This is the first of two papers on Dionysius to be published on HISTOS. The papers are free-standing but complementary: the second, entitled ‘From muthos to historia in Dionysius of Halicarnassus’, presupposes the conclusions of the first and may be regarded as a sustained practical demonstration of some of those conclusions.

The present paper is divided into the following sections:
1 Dionysius’ programme

2 The questions of scope and periodisation raised in 1.8.1
2.1 Larger perspectives: periodisation problematised

3 The feasibility of a ‘Roman Archaeology’ and Dionysius’ qualifications for writing it
3.1 Dionysius’ claim to exhaustive and meticulous reading
3.2 Sources cited by Dionysius

4 Dionysius’ citation procedure: an overview

5 The trustworthiness of the claim to exhaustive and meticulous reading
5.1 Arguments against Dionysius’ good faith and reliability
5.2 Arguments in favour of Dionysius’ good faith and reliability
5.3 Dionysius’ good faith vindicated

6 The difference of practice between Books 1-4 (especially Book 1) and the rest of the work

7 Dionysius’ creative engagement with his sources
7.1 Dionysius’ assessment of earlier authorities
7.2 Dionysius’ distinctions between muthikon and historikon
7.3 Dionysius’ criteria for the evaluation of evidence

8 Conclusions
(this section also functions as an abstract).

Any attempt to understand an ancient historian’s programme, claim to authority, self-definition,^[2] originality and ideas about history and historiography must begin with analysis of his prefatory statements.^[3] Dionysius’ preface (whose literary and intellectual quality has generally been underestimated) reveals how his authority rests at once upon his predecessors and upon himself. At the very outset, in a single long and impressive sentence, he marks his knowledge of, and simultaneously his distance from, those predecessors; he expresses his attitude to his role and his materials; and he asserts the logismoi (‘reasonings’) and empeiria (‘knowledge’) which underpin his work:

To render [apodidosthai] the accounts [logoi] customary in the prefaces of histories is not at all to my wish, yet I am forced to make a preliminary statement about myself. I do not intend to spin out my own praises, which I know would clearly seem burdensome to readers, nor am I deliberately making charges against other historians, as did Anaximenes and Theopompus in the prefaces of their histories. Rather, I am demonstrating my reasonings [logismoi], by which I was motivated when I started out [hormesa] upon this study, and am rendering [apodidous] an account [logos] of the starting-points [aphormon] from which I acquired the knowledge [empeiria] of what I intend to write. (1.1.1)

While in some ways conventional, Dionysius’ initial prefatory sentence also subverts established historiographical conventions (as he himself states). Unwilling to speak of his own person (to the point of not even registering his identity), he is yet in a position where by convention he has to do so; disavowing criticism, he nevertheless criticises. The effect of naming and blaming Anaximenes and Theopompus^[4] for their negative and critical approach is twofold: he brings to the fore source questions and invites comparison with some distinguished predecessors; he also formally avoids praise and blame,^[5] yet implicitly engages in the latter. This ambiguity anticipates his later practice: while maintaining an overtly positive disposition towards his predecessors and their work, he nevertheless on occasion expresses negative judgements (cf. section 7.1). Overall, the formal implication of the sentence is to play down his own role as historian^[6] and stress the objectivity of his own treatment. Starting out upon his work, the historian is at the inception of a journey through the text; although the basic conception of the text as a journey goes back to the very beginnings of ancient historiography, the vocabulary here (hormesa = ‘I started out’; aphormon = ‘the starting-points’) specifically recalls Polybius,^[7] another major predecessor of Dionysius and the one with whose views he will necessarily be most engaged in the preface. The effect of the verbal interplay between apodidosthai ... logous (the conventional historiographical ‘rendering of accounts’) and apodidous logon (his own seriously-motivated ‘rendering of an account’)^[8] is through rejection of the former to emphasise the latter, the subject-matter and its justification.

Dionysius proceeds to enlarge upon the qualifications for the historian: having first chosen noble and useful subject-matter (a choice which aligns Dionysius with yet another great predecessor, Herodotus),^[9] he must ...

... then provide himself with the necessary starting-points for the writing-up of the subject with much care and pains. (1.1.2)^[10]

By contrast, the choice of unworthy themes gives rise to the suspicion that the desire for mere fame is motivating the writer; on the other hand, if great themes are chosen, but there is inadequate care in the collection and composition of the material, the treatment is unworthy of the subject matter (1.1.3-4). These are general observations, not ad hominem ones; they extend the criticisms levelled against self-praise in 1.1.1, while of course implying that Dionysius himself will not fall short on either of these counts.

Next comes a justification of his choice of subject: 1.2-3 deals with the greatness of Roman power, and 1.4 with the particular importance of the early period,^[11] about which Greeks are either completely ignorant or else seriously misinformed. Dionysius therefore pledges that he will set right these misconceptions: he will show that the founders of Rome were Greeks and will show also how Roman deeds and institutions from the time of the foundation onwards explain their current great hegemony. This is the first formulation of the Romans-as-Greeks theme which characterises the whole work.^[12] The author will show the Romans’ origins in ‘this book [graphe]’, and from there will lead off (the journey metaphor is maintained) about their doings — by implication, in the remainder of the work:^[13]

I propose therefore to remove these (as I said) errant [peplanemenous]^[14] assumptions from the minds of the many and to substitute true ones. On the one hand, concerning the founders of the city: who they were, at what times the respective groups assembled, and what were the fortunes which made them leave their native foundations, I shall show in this book. Through this I undertake to demonstrate [epideixein] that they were Greeks and that it was not from the least or meanest nations that they assembled. On the other hand, concerning the deeds which they demonstrated [apedeixanto] immediately after the foundation, and the customs by which those after them reached so great a dominion: beginning with the book after this one I shall lead off with these, leaving out nothing (as far as my powers extend) that is worthy of history, so as to set before those who have learnt the truth an appropriate idea of this city ... (1.5.1-2)

The interplay between epideiknumi (of the role of the historian) and apodeiknumi (of the deeds of the historical agents) emphasises Dionysius’ positive, encomiastic, stance, as well as suggesting the ideal unity of theme and treatment to which he (like all ancient historians) aspires. The play also clearly recalls the similar play in Herodotus’ preface,^[15] though with elegant variation between epideiknumi and apodeiknumi: in this way Dionysius, without explicit self-advertisement, tacitly compares himself to the great Herodotus but suggests his own creative independence. Upon further examination, it becomes apparent that different modes of historical writing are promised here — what Polybius calls tropoi:

The genealogical mode attracts the man who loves a story; that concerning colonisations and foundations and kinships, the curious and particular; ... that concerning the deeds of nations and cities and dynasts, the politically aware (Pol. 9.1.4)

Polybius of course explicitly excludes from his own history

... that concerning genealogies and tales and colonisations, also kinships and foundations ... (Pol. 9.2.1)

Dionysius, on the other hand, intimates inclusiveness: ‘who the founders of the city were’ points at genealogy; when and why they came promises ktiseis (foundations); their doings and institutions indicate political and institutional history. In claiming by implication a wide appeal (a claim made explicitly later on at 1.8.3), he rejects the narrow exclusiveness of Polybius.

Dionysius goes on to say (1.6.1-2) that the early period has hitherto been inadequately handled by both Greek and Roman historians, seven of whom are named, these seven all being Greek-writing, an important point, as will soon become clear.^[16] This section thus introduces an important qualification of the initial, non-condemnatory stance: this stance has done its work of foregrounding the subject-matter and is now, in effect, discarded. All previous treatments have failed to achieve proper standards of accuracy and fullness; Dionysius will not only fill the gap but do so properly:

For these causes it seemed right to me not to pass (parelthein) by a fine history left aside uncommemorated by the older writers, from which, accurately written, the best and most just results will ensue ... (1.6.3).

Several elements here recall Thucydides’ justification (Thuc. 1.89) for his insertion of the Pentecontaetia narrative): the conception of history as space and of historiography as a journey; the idea of a specific period of history as a place within that encompassing historical space; the accusation that predecessors have treated the period only summarily and inaccurately and the counter-assertion that the present writer will remedy these deficiencies. The recall of these elements serves several ends: to add another great predecessor to the list of Dionysius’ influences; to suggest interesting parallels and contrasts between Thucydides’ project and his own (Thucydides commemorated the rise to power of the Athenians; Dionysius commemorates the rise to power of the Romans, who were Greek by origin but who became distinctively different, and whose world empire, incomparably greater than the Athenian empire, yet endures); and to hint at structural parallelisms between the two historians, a theme which Dionysius will later develop (section 2.1).

