentation and discussion in Moles, 'A False Dilemma' (n. [1]). For the particular case of Herodotus see n. [33]; for Hesiod see n. [35]; for another implicit competitor see n. [29]. [Return to text]
[20] LSJ s.v. Although Thucydides is the earliest example cited, it is unthinkable that subsequent examples derived from him. [Return to text]
[21] As Woodman (n. [11]) 8-9, rightly insists, though this insistence (like many of Woodman's correct 'insistences') continues to be widely ignored. [Return to text]
[22] See my papers 'A False Dilemma' (n. [1]) and 'Narrative Problems in Thucydides Book I' (n. [7]). [Return to text]
[23] The authenticity of the letter, as also of much (or even all) of Pausanias' medising, is naturally controversial; at least Thucydides is making a good fist at authentic phraseology, which is all that matters here: Hornblower (n. [8]) 216 (with bibliography); see also my discussion in the main text; on the other hand, as we shall see, this Persian phraseology is also being passed through a Greek filter. [Return to text]
[24] Crane (n. [7]) 13-14. [Return to text]
[25] See Hornblower in n. [23] and R. Meiggs and D. Lewis, A Selection of Greek Historical Inscriptions to the End of the Fifth Century B.C. (Oxford 1969), 20-2 (on the Letter of Darius; I discuss this latter inscription in the immediately subsequent main text). [Return to text]
[26] Presumably it might be possible also to give 1.129.3 some metaliterary force, more complex than, but along the same lines as, Croesus' reference in Herodotus 1.207 to apodexis ergon megalon. [Return to text]
[27] This notion of its being 'free' (relatively: some will have to buy their own copy) is reinforced by the contrast with the 'prize competition' (section 3) and with sophistic venality (section 9 and n. [61]). Note that in itself Thucydides' phrasing here gives little comfort to those (including Moles n. [33] 106) who suppose that he is writing for a select reading public: the implication, rather, is that more or less everyone is literate but they can choose whether or not to read Thucydides. Of course, those who might be in a position to make this choice might already be a select group, and of course also the extent and degree of literacy in fifth-century Athens continue to be highly controversial matters: see most recently T.J. Morgan, 'Literate education in classical Athens', CQ 49 (1999) 46-61. Nevertheless, for what it is worth, I feel that in general the many discussions of Athenian literacy take altogether insufficient account of the sheer fact of the fifth-century appearance of such huge texts as Herodotus and Thucydides.
Thucydides' representation of his work as an inscription might also seem at first sight consistent with the hypothesis that he constructed his work as a mnemonic text, a hypothesis advanced by G. Shrimpton, History and Memory in Ancient Greece (Montreal/London 1997) 191-98, but this hypothesis involves misinterpretation of key elements of 1.22 (as I have argued in 'A False Dilemma' [n. [1]], and fails also to take proper account: (a) of the History as a reading text (section 7); and (b) of the extent to which this reading text demands that the reader involve himself in (inter-)active interpretation (and not, therefore, in passive memorisation). [Return to text]
[28] For this idea see e.g. (and rather diversely) E. Havelock, A Preface to Plato (New Haven 1964) 53-4; V. Hunter, Past and Process in Herodotus and Thucydides (1982) 287-96; Svenbro (n. [16]) 148 n. 10; Crane (n. [7]) xi-xiii; 1-26; dissent of the 'it ain't necessarily so' variety in Hornblower (n. [8]) 60-1 (hypothesising selective Thucydidean recitation); Marincola (n. [18]) 21. [Return to text]
[29] Note the strong contrast between 'hearing' and 'looking' (cf. also n. [5]): the History is a sort of 'seeing text', with multiple implications: see my 'A False Dilemma' (n. [1]). This conception itself suggests yet another rival to Thucydides' claims in ch. 22: the visual arts. [Return to text]
[30] E.g. Thomas (n. [5]) 89-90; ead., Literacy and Orality in Ancient Greece (Cambridge 1992) 14; S. West, 'Herodotus' epigraphical interests', CQ 35 (1985) 278-305; the perspective of G. Crane (e.g. as cited in n. [28] and quoted in section 5) seems to me nearer the mark. [Return to text]
[31] The translations are my own, 'tweaked' for local purposes. Interpretation of Hecataeus' hos-clauses is controversial, but in the present context this does not matter. The suggestion sometimes made that the words 'Hecataeus the Milesian speaks thus' are not Hecataeus' but the quoter's founders: (a) on the parallels with Herodotus and Thuycides (and others, including Antiochus); (b) on the epistolary and inscriptional parallels; (c) on the Homeric quality of mutheitai. (I am aware there are circles here, but some circles are better, and more mutually-reinforcing, than others!) [Return to text]