There follows an outline of the many benefits that will accrue from a proper treatment of this fine theme (1.6.3-5), which Dionysius is peculiarly qualified to handle (1.7, discussed in detail in section 3). Then, at the start of the final section of the preface, Dionysius specifies the chronological limits of his proposed work:

I begin my history, then, with the most ancient tales [muthon], which the writers before me left aside as difficult to be investigated without great study; and I bring my narrative down to the beginning of the Punic war ... (1.8.1)

Dionysius intends, evidently, by those parallel sentences (archomai men ... katabibazo de) not only to indicate the starting and stopping points of his textual journey (the former vaguely, the latter with considerable precision), but also to define his work in relation to those his predecessors. He is making, moreover, a large, though modestly expressed, claim to originality of subject matter: he will be the first to deal properly with the very difficult phenomenon of ‘the most ancient tales’, and he will conclude his diegesis (narrative) at (as most of his readers would know) Polybius’ starting-point. The implication is that he will provide the ample and satisfactory treatment of the early Roman history that has so far been lacking; he further implies that a non-contemporary topic is as valid as a contemporary one, ^[17] an implication the more challenging for his implicit invocations of Polybius and Thucydides. In this passage, the implications of the term muthoi are of course open to debate: my rendering ‘tales’ is intended as a reasonably non-prejudicial ‘holding’ translation. I shall explore the implications of the term for Dionysius in section 7.2 and in my second paper.

Dionysius then proceeds to define the work’s content: he will handle wars, staseis (instances of civil strife), constitutions, customs, laws, and ‘in short, the whole ... ancient life of the city’) in a way which combines variety, edification and entertainment.^[18] Only at the very end of the preface does he name himself: the impact of this delayed self-naming is all the greater for its reversal of the historiographical norm as established by such writers as Hecataeus, Herodotus and Antiochus, for the artful, if inevitably somewhat disingenuous, playing down of his own role that characterised the beginning of the preface, and for the progressively greater emphasis on his own role which now climaxes in this perfectly placed self-naming.^[19]

This preface, then, is a remarkably complex piece of writing, intricately structured, creative in its reworkings and reorderings of the prefatory conventions of ancient historiography, ingenious in its verbal patterning and literary allusion and rich in content as Dionysius positions himself within the historiographical tradition (both Greek and Roman), justifies his particular project and adumbrates the historiographical issues raised by it. I now proceed to discuss how some of these issues are embodied in practice in the work as a whole. Some parts of this discussion will naturally provide further evidence of the literary quality of the preface.

The passage which precisely defines the work’s scope (the already quoted 1.8.1: ‘I begin my history, then, with the most ancient tales, which the writers before me left aside as difficult to be investigated without great study; and I bring my narrative down to the beginning of the Punic war’) is the culmination of a series of carefully staggered indications in the preface that Dionysius’ subject is to be ta palaia (‘the ancient things’) of Rome. (The fact that readers would already know this from the title externally attached to the roll is of less significance than the impact of their experience of the text.) The first pointer comes in a sugkrisis (comparison) context where readers are reminded of Rome’s achievements from earliest days:

The time of her might is not short, but such as has belonged to no other single city or kingdom. Straight from the beginning after the foundation she was attracting to herself the nearby nations — numerous, warlike ones — and was always advancing, enslaving every successive antagonist. (1.3.3-4)

Dionysius goes on to say that 745 years have elapsed since that foundation, in the course of which Rome mastered Italy and aspired to ‘rule over all’: this is surely another conscious^[20] allusion to Polybius 1.1.5 (‘all things ... under one rule’). She achieved this so totally that

no longer having as antagonist any race either barbarian or Greek, she continues for the seventh generation now in my time ruling over every region ... (1.3.5)

From this alone it would not yet be clear that Dionysius intends to cover the earlier rather than the later phase of Rome’s achievements, though the persistence of the echoes of Polybius must suggest that Dionysius is emulating Polybius’ achievement without retreading the same ground. However, it is made explicit at the outset of the following chapter:

That it is not without reasoning and sensible forethought [pronoias emphronos] that I turned (etrapomen) to the ancient parts of her history ... (1.4.1)

Dionysius has led his reader up to this point with unobtrusive dexterity.^[21] He now anticipates — and proceeds to answer — criticism for the fact that

I have diverted [apeklina] to the archaiologia which holds nothing notable. (1.4.1)

This language of ‘turning ’ and of ‘diverting’ here again suggests that the historian is on a journey, even that on the journey through history the early period constitutes, from one point of view, a digression. But there is an important qualification: this is no mere frivolous diversion^[22] but has been undertaken deliberately and with forethought. The overtones of pronoia here seem to be Herodotean rather than Stoic.^[23]

Greeks, it transpires, are both ignorant and misinformed about early Roman history:

for still among the Greeks, almost all of them, the ancient history (he palaia historia) of the Romans’ city is unknown ... (1.4.2)

Dionysius will set them right by covering the founders and their deeds (1.5.1-2, quoted above). Predecessors had of course dealt with what is variously described as he archaiologia (‘the archaeology’), ta palaia (erga) (‘the ancient things (deeds)’) and ta archaia (‘the old things’) (1.6.1-2), but in far from satisfactory fashion. Here, then, Dionysius continues to qualify his initial seemingly favourable stance toward his predecessors; his present work, based on a wide range of oral and written sources, will, he implies, supersede these accounts (1.7). There follows (1.8.1) the definition of the work’s scope.

No reader who had got so far could be in any doubt but that Dionysius’ work covered ta palaia. Whether the author himself had assigned it the actual title He Rhomaike archaiologia (‘The Roman Archaeology’) is not strictly known, but, despite the notorious uncertainties of ancient book-titles,^[24] it is very likely that he had in fact done so. The only passage in the preface where the term archaiologia is applied to Dionysius’ undertaking is the already quoted 1.4.1: this seems intended to convey a general idea of the content. The expressions ta palaia, ta archaia and he archaiologia, as used in the preface, evidently apply to the entire work rather than to Book 1 alone or to some other subdivision of the work. These considerations suggest that Dionysius himself gave — or implied — the title by which we know the work today. Further support for this hypothesis is provided by the fact that, as is widely acknowledged, Josephus’ Jewish Antiquities (He Ioudaike archaiologia) is to some extent modelled on the work of Dionysius.^[25] At the very least, therefore, the title must have been generally applied to the work somewhat before the time of Josephus. If, indeed, Dionysius entitled his work ‘The Roman Archaeology’, there are implications for his view of the periodisation of history. (I shall return to this point below and in section 2.1.)

Given the enormous chronological scope of Dionysius’ work (‘the most ancient tales’ of 1.8.1 turn out in practice to mean the Oenotrian settlement of Italy described in 1.11), periodisation was of course essential. From the Oenotrians to the start of the first Punic war requires the coverage of over 1300 years and coverage, moreover, of such a nature as to validate Dionysius’ implied claim to write a history linked with that of Polybius. For Dionysius was writing to extend Polybius’ history back to encompass the earliest of the pre-foundation legends of the ancestors of the Romans and their Latin kin. The scope and economy of the work must now be briefly outlined.

In Book 1 Dionysius traces the (Greek) origins of the families and ethnic groups which had contributed to the Roman and Latin stock. He then describes the arrival of the Trojans, the foundation of Lavinium and Alba Longa, and the birth and recognition of Romulus and Remus. By the end of Book 1, Rome has been founded and Romulus has become its sole ruler; his institutions and military exploits comprise the first half of Book 2. The activities of the other six kings occupy the equivalent of two books (2.57 to 4.53). The last third of Book 4 narrates the expulsion of the Tarquins and the constitutional discussions leading to the establishment of the republic. By the end of Book 4 Dionysius has traversed one-fifth of his intended 20-book compass. He has in fact covered 244 years or 61 Olympiads (Ol. 7,1 to 68,1) in 3 books. Virtually the same number of years (243) remain down to the arche (‘beginning’) of the First Punic War in Ol. 128,4.^[26]

However, Olympiads and years are completely inappropriate to the content of Book 1. When the whole work is looked at in generational terms, a clear pattern appears: from the arrival of the Oenotrians in Italy to the fall of Troy is seventeen generations (1.11.2); from Troy’s fall and Aeneas’ flight to Italy down to Romulus’ founding of Rome is also seventeen generations (counted inclusively).^[27] All this is comprised in Book 1. There follow seven generations of kings plus a further seven (inferred) generations to the First Punic War.^[28] Such a period could not be considered historikon (‘historical’) in any uniform sense of the word, both by reason of its sheer size and, especially, because of the strongly muthikon (‘mythical’) quality of the start.

Now, not long before Dionysius’ time, Varro had grappled with the periodisation of prehistory — a problem which for him arose in the context of his antiquarian-anthropological researches. According to Varro, there were three epochs: after the adelon (‘the obscure’) and the mythikon (‘the mythical’), both of which are pre-historic, there came the historikon (‘historical’) epoch.^[29] The adelon is from the origins to Ogygus and the first flood; the mythikon, from Ogygus to the first Olympiad; then the historikon runs to the present.^[30] Although Dionysius knew and used Varro’s works (see sections 7.1 and 7.2), he plainly does not adopt this periodisation as it stood. The adelon is of course irrelevant to any historian, since it is defined precisely as that which is not available to enquiry. And the first Olympiad has no particular significance in Alban or Roman history. Yet Dionysius’ system clearly bears some resemblance to that of Varro. On Dionysius’ chronology of the regal period, with the foundation in Ol. 7,1 (752/1 B.C.), Romulus and Remus, about eighteen years old when their identity was revealed (1.79.12), were born at some time in the second Olympiad. Hence Dionysius might be said to be in the same chronological area as Varro with regard to the onset of the historikon,^[31] although we know that he had also composed his own chronological work where a consistent Roman dating system was established, complete with Olympiad equivalences.^[32] He is in any case working with a centrally important myth/history distinction inherited from Herodotus and Thucydides.^[33] It will therefore be necessary to ask whether Dionysius approaches the material provided by his sources in a different manner when he deals with muthoi from when he deals with historia. Where do the tales end and the history begin, and, in particular, what happens where they overlap? These are questions which I shall consider in section 7.2 of the present paper and in my second paper.

It is all too easy to underestimate the sophistication of Dionysius’ literary allusions and of his historiographical thinking. The periodisation outlined above is intrinsic to Dionysius’ historical thinking and to his organisation of material. But Dionysius deftly suggests much larger perspectives. The broad periodisation implied by characterising history down to the Punic Wars as archaiologia is interestingly complicated by the presence of 1.9-75 (the narrative of events from the first settlement of Rome to just before the foundation proper). Within the work as a whole, this section seems to constitute a kind of ‘mini-archaiologia’ which plays a role analogous to that of Thucydides’ archaiologia within his history^[34] (both ‘mini-archaeologies’ also occurring in Book 1 of their respective authors). To this structural parallel between Dionysius’ ‘mini-archaeology’ and Thucydides’ ‘archaeology’ must be added the parallel between Dionysius’ work as a whole and Thucydides’ Pentecontaetia narrative (1.6.3, discussed in section 1) and the notion that the work as a whole is a ‘digression’ (1.4.1). There are consequences of different kinds. These Thucydidean imitations and echoes ^[35] help to build up Dionysius’ composite historiographical persona. They also help to demarcate the less true from the historically true. Inasmuch as Thucydides’ archaiologia is explicitly (Thuc. 1.1.3; 1.20.1) on a lower truth level than his narrative proper, the narrative of 1.9-75 is an ‘archaeology’ in relation to the subsequent narrative, which is substantially historical. But on the other hand, Dionysius’ whole work, which is historical apart from the ‘mini-archaeology’ of 1.9-75, is also an ‘archaeology’ (and parallel to Thucydides’ Pentecontaetia retrospect): in relation to what? Obviously, on one level, in relation to the whole narrative of Roman history, which is (as it were) bigger than Dionysius’ project but of which that project is an indispensable part. (The effect is simultaneously to decrease and increase the importance of the project.) But also, inasmuch as Rome is represented as the supreme world power, the last and greatest of the succession of empires (1.2.1-4), and incomparably greater than any of the Greek powers, there is a sense in which Dionysius has, as it were, written the definitive ‘archaeology’ of that Universal History which is the synthesis of all previous histories.^[36] The shifting boundaries of the notion of ‘archaeology’ also suggest, unsettlingly, that while periodisation is an inevitable practical response to the problems of writing history (because of the evidence problem faced by any particular generation), it is always relative, not to say arbitrary.^[37] Those same shifts underline the fact that all periods of history are causally connected with all others, a thought all the easier in Greek historiography because of the perceived relationship between ‘beginning’ (archomai/arche) and ‘cause’ (arche/aitia).^[38] Within the periodisation that he actually adopts Dionysius, by ancient standards, is a rigorous and accurate chronologist (sections 2 and 7.3), yet his construction of this hierarchy of ‘archaeologies’ indicates also a rather profound historical thinker.

A historian’s justification of his project involves two things: justification of the project’s importance and justification of its feasibility. The former poses Dionysius no problem (1.1.2; 1.2-4), the latter is problematic in the extreme. The main difficulty is one of evidence. It emerges in the course of Book 1 that Dionysius, an admirer^[39] of Theopompus’ efforts at historical investigation (however much, as we have seen, he disapproved of the latter’s vindictive criticisms [1.1.1]), has himself carried out a certain amount of original research. For example, he sometimes attests to the survival of a monument or a custom on the basis of autopsy;^[40] and I have already mentioned his work on Roman chronology. But however active his antiquarian investigations and however original his chronological researches, the nature of Dionysius’ subject was bound to make him mostly an armchair historian, vulnerable, mutatis mutandis, to the famous criticisms of Timaeus made by Polybius, who, as we have repeatedly seen, was one of Dionysius’ own historiographical models and with whom the preface continually engages:

Having settled at Athens for almost fifty years and having access to the records [hupomnemasi] of his predecessors he assumed that he had the most important starting-points [aphormas] for history — in ignorance, as it seems to me at least. (Pol. 12.25d.1)^[41]

Consequently, claiming as he does to provide the work which will complement — even, in a sense, complete — that of Polybius, Dionysius has to meet the latter’s criticisms and justify the proposition that the work of the historian of the early period is as valid and as worthy of recognition as that of the historian who grapples with a contemporary (or near-contemporary) topic. In order to establish his own authority, he has to position himself and his graphe (writing) in relation to its aphormai (‘starting-points’)^[42] and to justify the particular ones that he has chosen:

I wish too to speak about the starting-points which I used when I intended to put my hand to this writing. (1.7.1)

His own authority is primarily based on, but also exceeds, that of his sources: he commands, and can deploy, the information provided by all of them, and consequently the fact that his account is fuller than those of supposedly authoritative predecessors should not give rise to doubts in the minds of his readers (yet again criticism of his predecessors is delicately implied). In order to prevent suspicions of invention

it is better to make a preliminary statement about both the accounts [logon] and the records [hupomnematismon] from which I started [hormethen] (1.7.1).

These statements about his ‘starting-points’ and the ‘records’ he used seem indeed directly to echo (and thereby directly to confront) Polybius’ criticisms of Timaeus.

The predecessors whose works his readers may already have encountered are Hieronymus, Timaeus, Polybius ‘or any of the other historians (suggrapheon) whom I just now mentioned as having slurred over their writing’ (1.7.1). The back-reference is to the critical allusions in 1.6.1, and the choice here of these three in particular is significant: Hieronymus as the first to allude to the Roman archaiologia; Timaeus as the best-known historian of the west, and Polybius as the historian of whom Dionysius will be, as it were, a pre-continuator.^[43] Dionysius’ long-term residence in Rome, his knowledge of Latin, and his sustained study particularly qualify him to treat an early Roman theme:

Having sailed across to Italy at the time when the civil war was resolved by Augustus Caesar in the middle of the 187th Olympiad, and from that time having spent in Rome twenty-two years until the present, learning the Roman language and acquiring knowledge of the local writings, I have in all this time been working continuously upon this subject (1.7.2)

Apart from the solid arguments here presented, this description seems to reverse the picture of Timaeus drawn by Polybius and thus to continue Dionysius’ intertextual debate with Polybius.

Dionysius has gathered information in two ways:

receiving some by teaching from the most learned men^[44] with whom I came into association, and some by a process of selection from the histories written by those praised by the Romans themselves: Porcius Cato, Fabius Maximus, Valerius Antias, Licinius Macer, the Aelii, Gellii and Calpurnii and many other not obscure men besides these; starting from those studies ... it was then that I put my hand to this book. (1.7.3)

Thus, as at 1.6.1-2, seven names are again cited, but this time of prominent Latin-writing authors.^[45] The preface has already revealed his thorough acquaintance with the Greek historians; moreover, the sequel will demonstrate his extensive knowledge of a wide range of other works. Dionysius here deftly demonstrates the bilingual (and in a sense bicultural) mastery which forms one of his unique qualifications for the task which he has set himself. Towards the end of Book 1 he reiterates his procedure:

These, then, are the things that my powers have sufficed to discover — by reading with great care many books of both Greeks and Romans — about the Romans’ origin. (1.89.1)

The claim to exhaustive (suchnas, cf. suchnoi at 1.7.3)^[46] and meticulous reading is thus the most important single element of Dionysius’ claim to authority.^[47] From this claim two questions arise (treated respectively in sections 4 and 5): in what ways does this allegedly exhaustive and meticulous reading manifest itself in Dionysius’ deployment of his source material? And is the claim to exhaustive and meticulous reading to be trusted?

Broadly speaking, the first question bears on Dionysius’ literary and rhetorical procedures, the second on his integrity and reliability as a historical researcher. But the two questions obviously overlap: Dionysius’ rhetorical case will be the stronger the more credible he appears as a historical researcher, and, if he actually is a reliable and conscientious historical researcher, the case will be at its very strongest; conversely, some of Dionysius’ literary procedures may provide real, as opposed to merely specious, support for his credentials as a reliable historical researcher. The question of whether Dionysius actually is such a researcher ultimately matters most to modern historians, who are concerned either to use Dionysius himself as a historical source or to use him as a source for other sources or as one element in the construction of source relationships. It should perhaps also be said that if our answer to the second question were that some (or many) of Dionysius’ citations were tralatician, this would not utterly destroy his credibility as a historical researcher but obviously it would weaken it and it would also impugn his personal integrity. It should also be recognised that, as with modern scholars, between the polarities of total reading by oneself and reliance on intermediaries (compendia, epitomes, diligent slaves and the like) there are many shades of grey.

Nevertheless, the analysis will be cleaner if the two questions are first kept separate and this way of proceeding will also help to highlight Dionysius’ literary procedures. Since the sources cited are so numerous, it is useful here to append a list.