[32] E.g.: Conz, Museum für griech. und rÖm. Liter. 2 127 (who Conz? non vidi; Conz is cited by Creuzer in the next reference); F. Creuzer, Deutsche Schriften, neue und verbesserte III.2 (Leipzig and Darmstadt 1847) 642 (1789 original); H. Diels, 'Herodot und Hekataios', Hermes 22 (1887) 118 n. 1; J.L. Myres, Herodotus: Father of History (Oxford 1952) 82; Svenbro (n. [15]) 149-50; A. Corcella, «Ecateo di Mileto cosÌ dice», Quaderni di storia 43 (1996) 295-301 (disciplined discussion, with useful bibliography); L. Porciani, La Forma Proemiale: Storiografia E Pubblico Nel Mondo Antico (Pisa 1997), esp. 5-6; 27-33; 34-37; 70-78; 160-62 (diffuse and repetitive but with masses of information); T.J. Luce, The Greek Historians (London 1997) 26; H. Sancisi-Weerdenburg, 'The Persian Kings and History', in C.S. Kraus (ed.), The Limits of Historiography (Leiden/Boston/Köln 1999), 91-112, does not consider the question of the relationship between Persian historiography and Greek; 'for always' in inscriptions: e.g. Meiggs-Lewis (n. [24]) nos. 10, 47, 63 and 64; use of keimai occurs in funerary inscriptions ('here lies so-and-so'), but given Thucydides' formal suppression of himself (in 1.22.4) in favour of his work, which keitai, this resonance is not necessarily inapposite; Herodotus' familiarity: extensive documentation in Corcella, art. cit., 297; Thucydides' familiarity: Xerxes' alleged letter to Pausanias (section 5). No single discussion offers the sort of cumulative case that I put together in the main text, even though it should be obvious that only a cumulative case has any chance of carrying conviction. [Return to text]
[33] Herodotus ~ Hecataeus: see e.g. F. Jacoby, 'Die Entwicklung der griechischen Historiographie', Abhandlungen zur griechischen Geschichtsschreibung (Leiden 1956) 16-64 at 37-9 = Klio 9 (1909) 80-123 at 99-100; H. Pelliccia, 'Sappho 16, Gorgias' Helen, and the preface to Herodotus' Histories', YCS 29 (1992) 63-84 at 74-5; C. Pelling, 'Epilogue', in Kraus (n. [32]) 325-57 at 331-32 (stimulating observations with useful bibliography); Thucydides ~ Herodotus: fairly full discussion and documentation (missing, however, some tricks) in my paper, 'Truth and Untruth in Herodotus and Thucydides', in C. Gill and T.P. Wiseman (edd.), Lies and Fiction (Exeter 1993), 88-121; also the useful list of parallels, real or alleged, in S. Hornblower, A Commentary on Thucydides: Volume II: Books IV-V.24 (Oxford 1996) 19-38 (likewise missing important tricks) and Hornblower's extended (but highly restricted) discussion, 'Thucydides' use of Herodotus' (op. cit. 122-45). [Return to text]
[34] So, for example, M. Pohlenz, 'Thukydidesstudien', Nachr. Ges. Wiss. zu GÖttingen, 1920, 74f.; Moles (n. [33]) 106; Shrimpton (n. [27]) 170; the fact that Thucydides' wording has internal logic (contrasting with his reconstruction of the speeches 'as it seemed to me') does not preclude this additional resonance, which is thoroughly compatible with the density of Thucydides' literary allusions in ch. 22; Gomme's objections to Pohlenz (n. [10] 141) seem to me pedantically misconceived. [Return to text]
[35] In 21.2, 22.1 and 22.4 Thucydides evokes the Hesiodic play on aiei/aeidein: Moles, 'Truth and Untruth' (n. [33]) 109f. Hesiod should also be seen as an influence upon Thucydides' complex analysis of causality in Book 1: Moles, 'Narrative Problems in Thucydides Book I' (n. [7]). Svenbro (n. [16] 150), quoted below in this section, notes the parallel with Hesiod as regards the prefaces. M.L. West, The East Face of Helicon (Oxford 1997) 287, adduces two Babylonian parallels for divine revelation of a poem, one of which mentions a 'tablet', so if this sort of 'parallel' was real to Hesiod, he might have been suggesting the written quality of his own poem; to me, however, the parallels look distinctly strained. (One can think this while still supposing that Hesiod himself did use writing.) [Return to text]
[36] E.g. (a): G. Nagy, 'Herodotus the Logios', Arethusa 20 (1987) 175-8; (b): D. Lateiner, The Historical Method of Herodotus (Toronto 1989) 9 (a notably pregnant formulaton); (c): Pelliccia (n. [33]) 80-4, adducing Gorgias, Helen 5; this parallel seems to me certain in itself, but Pelliccia formally leaves open the question of relative priority, though in n. 45 he teases us with the possibility of Gorgianic priority; however, alike on general grounds (known literary activity); because the arrangement of material has richer interpretative implications in Herodotus (e.g. in making the key parallel with Odysseus [n. [59]); because imitation of Herodotus does more for Gorgias (imparting a pleasingly specious 'historiographical' tone) than imitation of Gorgias would for Herodotus, I think Herodotus must be prior, in which case the Gorgianic parallel, though interesting, has no interpretative consequences for Herodotus; also R.L. Fowler, 'Herodotus and his Contemporaries', JHS 116 (1996) 62-87 at 86-7, adducing Theognis 769-72 and Plato, Protagoras 320b. [Return to text]
[37] Svenbro (n. [16] 150). [Return to text]
[38] Corcella (n. [32]). [Return to text]
[39] Pelliccia (n. [33]) 75, followed by J. Moles, 'Herodotus warns the Athenians', PLLS 9 (1996) 259-84 at 282 n. 36.