In the whole of the surviving text of the AR, Dionysius mentions over fifty writers (here arranged by category, with F. Jacoby’s FGH numbers for the Greek historians):

Greek poets: Homer, Arctinus, Aeschylus, Sophocles;

Greek philosophers: Aristotle, Theophrastus;

Greek historical writers: Herodotus, Thucydides, Polybius, Pherecydes (FGH 3), Hellanicus (4), Damastes (5), Satyrus (20), Hegesianax = Cephalon (45), Anaximenes (72), Theopompus (115), Hieronymus (154) Silenus (175) Pyrrhus (229) Xenagoras (240), Eratosthenes (241), Ariaethus (316), Phanodemus (325), Hegesippus (391), Domitius Callistratus (433), Myrsilus (477), Antiochus of Syracuse (555), Philistus (556), Callias (564), Timaeus (566), Proxenus (703), Xanthus (765), Menecrates (769), Antigonus (816), Zenodotus (821), Dionysius of Chalcis (840 F 10), Demagoras (840 F 22), Agathyllus (840 F 22, 27).

Roman poets: Euxenus (? = Ennius);

Roman antiquarian and specialist writers: Fabius Maximus Servilianus, L. Mallius, Varro; historical writers: Fabius Pictor, Cincius Alimentus, Cato, C. Acilius, Calpurnius Piso, Sempronius Tuditanus, Gellius, Vennonius, Licinius Macer, Valerius Antias, Aelius Tubero.

These named sources, together with many anonymous references and variants, provide a convenient starting-point for an examination of Dionysius’ procedures in citing sources.^[48] When these citations are considered in context, a number of points emerge:

(i) the sheer number of writers cited (wholly contrary to the norm in ancient historiography), a factor which obviously boosts the credibility of Dionysius’ narrative and specific historical claims.

(ii) the difference between Books 1-4 and the rest of the work; Dionysius’ practice shows a clear contrast between the part covering the Republic (Book 5 onwards), where very few authors are cited or variants alluded to, and the earlier books on the origins and the monarchy.

(iii) the especially dense character of Book 1. Even within Books 1-4, Book 1 stands out for the number and variety of sources cited.^[49] Excluding those merely named and discussed in a general way in the prefatory chapters 1.1-8, over forty writers are cited for specific items of information in Book 1, and several are mentioned more than once. In addition, there are numerous vague references to ‘others...’, ‘many...’ etc., who are not identified further.^[50]

(iv) the variety of sources in Book 1. In support of his contention that Rome is by origin a Greek city, Dionysius cites epic and dramatic poets; chronological, mythological and genealogical writers; philosophers; and many historians; the reader is overwhelmed by the wide range of the evidence.

(v) their range from the very well known (whose works will have been readily accessible and who have natural persuasive power) to the quite obscure.

(vi) the obscurity of many of these authors. Dionysius is the only surviving writer of his generation (or, indeed, of the first centuries BC or AD) to cite certain of these authors, whose works are to us otherwise known only from references in scholia, lexica, and similar compilations. Although they must have been better known to educated readers of Dionysius’ own day, they will still have been relatively obscure: hence the very citation of such recherché material helps to bolster the impression of Dionysius’ erudition.

(vii) Dionysius’ practice, in Book I, of direct quotation (rare in historical writing),^[51] a device which he employs with great deliberation and emphasis. Significantly, he generally avoids it elsewhere. But Book I shows Dionysius in his most antiquarian mode; moreover, he is refreshingly unconcerned with the divergences in style which result and which are usually thought to have deterred self-conscious stylists from employing the device. A very striking instance is discussed below.

(viii) seeming accuracy of quotation, strikingly illustrated by the Ionic forms which appear in 1.48.3, from Menecrates of Xanthus.^[52]

(ix) paraphrase of sources (as opposed to more or less verbatim quotation)

(x) the impression given of constant evaluation of sources by intelligent criteria (a topic I shall examine separately in sections 7.1-3).

To the cumulative impression of this rich and varied source picture we may perhaps add the persuasive effect of Dionysius’ original research involving autopsy of monuments and customs.

The sheer volume of these citations, coupled with the obscurity of many of them, has naturally led scholars to pose the same question of Dionysius as of other seemingly polymathic writers such as Plutarch: did he consult all his sources at first hand or did he sometimes use short-cuts (‘inherited’ citations, compendia, epitomes, diligent slaves and so forth)? While the great Jacoby was sceptical of Dionysius’ credentials in this regard, modern scholars have generally been more ready to accept Dionysius’ good faith,^[53] with, as I think, good reason, as I shall now try to show.

The apparent obscurity of some of these works raises the question of their availability in Rome, and the possibility that Dionysius might have consulted them (if, that is, he did consult them) at some centre of learning in the Greek-speaking world. This possibility cannot be certainly disproved, but it seems highly unlikely. There is no indication that Dionysius ever studied anywhere except Rome; he does not even seem to have visited Athens.^[54] Since he lived continuously in Italy from the time of his arrival at the close of the civil war (1.7.2), any possible period of study must have preceded 30/29 BC, whether in Halicarnassus or elsewhere. But it seems more than doubtful — indeed virtually impossible — that the theme of demonstrating Rome’s Greek character could have been conceived before Dionysius had some personal acquaintance with the Romans, their language, culture, and the religious, political and other practices which seemed to him to be Greek. Resources in Rome were presumably ample: there were several more or less public libraries, containing many works acquired from the Greek world, and under the direction of Greek scholars competent in library skills.^[55] Dionysius could also have drawn upon private resources: works owned by his Roman patrons and Greek friends.

If, then, the works cited by Dionysius would have been available to him in Rome, the question must be pursued by consideration, first (section 5.1), of the case against Dionysius’ good faith and reliability, and, second, (section 5.2), of arguments in favour of them. For there are really two questions here: did Dionysius do what he claims to have done? and, if he did, was he reliable in his reportage of the sources?

Two critical items require consideration.

In 1.29.3 Dionysius reproduces Herodotus 1.57.3 word for word,^[56] apparently with a text of Herodotus open in front of him — except that he makes the passage refer to the Krotonietai, the people of Umbrian Cortona (called Kroton by Dionysius),^[57] whereas the manuscripts of Herodotus read Krestonietai (the people of Creston in Thrace). The discrepancy has long been a matter of dispute: should the text of Herodotus be emended on the basis of Dionysius’ reading? or did Dionysius incorporate a different reading, either fraudulently in order to strengthen his own case, or deliberately, believing it to be a better reading, or carelessly, by following another writer’s citation of Herodotus? Briquel has thoroughly investigated the possibilities, and has shown good reason to accept the first suggestion: that Dionysius accurately transmits what he found in his text of Herodotus, and that his reading is the correct one.^[58] If this is indeed so, this example actually supports Dionysius’ accuracy.

Herodotus provides a second test of Dionysius’ precision: the topic under discussion is the question of Etruscan origins.^[59] When paraphrasing the text of Herodotus 1.94, Dionysius introduces (1.27) a reference to Lydus son of Atys, whereas Herodotus at that point mentions only the one son, Tyrrhenus, who left the country with the party of emigrants while his father Atys stayed behind. The explanation could be that Dionysius is working from memory or, perhaps more likely, that he is combining accounts as he composes: Herodotus has indeed mentioned Lydus son of Atys at 1.7.3.^[60] Dionysius thus retains Lydus as eponym of the Lydians, no doubt on the assumption that someone must have succeeded Atys, although this is not specifically mentioned by Herodotus at 1.94. The slight distortion is due to combination and inference, and can be counted as a normal procedure in learned writing. No doubt the same may well have happened elsewhere. There is nothing here for the prosecution. Indeed, this case, too, speaks in Dionysius’ favour.

Obviously, items (vii), (viii) and (x) in section 4 (respectively the practice of direct quotation, seeming accuracy of quotation, constant evaluation of sources by relatively intelligent criteria) do not only look good: they make a genuinely favourable impression, an impression enhanced by the confidence with which Dionysius constantly lets us ‘see his workings’. Again, two cases must be examined.

The instance of verbatim quotation from Antiochus of Syracuse (mentioned above under item (vii) in section 4) provides a striking example of Dionysius’ readiness to allow the reader to evaluate his usage of his authorities:

These things are recorded not by any of the haphazard or recent historians but by Antiochus of Syracuse, whom I have mentioned before also. He says that when Morges was reigning in Italy (at that time Italy was the coastal area from Tarentum to Posidonia) a man came to him who was an exile from Rome. He says this: “When Italus was growing old, Morges became king. In his reign a man arrived from Rome, an exile: Sicelus was his name”. (1.73.4)

The quoted words add only the actual name of Sicelus to the information already given in the paraphrase but they bring the reader into direct contact with the raw material of the narrative and they contribute to Dionysius’ authority by fusing him with the older writer: as it were, Dionysius ‘states’, and Antiochus ‘speaks’. In addition, the clipped and simple style instantiates the antiquity of the source: Dionysius considers brevity, a simple style and unadorned sentences to be characteristic of the early Greek historical writers.^[61]

Dionysius’ reading of antiquarian works and, even more, his own practice of quoting in his literary treatises, where he sometimes refers to quotation kata lexin (‘word for word’) and on other occasions quotes from memory, may have inclined him to apply this demonstrative method to the writing of history.^[62] The passage in which Dionysius elucidates and expands Fabius Pictor^[63] on the Roman games by adding long illustrations from Homer is reminiscent of the De Compositione Verborum.^[64] However, the elaboration of the passage on the games is exceptional. Other quotations tend to be only six or seven lines;^[65] the majority are from prose works, though there are also a few very brief passages of verse.^[66]

The second notable case of ‘showing the workings’ occurs when Dionysius actually does make a deliberate change to the reading of his source and draws attention to the fact. This is with an item attributed to Myrsilus of Lesbos:

These things Myrsilus of Lesbos has recorded, writing in almost the same words as I do now, except to the extent that he calls the men not Pelasgians but Tyrrhenians: I shall state the reason a little later. (1.23.5)

He proceeds to explain the confusion between Tyrrhenians and Pelasgians in 1.25-29. This controversial point is argued out at considerable detail, and various authorities are adduced (1.28-29).^[67] Since texts of Myrsilus must have been much less readily available than those of Herodotus, this indicates that Dionysius’ readiness to offer checkable proofs for the stages of his argument is genuine.

At this point it is finally legitimate to bring the question posed at the start of section 5 (the validity of Dionysius’ claim to wide reading) together with the issues outlined in section 4 (the different ways he deploys his sources to substantiate the claim). Unless Dionysius be regarded as a complete charlatan or as a complete incompetent (views both a priori implausible and incompatible with the evidence that is directly checkable), these two aspects are interrelated. The whole basis of Dionysius’ argument in Book 1 relies on the collation of authorities and on his proving his thesis of the Greek origin of Rome by a step-by-step examination of the scattered references to immigrant leaders and peoples.^[68] His method, it seems, was to systematise these various allusions; substantial distortion of the traditions was neither necessary nor likely to benefit his case. He is perfectly ready to argue out controversial instances, and in so doing he shows considerable familiarity with both Greek and Roman sources; it is utterly implausible to envisage a predecessor, especially as the existence of such a predecessor would make Dionysius’ claims for himself extremely vulnerable.

The overall conclusion, therefore, must be that Dionysius did do what he claimed to do in regard to his sources and that by ancient standards he is reliable in his collation and reporting of them, although not necessarily always verbally precise. Of course, the above discussion has been on a somewhat narrow basis and a full assessment of Dionysius’ competence in his use of sources must bring in much broader questions of critical judgement (see sections 7.1-3); nothing I have said here is meant to deny that Dionysius is capable of carelessness and error on occasion, of misconception, of rather surprising misjudgement and, no doubt, of a certain amount of misrepresentation in the interests of the thesis that the Romans were originally Greeks.^[69] Yet even when these things are taken into account, the overall picture remains a positive one — certainly much more positive than some scholars have made it out to be.

Any discussion of Dionysius’ use of sources must take account of the special status within the AR as a whole of Book 1, which is largely concerned with the ‘mythical period’ and is the most important single site for Dionysius’ demonstration of the Greek origins of the Romans. Items (ii) and (iii) in section 4 highlighted the marked difference between Dionysius’ practice of source citation and acknowledgement of variants in Book 1-4 (especially Book 1) and subsequently. In the later books, the sources are naturally mostly Roman ones, comprising both annalists and antiquarians. Again, Dionysius seems to have read widely and such reading must have taken place in Rome, after he had achieved a competent command of Latin (1.7.2-3). But his practice in these books is to name authors only for specific points of information, or in cases of conflict.

The reason for the difference of practice is obvious enough, although itself eloquent of the extent to which Dionysius is in control of his source material and of the argumentative structures which underpin his narratives. The subject-matter of Book 1 is of its very nature intensely problematic and hence requires extensive citation. Such citations demonstrate the diversity of the ‘historical’ record; when they have been properly sifted, some prove authoritative and prove the historian’s judgement. For the remaining books, the underlying late annalistic material is generally agreed to derive from Licinius Macer and /or Valerius Antias; either or both of these may have been mediated through Aelius Tubero. But including the preface, these writers are merely named (respectively) seven times, twice, and twice,^[70] almost as if Dionysius is following a convention, a principle of not naming the most used sources^[71] — surely not merely because this breaks the narrative flow.

To the explicit citations a few additions can reasonably be made. Thus, the question of Dionysius’ source or sources for any given item raises all the notorious problems of Quellenkritik. ^[72] Even in the cases where a well-defined factual point is involved, there is no certainty^[73] about many of the suggested attributions to Fabius Pictor,^[74] Ennius,^[75] Cato,^[76] Fabius Maximus,^[77] Piso,^[78]Polyhistor,^[79] and, above all, Varro.^[80] In the account of the exposure, survival and recognition of Romulus and Remus (1.79.4-83.3), explicitly stated by Dionysius to be attributable to Fabius Pictor (gegraphe: ‘he has written’), whom a number of others ‘followed’ (ekolouthesan), the difficulty of positively identifying the Fabian material is generally recognised.^[81] For the long anonymous tracts of text where Dionysius’ sources were the later annalists, opinions over identification in particular cases vary even more widely, despite the general consensus concerning Macer and / or Antias as providing the main narrative threads.

Dionysius has presented himself in the preface as one who engages fairly with his predecessors but is capable both of assessment and of justified criticism; further self-definition occurs both implicitly and explicitly later. Such definition may be in programmatic statements (such as secondary prefaces),^[82] or it may be manifested by means of declared judgement and choice, or it can be implied in the ‘action’ of the author. Given a notion that the past is a common heritage, shared by all and not monopolised by any individual, it is possible, even easy, to avoid acknowledgement of one’s predecessors: this may account for some ancient silences.^[83] Thus one form of action is to engage (with little or no explicit mention of predecessors) in creative adaptation within that common heritage; such creative engagement is there to be recognised by the cognoscenti among readers. In addition to these considerations, which apply generally to more or less all ancient historians, Dionysius had a very specific project: that of demonstrating the Greek origins of the Romans. Accordingly, the next section will examine Dionysius’ explicit evaluation of his sources: his terms of praise or criticism, the manner in which an author is introduced and the reasons adduced for deciding — or failing to decide — between variants give a useful indication of the criteria by which Dionysius judged his predecessors and further define and illustrate his conception of his historiographical task.

Dionysius quite often characterises named authors. Sometimes he does this by a term describing the writer himself,^[84] sometimes by the addition of the work’s title^[85] or of a phrase summarising its content.^[86] In other cases no help is given beyond the name: Dionysius apparently expected his readers either to recall readily who was meant^[87] or not to have a greater interest in an author’s context than will be satisfied by a brief allusion. He does however sometimes refer to an author’s credentials: to belong to an early epoch^[88] (hence of course to be nearer the historical events) or to have a claim to high personal status are the most important ones.

The antiquity of a source is felt as inherently important. There is a note of apology in the admission that the Romans have no single ancient historian; on the other hand, the Roman historians did enjoy access to some ancient records:

Not one single ancient historian or logographer do the Romans have: from ancient accounts preserved on sacred tablets each took something and wrote it up. (1.73.1)^[89]

Fabius Pictor naturally possesses the authority deriving from his position as most ancient Roman historian, but Dionysius also emphasises that Fabius was writing on the basis of his own knowledge:

I shall adopt my evidence from that time when they did not yet hold dominion over Greece nor any other overseas rule at all, using Quintus Fabius in support and not needing any further proof: for that man is the most ancient of those covering Roman matters, providing proof not only from what he heard but also from what he himself knew. (7.71.1)

Age and reliability are mentioned together (1.49.1) in the cases of Cephalon of Gergis^[90] and Hegesippus of Mecyberna, ‘men early and worthy of account’.^[91] Antiochus of Syracuse is specially mentioned as not being one ‘of the haphazard or recent historians’ (1.73.4). On several other occasions Dionysius characterises a writer as ‘ancient’ (palaios) or ‘early’ (archaios).^[92]

Personal status constitutes the other main claim. Dionysius considers censorial or senatorial rank as worth mention (though by no means in all relevant cases).^[93] He also alludes to general standing (L. Mallius at 1.19.3 is ‘not unworthy of note’),^[94]and describes Xanthus of Lydia^[95] as being both a native and well skilled in ancient history:

as skilled as any about ancient history, and for that of his native land reckoned a supporting authority inferior to none ... (1.28.2)

In general, as a Greek writing about Rome for a largely Greek audience, Dionysius naturally needs to stand by the native Roman tradition.^[96] Nevertheless, he does not automatically accept the testimony of any individual Roman as particularly valuable merely because he is a local.^[97] He defines his methodological position (mutatis mutandis echoing Thucydides 1.20-22) thus:

not regarding it as sufficient for those writing up early and local histories to go through recounting them in a manner worthy of credence as they have received them from the locals, but thinking that they need many indisputable testimonies if they are going to appear credible. (7.70.2)

Local traditions — perhaps particularly those which deal with difficult and confused archaia (‘early things’, cf. 1.8.1) — require substantiation. Here Dionysius may well be thinking of oral accounts as well as written ones: either ‘tourist guide’ information about place-names and monuments at a popular level, or scholarly theories passed on by the learned men of his acquaintance. However, most of the accounts used are in fact written ones. The sort of extra testimony needed is shown in 7.72, where the comparison of Homer and Fabius Pictor on the Roman games is carefully reasoned. Being ‘worthiest of credence and earliest’ (7.72.3), Homer must of course depict the purest, most typically Greek, practice; Pictor, an eyewitness, dates to an epoch at which no possible influences derived from Rome’s later conquest of Greece can be supposed to have affected Roman ritual. Thus, if the ritual was Greek in Pictor’s time, this proves that the tradition had come down unimpaired from much earlier times.