Franco Basso will shortly be publishing a paper (destination as yet unknown) which will thoroughly examine exitela and its possible implications and will indeed offer what seems to be a distinctively new interpretation. In what follows I am particularly grateful to discussion with Franco Basso. [Return to text]
[40] Note Pelling (n. [33]) 334, 'things originating from humans' (with exegesis in n. 32), though Pelling might not regard this as 'biological'; 'biologically'/exitela ~ telos: Moles (n. [39]) 282 n. 36. Franco Basso alerts me to Dionysius To Pompeius 2, 'the very proem is both beginning and end [telos] of the History', which suggests that the estimable Dionysius (n. [17]) here anticipated Moles. [Return to text]
[41] Documentation in Basso (n. [39]). [Return to text]
[42] Luce (n. [32]). The gloss is my own, partly founded on a reported conversation between Jim Luce and Tony Woodman. [Return to text]
[43] Cf. T.C.W. Stinton, CQ 25 (1975) 226: 'it is not a question of hamartia having different senses ... hamartia is a word which covers a range of situations, and will be interpreted in each case according to the situation to which it is applied' (a crucial point about language and its interpretation and one which has enormous general application). [Return to text]
[44] LSJ s.v.; Pollux 1.44 (re dye). [Return to text]
[45] E.g. C.D. 57.16.2 (an example which shows that exitelos can be used in contrast to (altogether) 'dead', 'destroyed'. [Return to text]
[46] E.g. Paus. 10.38.9. [Return to text]
[47] Meiggs-Lewis (n. [25]) 20; A.W. Gomme, A. Andrewes and K.J. Dover, A Historical Commentary on Thucydides: Volume IV Books V 25-VII (Oxford 1970) 331 (less trenchant). [Return to text]
[48] Pollux 5.150. [Return to text]
[49] LSJ s.v.; P. Chantraine, La Formation Des Noms En Grec Ancien (Paris 1933) 241. [Return to text]
[50] Herodotus' general 'openness' can be (and often is) exaggerated; certainly at the beginning of his work he is fairly (albeit also ambiguously and ironically) authoritative: after undertaking to investigate the 'cause' of Greco-barbarian warring, he gives other accounts, which if not necessarily untrue are at least relatively trivial and unverifiable, before commencing his narrative proper with 'the man I know to have begun unjust deeds to the Greeks' (1.5.3). [Return to text]
[51] The traditional view, e.g.: Gomme (n. [10]) 148 (qualified); Hornblower (n. 10]) 61; Moles (n. [33]) 106; contra e.g. Marincola (n. [18]) 21. [Return to text]
[52] See n. [33]. [Return to text]
[53] Critical discussion (a little too critical for my taste) in S. Flory, 'Who Read Herodotus' Histories?', AJP 101 (1980) 12-28 at 14-15. [Return to text]
[54] O. Murray, 'Herodotus and Oral History', in H. Sancisi-Weerdenburg and A. Kuhrt (edd.), Achaemenid History: Proceedings of the Groningen Achaemenid History Workship II. The Greek Sources (Leiden 1987) 93-115; J.A.S. Evans, Herodotus, Explorer of the Past (Princeton 1991) 90, 100; R. Thomas, Literacy and Orality in Ancient Greece (Cambridge 1992) 125-6; id., 'Performance and Written Publication in Herodotus and the Sophistic Generation', in W. Kullmann and J. Althoff (edd.), Vermittlung und Tradierung von Wissen in der griechischen Kultur (Tübingen 1993) 225-44. [Return to text]
[55] E.g. J.C. Kamerbeek, The Plays of Sophocles. Part III. The Antigone (Leiden 1978) 158-60; A. Brown, Sophocles: Antigone (Warminster 1987) 199-200; H. Lloyd-Jones and N.G. Wilson, Sophoclea; Studies in the Text of Sophocles (Oxford 1990) 138. I accept authenticity and find the arguments of the 'antis' rather crude. [Return to text]
[56] Evans (n. [35]) 100; Moles (n. [33]) 119 (independently). [Return to text]
[57] Arion: Moles (n. [39]) 281 n. 26 (accepting a suggestion of Alan Griffiths); Solon: Moles, art. cit.; C. Pelling (n. [33]) 332 (with bibliography); Anacharsis clearly comes into the same general category as the other two and Herodotus' language in relation to him is surely suggestive: 4.76.2 gen pollen theoresas kai apodexamenos kat' auten sophien pollen. [Return to text]