So another might assume that the things now done in the city might suffice as no small indication of ancient customs. But I, lest anyone should assume this to be a weak proof, according to the unpersuasive assumption that having conquered the whole Greek world they gladly learnt better customs, despising their native ones, I shall adopt my evidence from that time when they did not yet hold dominion over Greece nor any other overseas rule at all ... (7.71.1)

There follows the passage quoted above. Therefore the Greek elements found in use at Rome prove the Romans’ noble Greek ancestry. Dionysius’ numerous misconceptions — about Graeco-Roman contacts in the third century and earlier, about Pictor’s history, about the nature of religious practice^[98] — are beside the point: Pictor’s is the sort of testimony which, for Dionysius, constitutes proper corroboration. Other native traditions are confirmed in the same way: e.g. the term ‘Saturnian hill’ did not derive from ‘hill of Kronos’ in Heracles’ time (1.34.4) but bore that name earlier, as the terminology of oracular literature and many place-names can confirm (1.34.5)

Certainly, in Books 1 and 2, no special authority resides in the local accounts as such. Allusions to ‘native’ or ‘Roman’ accounts do no more than combine with and support traditions known to Dionysius from elsewhere, forming a Hellenocentric pattern of origines gentium.^[99] He consistently seeks agreement, or at least the possibility of coordination, with Greek traditions, as in his discussion of the origin of the Sabines. The accounts of Zenodotus of Troezen and of Cato are followed by one which made the Sabines Spartans by origin: ‘there is also another account about the Sabines stated in the native histories’ (2.49.4).^[100] Dionysius does not explicitly decide among the versions, but appears to lean towards the last: this is told at greater length, is connected with a Greek etymology, and links Spartan and Sabine frugality.

Sometimes Dionysius shows a tendency to represent native traditions as unanimous or virtually so. He may do this by contrasting a Greek view or views with ‘(all) the Romans’ (1.31.1, 1.49.3). On other occasions there is no question of such polarised Greek-Roman views: Dionysius simply asserts that the Roman tradition is unanimous, e.g. ‘as all the Roman historians say’, 5.11.2; ‘as the majority of Roman historians write’, 3.62.1);^[101] and on the Capitoline temple:

It is worth going through the happenings before its building which have been relayed by all those compiling local histories. (3.69.3)

However, the story of Terminus and Juventas which follows is in fact found in other sources with considerable differences of chronology and detail.^[102] In this case, then, Dionysius is either deliberately misleading or at any rate unusually careless in representing the tradition as unanimous.

Outright praise or blame is fairly sparing. Clearly Dionysius intends approval of Xanthus at 1.28.2; other writers who receive explicit endorsement are Varro (2.21.2 ‘the man most knowledgeable of those flourishing in the same age’), Cato, and Aelius Tubero. Cato, who is regarded as ‘more worthy of credence than either’ of Fabius Pictor and Vennonius (4.15.1), is twice praised for his carefulness: ‘compiling most carefully the genealogies of the cities of Italy’ (1.11.1) and ‘careful if any man was in the compilation of the earlier period of history’ (1.74.2). Tubero is similarly characterised: ‘a clever man and careful in the compilation of history’ (1.80.1). Care in bringing data together evokes Dionysius’ admiration, while he condemns lack of care and lack of proof. A group of Greek writers is blamed for insufficient enquiry (and the terminology again recalls Thuc. 1.22):

... each of them wrote up a few things, not even accurately investigated himself, but putting them together from chance hearings (1.6.1)

Similarly on the question of Vesta:

For there are things in this area thought worthy of enquiry by many Roman writers, of whom those who have not scrutinised the causes carefully have brought out rather useless books. (2.64.5)

Some Roman historians provide no proof of their assertions:

... while using a Greek tale they provided as support none of those writing of Greek affairs. (1.11.1)

Like Polybius, Dionysius blames Timaeus for chronological inexactitude (1.74.1),^[103] while apropos the discussion of the sacred objects at Lavinium Timaeus is certainly included among those who try to enquire into what it is not themis to know:

I resent, too, those others who see fit to enquire into or to know more than is allowed. (1.67.4)

Dionysius also theorises about the methodology of those who give diverse accounts of the objects in the temple of Vesta. It is known that Metellus saved them from burning in 241:

taking this as agreed, they attach some conjectures of their own. (2.66.5)

Dionysius refuses to contribute any further conjectures as to the nature of these holy things (2.66.6). In general, compared with the practice of Theopompus, Polybius, or Timaeus/Epitimaeus himself, Dionysius’ criticisms of his predecessors are neither immoderate nor unjust.^[104]

Where chronological problems are involved, Dionysius is confident in determining disputes on the basis of his own preliminary chronological work. He has also worked out the relationship between the two Tarquinii and their various connexions and descendants.^[105] These data are used on different occasions to identify an individual or to reject an account on grounds of chronological / genealogical improbability.^[106] Feeling himself to be on secure ground, he is therefore ready to draw attention to the carelessness in this respect of previous writers, including Fabius Pictor:

Here again I am forced to recall Fabius and to confute his carelessness in the scrutiny of times (4.30.2-5, cf. 4.64.3)

Licinius Macer and others incur similar reproaches:

For Licinius and those with Gellius, not having scrutinised the probabilities or possibilities at all, introduce King Tarquinius himself as involved in the fighting ... (6.11.2)

Numa’s association with Pythagoras (2.59)^[107] and the gift of corn from Dionysius of Syracuse (7.1), are similarly rejected in an unusually decisive fashion. Where no question of chronology is involved, Dionysius only twice rejects a named source without any support from another author.^[108] Chronological exactitude is thus a prime requisite for a historian, its absence grounds for blame.

To some extent, even where explicit expression of praise or blame is lacking, Dionysius’ terminology will suggest his evaluation of a given logos (‘account’).^[109] His usage is not hard-and-fast: graphei (‘writes’) is often used apparently without nuance of a named author but, when applied to an anonymous account, it seems to imply neutrality or slight disapproval.^[110] Most unattributed tales are left with the neutrality of legousi (‘they say’) / legetai (‘it is said’) / phasi (‘they say’).^[111] Muthologousi (‘they tell the story’) usually but not always conveys slight disparagement;^[112] apophainei (‘indicates’)^[113] and historei (‘relates’)^[114] are neutral or approving. There are a few rarely used terms of strong assertion, such as bebaioi (‘guarantees’),^[115] marturei (‘attests’)^[116] and their cognates. The verb deloun (‘make clear’), occasionally applied to other sources, is more often reserved for a very positive statement by Dionysius himself.^[117] Quite frequently he refers to a pithanotatos (‘most persuasive’) or pithanoteros (‘more persuasive’) account;^[118] sometimes to an alethesteros (‘truer’) one.^[119] These terms may be in contrast with a muthodesteros (‘more story-like’ / ‘more mythical’) version.^[120]

This brings us to the crucial, but slippery, muthos-historia distinction. Dionysius appears to be handling the muthikos (‘story’ / ‘myth’) element firstly by establishing a broad division which corresponds to the earliest and later epochs; then he goes on to redefine^[121] what is more or less a matter of tales, more or less a matter of history proper. In connection with Heracles, he seems to imply that the muthikos does not befit history:

Such, then, is the mythical account about him which has been handed down. But the truer one, which many of those who narrated his deeds in the form of history have adopted, is this ... (1.40.6-1.41.1)^[122]

Such redefinition allows Dionysius to include many variants inclining to the muthodes (‘story-like’ / ‘mythical’) end of the scale. On some occasions he tends to prefer a rationalised explanation, for example on Numa and Egeria, with their Greek analogues (2.61.2). In the case of Servius Tullius’ miraculous birth, it seems as if two — or three — improbabilities make a probability. A rationalised account comes first (4.1), the muthodes (‘story-like’ / ‘mythical’) version second (4.2.1-3); Dionysius then continues:

While this myth seems not entirely credible, another divine appearance — a wonderful and surprising one — relating to this man renders it rather less to be distrusted. (4.2.3)

Dionysius notes that several people witnessed this second occurrence. The last miracle (the unburnt statue) occurred after Servius Tullius’ death:

And in fact another divine act made it clear that he was a man dear to the gods, as a result of which the incredible tale assumed concerning his birth, as I mentioned before, was trusted by many as true. (4.40.7)