[58] Moles (n. [39]) 262-65. In the translation, the inverted commas enclosing the word "sophists" register the fact that the Greek word is sophistai, ordinarily here translated as 'wise men' (vel sim.) but already in the fifth century capable of the resonance 'sophists' (which has of course to be justified in context: see my arguments). [Return to text]
[59] Moles (n. [32]) 96 and (n. [39]) 264-6; Marincola, 'Odysseus and the Historians', HISTOS 1 (1997) (October 1996). [Return to text]
[60] Hornblower (n. [8]) 62, cl. Thuc. 3.38.3-4 (a passage which in fact has metaliterary implications for the general interpretation of 1.22). [Return to text]
[61] Cf. nn. [9] and [27]. [Return to text]
[62] Moles (n. [39]) 277-79. [Return to text]
[63] And was indeed forcefully urged as insuperable by Franco Basso in conversation with me at the Leeds Conference. [Return to text]
[64] I accept the traditional modern dating and believe that there are good reasons for rejecting later datings: Moles (n. [39]) 276. [Return to text]
[65] I am aware that they have been disputed, e.g. by those modern scholars (including so distinguished a Herodotean as Stephanie West) who continue to believe that the ending of Herodotus is incomplete; for a partial disclaimer of the Pericles-Cleon parallel/contrast see P.J. Rhodes, ' "Epidamnus is a City": On Not Overinterpreting Thucydides' (n. [3]), at the very end. [Return to text]
[66] Beginning and end of Herodotus: Moles (n. [39]) 271-77 (with bibliography at 283 n. 45); C. Dewald, 'Wanton Kings, Pickled Heroes, and Gnomic Founding Fathers: Strategies of Meaning at the end of Herodotus' Histories', in D.H. Roberts, F.M. Dunn and D. Fowler, edd., Classical Closure: Reading the End in Greek and Latin Literature (Princeton 1997) 62-82. [Return to text]
[67] The apparent paradox that the editor of an Internet journal should hold this conviction is only apparent: readers of HISTOS need only to download to enjoy most of the advantages of conventional publication (nevertheless, HISTOS will soon go 'hard-copy'). Some of the issues raised for classicists by the Internet are explored by Craige Champion's review of J.J. O'Donnell, Avatars of the Word: From Papyrus to Cyberspace, HISTOS 3 (1999). Excellent as that review is, I think that it does not fully register the extreme radicalness of the threats to learning and understanding - even to what it is to be an intelligent and fully sentient human being - posed by the Internet. As literary classicists, we should never abandon our belief in the supreme intellectual value of reading great texts, an activity to which all the paraphernalia of the resources of the Internet, great and ever-increasing as they are, should always remain strictly ancillary. [Return to text]
[68] I thank Stephen Wheeler for drawing my attention to this passage. [Return to text]
[69] One manifestation of Polybius' elaborate development of Thucydidean causality (1.23.5-6). [Return to text]
[70] An echo of Thucydides 1.22.2 (quoted in section 1). [Return to text]
[71] F.W. Walbank, A Historical Commentary on Polybius: Volume I Commentary on Books I-VI (Oxford 1957) 358-61; id., 'Symploke: its Role in Polybius' Histories', YCS 24 (1975) 197-212; K. Sacks, Polybius on the Writing of History (Berkeley 1981) 110-14; S.M. Wheeler, 'Time in Ovid's Metamorphoses and Ancient Historiography', paper delivered at the Durham Conference on 'Augustan Poetry and the Traditions of Ancient Historiography' on 1st September, 1999. For Polybian symploke cf. also Polyb. 1.3.4 and 1.4.11; Diodorus 4.60 and 4.63, with K. Clarke, 'Universal Perspectives in Historiography', in Kraus (n. [32]) 249-79 at 265-76; and (so Wheeler) Ovid, Metamorphoses 1.1.4 perpetuum deducite ... carmen. Incidentally, given that the 'actions' of 3.22.2 are presumably the 'threads' of the 'weaving' of 'universal' history, does 'follow along clearly' image Polybius' History as (also) a labyrinth, negotiable by (as it were) 'following Ariadne's thread'? For history as a labyrinth see M. Jaeger, 'Guiding Metaphor and Narrative Point of View in Livy's Ab Urbe Condita', in Kraus (n. [32]) 169-195. [Return to text]
[72] So also e.g. G.E.M. De Ste. Croix, The Origins of the Peloponnesian War (London 1972) 29-33 (with both substantial arguments of his own and final appeal to Lucian); contra e.g. Marincola (n. [18]) 21. I hope to have validated this interpretation in 'A False Dilemma' (n. [1]). [Return to text]
[73] Moles, 'Livy's Preface', PCPS 39 (1993) 154. [Return to text]
[74] A.B. Bosworth, A Historical Commentary on Arrian's History of Alexander (Oxford 1980) 104ff.; J.L. Moles, 'The Interpretation of the "Second Preface" in Arrian's Anabasis', JHS 105 (1985) 162-68; J. Marincola, 'Some Suggestions on the Proem and "Second Preface" of Arrian's Anabasis', JHS 109 (1989) 186-9; V. Gray, 'The Moral Interpretation of Arrian's Anabasis', JHS 110 (1990) 180-6; Marincola (n. [18]) 253-4. [Return to text]
[75] Translation by Moles (n. [73]) 163 (slightly adapted). [Return to text]
[76] Besides the references in n. [73], Marincola (n. [18]) 271-5 = Appendix II discusses the historiographical convention of naming (or not-naming oneself); the 'one reading' is Moles (n. [74] 164 with earlier bibliography in n. 13), followed by Marincola (n. [18] 146 n. 80); 'another' is Gray (n. [74]). [Return to text]
[77] Moles (n. [74]) 166 (as we shall see, wrongly citing anagrapho in Praef. 1 as an example of the 'colourless' usage). [Return to text]
[78] Marincola (n. [18]) 275. [Return to text]
[79] The questions of beginnings and genre (along with other questions) are treated in two excellent contributions to Kraus (n. [32]), John Marincola's 'Genre, Convention and Innovation in Greco-Roman Historiography' (op. cit. 281-324) and the already much-cited Christopher Pelling's 'Epilogue' (op. cit. 325-60). From the point of view of this paper, however, their impressive insights need to be supplemented by recognition: (a) of the influence of Persia and Persian (in some sense) historiographical modes; (b) of the impact of writing and reading and the immense interpretative possibilities that they offer; (c) of the impact of the physical inscription as a metaphor for historical writing.
I say 'post-Thucydides' because of the several writers ('historians') who were in some sense continuators of Thucydides; 'post-Herodotus' because Thucydides was in some obvious senses a continuator of Herodotus: 'post-Hecataeus' because Herodotus was in some obvious senses a continuator of Hecataeus; it is also worth recalling that on one occasion (7.96.1) Herodotus comes very close to using historie of a distinctive form of intellectual activity, which we might well call 'history', even though Thucydides avoided the use of historia to describe his own activity, because it would make him look too like Herodotus (Hornblower [n. [8]] 1). [Return to text]
[80] I suspect that Herodotus' (certainly to some extent) ambiguous and ironic appeal to the authority of Persian and Phoenician logioi for previous accounts of Greek-barbarian enmity (1.1.1; 5.3), logioi in whose existence many modern Herodoteans (including myself) devoutly disbelieve, cocks a snook at Hecataeus' implicitly 'Persian' stance; for nuanced formulations of the well-aired hypothesis that 1.1.1-5.3 alludes to/targets Hecataean material (or at least Hecataeus' manner) see Pelliccia and Pelling (both cited in n. [33]). I note in passing that any serious attempt to engage with 'the liar school of Herodotus' must factor in Herodotus' irony, ambiguity, humour and sheer sense of fun. [Return to text]
Back to HISTOS: Contents of the current volume
Back to HISTOS Main Menu
To Durham Classics Department Home Page