There is some ambiguity as to Dionysius’ own position: edelose (‘ made it clear’) might indicate that he agreed with this view, but the end of the sentence and hupo pollon (‘by many’) have a distancing effect. Whatever his own views, he has it both ways in his history — perhaps trying to suit everyone (cf. 1.8.3), perhaps in the belief that these heroic and mythical logoi (‘accounts’) illuminate the Roman character.^[123]

Occasionally Dionysius shows that he thinks of different logoi (‘accounts’) as having a core of truth or reliability, despite variants or accretions. There is a common factor in Antiochus’ and Hellanicus’ accounts of the naming of Italy: their derivations are different but ‘that at least is clear from both’ (1.35.3) — that the name dates back at least to Heracles’ time. Then Dionysius turns to another local legend:

There is also another account told as a tale by the locals ... (1.36.1)

This refers to the belief that during the reign of Saturn (whom Dionysius calls Kronos) Italy had been particularly favoured. Again, Dionysius finds it possible to ask his reader to set aside the mythical aspect in favour of rational examination:

And anyhow, if anyone, putting aside the story-like element of the account, were willing to scrutinise the excellence of the country ... (1.36.2)^[124]

The result will be an unprejudiced assessment on rational criteria of the fertility of Italy. The passage which follows (1.36.2-37.4) constitutes an extensive laus Italiae (‘encomium of Italy’). Dionysius then concludes:

It was no wonder, therefore, that the ancients assumed this country to be sacred to Saturn ... (1.38.1)

The ancient name of Saturnia (1.35.3) can thus be given a rational justification.^[125]

Dionysius’ approach to his sources is on the whole positive and candid.^[126] He is looking for the acceptable elements in what they have to offer, rather than seeking to criticise or condemn. However, particularly in Book 1, where his compositional method often brings out contradictions, he finds himself having to choose between variant accounts. This often involves him in discussion, and it is apparent that his main criteria for resolving discrepancies fall roughly into three categories: (1) supporting evidence (including documents and monuments); (2) reasoned argument (including consistent argumentation for genealogical and chronological questions); (3) general credibility (an appeal to likelihood and common sense versus the improbable or mythical). In any particular case, however, it is apparent that the three categories are not rigidly exclusive: there is often slippage between, or combination of, two types. Finally, where Dionysius feels the proof to be inadequate or inconclusive, he often leaves the question open for the reader.^[127]

(1) Supporting evidence includes the testimony of credible authors:

But I do not know how this could be demonstrated, not being handed down in history, as far as I know, by any of the Romans or Greeks worthy of account. (2.59.5)

Axiologos here (‘worthy of account’) is somewhat reminiscent of the praise of Cato in 4.15.1 (quoted above) and the thought recurs in the other general references.^[128] Absence of such testimony is several times noted; cf. 1.11.1 (quoted above). Support is also given by religious rituals (e.g. 1.49.3, 7.70.2) and monuments: this may either be implicit - Piso’s ‘proof’ from Tarpeia’s tomb (2.40.2-3) is a good example - or explicit:

... as the inscription on his statue situated on the Capitol testifies (2.66.4)

(2) Reasoned argument may of course sometimes involve adducing testimony, as above; on other occasions it is more abstract: the proof of Rome’s foundation date, or the genealogical mathematics covering the relationships of the whole extended Tarquinian family. Here Dionysius feels himself to be dealing with fairly firm dating evidence, as 4.6.4 shows: he notes that authorities vary as to whether Tarquinius arrived in the first or eighth year of Ancus Marcius’ reign, but he was certainly there by the ninth year, commanding the cavalry - that is agreed, and stands as a fact.

The dispute over Cassius’ trial is an excellent example of Dionysius’ method. He thinks the condemnation by his father is less probable, not because he does not believe a Roman father capable of such a stern act (8.79.2), but because there could be no dedication from the property of one still in patria potestate, and the state would not have confiscated the elder Cassius’ property (8.79.3).^[129] He can even point to the inscribed offerings which prove the dedication: thus some supportive evidence backs up his argument.^[130]

(3) The mode of argument just discussed has a tendency to shade into argument from general probability^[131] (elements of this are present in 4.6). A lengthy demonstration (9.22) that there must have been more than one surviving Fabius after the massacre at the Cremera depends entirely on probability and common sense. Though, as we have seen, Dionysius is, on the whole, not very ready to condemn his predecessors, he is prepared to do so on some occasions. Here he feels that (some of) his authorities cannot be taken seriously and is strongly condemnatory:

What is attached to this by some, while neither true nor persuasive but fabricated by the multitude from some misreport, is not to be left on one side unscrutinised. (9.22.1)

After outlining all the unlikely conditions which would have to obtain, Dionysius concludes:

Scrutinising the account in this particular way I have reckoned that it is not true, but that this is true ... (9.22.5)

Similarly, he is prepared to argue on grounds of the improbability of a secret conspiracy, and of the impossibility of divine approval, against the view that Tullus Hostilius was assassinated (3.35.5-6).^[132]

The variants which are discussed by Dionysius are strikingly about matters of fact, where argument or proof of various kinds can be employed to decide such questions as disputes over date or identity. This is very marked in the later books, where fewer than a dozen variants are mentioned, several of which appear to be traditional instances of dispute.^[133] It is apparent that Dionysius’ historiographical practice implies a different attitude in regard to the earliest and most obscure subject matter compared with that of the truly historical period. In the later books, the engagement will be by means of speeches; the variants will be conveyed through the personae of the opposed speakers in many a debate.^[134] The end of Book 1, where Dionysius negotiates the transition between the mythical and the historical, will be a crucial area for examination and requires a separate paper.

Like all ancient historians, Dionysius had to position himself within the historiographical tradition, hence, in part, the density of the allusions in the preface to such great predecessors as Herodotus, Thucydides and Polybius. His specific project of writing an ‘archaeology’ of Rome required particular engagement with Polybius, whose work he set out to complete by the paradoxical provision of a ‘prequel’ and many of whose canons he accepted but two of whose historiographical positions (his choice of contemporary or near-contemporary history and his rejection of ‘arm-chair’ historiography) ran flatly counter to Dionysius’ project. Not only had Dionysius to justify the importance of his ‘Roman archaeology’ as a historical theme, hence his deceptive playing down of his own role at the beginning of the preface, but he also had to justify its feasibility. Necessarily largely dependent upon written sources, he had to show that he had read everything relevant to his theme, exercised proper critical judgement and been able to extrapolate solid historical material. The problem was the more acute because he was also committed to demonstrating the Greek origins of the Romans, hence to a serious reconstruction of the very earliest period. The result, in Book 1 and, to a lesser extent, in Books 2-4, is an almost unprecedented parading of his workings, the necessity for which disappeared in the rest of the work. Dionysius’ periodisation of his vast chronological theme, his deployment and evaluation of his sources, and his general historiographical criteria show him to be a historian who is not only highly conscientious but also, on the whole, thoughtful and reliable and much more fair-minded in his judgement of his fellow-historians than most, and one, moreover, whose material is at once rigorously directed towards the validation of his historical thesis and (in a further contrast with Polybius) generously inclusive of all the different historiographical modes.

A. Andrèn, ‘Dionysius of Halicarnassus on Roman monuments’, Hommages à L. Herrmann (Brussels 1960) 88-104

H. Veit Apfel, Literary Quotation and Allusion in Demetrius peri hermeneias and Longinus peri hupsous (Diss. Columbia 1935)

D. Asheri, La Lidia e la Persia. Erodoto. Le storie. Libro I. Testo e commento (Milan 1988)

E.J. Bickermann, ‘Origines gentium’, CPh 47 (1952) 65-81

R. Blum, Kallimachos. The Alexandrian Library and the origins of bibliography, tr. H.H. Wellisch, (Madison 1991)

O. Bocksch, ‘de fontibus libri V et VI Antiquitatum Romanarum Dionysii Halicarnassensis quaestiones variae’, Leipziger Studien (17) 165-274

G. Bowersock, Augustus and the Greek world (Oxford 1965)

D. Briquel, Les Pelasges en Italie. Recherches sur l'histoire de la légende (Rome 1984)

D. Briquel, ‘Denys d’Halicarnasse et la tradition antiquaire sur les Aborigènes’, Pallas 39 (1993) 17-39

G. Capdeville, ‘Les institutions religieuses de Rome selon Denys d’Halicarnasse’, Pallas 39 (1993) 152-72

O. de Cazenove, ‘La chronologie des Bacchiades et celle des rois étrusques de Rome’, MEFRA 100(2) (1988) 615-48

M. Chassignet, L’annalistique romaine. t. 1. Les annales des pontifes et l’annalistique ancienne (Paris 1996)

K. Clarke, ‘Universal perspectives in historiography’, in C.S. Kraus (ed.), The limits of historiography. Genre and narrative in ancient historical texts (Leiden 1999), 249-79

G. Cobet, Observationes criticae et palaeographicae ad Dionysii Halicarnassensis Antiquitates Romanas (Leiden 1877)

F. Della Corte, La mappa dell’ Eneide (Florence 1972)

F. Della Corte, ‘L’idea della preistoria in Varrone’, in Atti del congresso internazionale di studi varroniani Rieti settembre 1974 (Rieti 1976)

A. Dubourdieu, ‘Denys d’Halicarnasse et Lavinium’, Pallas 39 (1993) 71-82

S. Ek, Herodotismen in der Archäologie des Dionys von Halikarnass (Lund 1942)

D. Fehling, Herodotus and his ‘sources’: citation, invention and narrative art, tr. J.G. Howie (Leeds 1989)

L.H. Feldman, ‘Hellenizations in Josephus’ Jewish Antiquities: The portrait of Abraham’, in L.H. Feldman and G. Hata (eds.), Josephus, Judaism and Christianity, (Leiden 1987), 133-53

C.W. Fornara, The nature of history in ancient Greece and Rome (Berkeley 1983)

G. Forsythe, The historian L. Calpurnius Piso Frugi and the Roman annalistic tradition (Lanham 1994)

M. Fox, ‘History and rhetoric in Dionysius of Halicarnassus’, JRS 83 (1993) 31-47

B.W. Frier, Libri annales pontficum maximorum. The origins of the annalistic tradition (Rome 1979) (The second edition (Ann Arbor 1999) reprints a text unchanged except for small corrections, and includes a new introduction.)

V. Fromentin, ‘La definition de l’histoire comme “mélange” dans le prologue des Antiquités Romaines de Denys d’Halicarnasse’, Pallas 39 (1993) 177-92

E. Gabba, ‘Studi su Dionigi da Alicarnasso. III. La proposta di legge agrarie di Sp. Cassio’, Athenaeum n.s. 42 (1964) 29-41

E. Gabba, ‘Dionigi d'Alicarnasso sul processo di Spurio Cassio’, in La storia del diritto nel quadro delle scienze storiche. Atti del I. congresso internazionale della società italiana di storia del diritto (1966) 143-153

E. Gabba, ‘Considerazioni sulla tradizione letteraria sulle origini della Repubblica’, in Les origines de la république romaine. Entretiens Fondation Hardt 13. (Vandoeuvres-Genève 1967) 133-74

E. Gabba, ‘Dionigi e la “storia di Roma arcaica”’, Actes 9 Congrès Association G. Budé (1975) 218-29

E. Gabba, ‘Mirsilo di Metimna, Dionigi e i Tirreni’, RAL 30 (1975) 35-49

E. Gabba, Dionysius and The History of Archaic Rome. (Sather Classical Lectures, 56) (Berkeley 1991)

B. Gatz, Weltalter, goldene Zeit und sinnverwandte Vorstellungen (Hildesheim 1967)

G.P. Goold, ‘A Greek professorial circle at Rome’, TAPhA 92 (1961) 168-192

S. Gozzoli, ‘Un teoria antica sull’origine della storiografia greca’, SCO 19-20 (1970-1) 158-211

C. Hedrick, J. Marincola, E. O’Gorman, J. Moles. ‘Exchange and reply’ [to Moles 1999] HISTOS 3 (1999) (June, 2000). 19 December 2000. <http://www.dur.ac.uk/Classics/histos/1999/molesexchange.html>

E. Herkommer, Die Topoi in den Proömien der römischen Geschichtswerke (Diss. Tübingen 1968)

N. Horsfall, ‘Some problems of titulature in Roman literary history’, BICS 28 (1981) 103-14

N. Horsfall, ‘Empty shelves on the Palatine’, G&R 40 (1993) 58-67

N. Horsfall, ‘Rome without spectacles’, G&R 42 (1995) 49-56

F. Jacoby, Atthis: the local chronicles of ancient Athens (Oxford 1949)

C. Joyce, ‘Was Hellanikos the first chronicler of Athens?’ HISTOS 3 (1999) (July 1999). 19 December 2000. <http://www.dur.ac.uk/Classics/histos/1999/joyce.html>

R. Jumeau, ‘Un aspect significatif de l’exposé livien’, in M. Renard and R. Schilling (eds.) Hommages à Jean Bayet. Collection Latomus 70 (Brussels-Berchem 1964) 309-33

A. Klotz, Livius und seine Vorgänger (Leipzig-Berlin 1940-1)

R.A. Laroche, ‘The Alban king-list in Dionysius I, 70-71: a numerical analysis’, Historia 31 (1982) 112-20.

R.A. Laroche ‘Popular symbolic / mystical numbers in antiquity’, Latomus 54 (1995) 568-76

D. Levene, ‘Sallust’s Jugurtha: an ‘historical fragment’’, JRS 82 (1992) 53-70

W. von Leyden, ‘Spatium historicum’, DurhamUJ n.s. 11 (1950) 89-104

T.J. Luce, Livy:the composition of his history (Princeton 1977)

E. Maass, ‘Tibullische Sagen’, Hermes 18 (1883) 322-42

G. Marenghi, Dinarco (Milan n.d. [1970])

J. Marincola [= Marincola (1997a)], Authority and tradition in ancient historiography (Cambridge 1997)

J. Marincola [= Marincola (1997b)], ‘Odysseus and the Historians’, HISTOS 1 (1997) (October 1996). 19 December 2000.

<http://www.dur.ac.uk/Classics/histos/1997/marincola.html>

A.J. Marshall, ‘Library resorces and creative writing at Rome’, Phoenix 30 (1976) 252-64

F. Millar, A Study of Cassius Dio (Oxford 1964)

J.L. Moles, ‘Herodotus warns the Athenians’, in F. Cairns (ed.), Proceedings of the Leeds International Latin Seminar 9 (1996) 259-84

J.L. Moles, ‘Cry Freedom: Tacitus Annals 4.32-35 ’, HISTOS 2 (1998). (October, 1998). 19 December 2000.

<http://www.dur.ac.uk/Classics/histos/1998/moles.html>

J.L. Moles, ‘Anathema kai ktema: the inscriptional inheritance of ancient historiography’, HISTOS 3 (1999). (October, 1999). 19 December 2000. <http://www.dur.ac.uk/Classics/histos/1999/moles.html>

J.L. Moles, ‘Livy’s preface’, PCPhS 39 (1993) 141-168

T. Mommsen, ‘Mamilius Sura, Aemilius Sura, L. Manlius’, RhM 16 (1861) 282-7

D. Musti, Tendenze nella storiografia romana e greca su Roma arcaica. (Quaderni Urbinati di Cultura Classica 10) (Rome 1970)

R.M. Ogilvie, A commentary on Livy Books 1-5 (Oxford 1965)

L. Pepe, ‘L’annalista Q. Fabio Massimo Serviliano’, StudUrb 49.1 (1975) 95-108

J. Perret, Les origines de la légende troyenne de Rome (281-31) (Paris 1942)

L. Porciani and C.J. Joyce, ‘Exchange’, HISTOS 3 (1999). (February, 2000). 19 December 2000. <http://www.dur.ac.uk/Classics/histos/1999/porcianijoyce.html>

J. Poucet, ‘Varron, Denys d’Halicarnasse, Macrobe et Lactance. L’oracle rendu à Dodone aux Pelasges’, Pallas 39 (1993) 41-69

W.K. Pritchett, Dionysius of Halicarnassus: On Thucydides (Berkeley 1975)

W.K. Pritchett, The liar school of Herodotus (Amsterdam 1993)

E. Rawson, Intellectual life in the late Roman republic (London 1985)

W. Rhys Roberts, Dionysius of Halicarnassus. On Literary Composition (London 1910)

W.A. Schröder, M. Porcius Cato. Das erste Buch der Origines. Ausgabe und Erklärung der Fragmente (Meisenheim 1971)

C.E. Schultze, ‘Dionysius of Halicarnassus and Roman chronology’, PCPhS 41 (1995) 192-214

H.H. Scullard, ‘Two Halicarnassians and a Lydian’, in E. Badian (ed.), Ancient society and institutions. Studies presented to Victor Ehrenberg (Oxford 1966)

V. Strocka, ‘Römische Bibliotheken’, Gymnasium 88 (1981) 298-329

J.-P. Thuillier, ‘Denys d’Halicarnasse et les jeux romains (Antiquités romaines VII 72-73)’, MEFRA 87 (1975) 563-81

O. Tomasini, ‘Per l’individuazione di fonti storiografiche anonime latine in Dionisio d'Alicarnasso’, AFLT 1 (1964-5) 153-74

D. Toye, ‘Dionysius of Halicarnassus on the first Greek historians’, AJP 116 (1995) 279-302)

L. Trojani, ‘Contributo alle problematica dei rapporti fra storiografia greca e storiografia vicino-orientale’, Athenaeum 61 (1983) 427-438

A. Valvo, ‘Le vicende del 44-43 a.C. nella tradizione di Livio e di Dionigi su Spurio Melio’, CISA 3 (1975) 157-83

H. Verdin, ‘Notes sur l’attitude des historiens grecs à l’égard de la tradition locale’, AncSoc 1 (1970) 183-200

P. Veyne, Did the Greeks believe in their myths?: an essay on the constitutive imagination, tr. P. Wissing (Chicago 1988)

K. von Fritz, Die griechische Geschichtsschreibung (Berlin 1967)

F.W. Walbank, Polybius (Sather Classical Lectures, 42) (Berkeley 1972)

F.W. Walbank, ‘Polemic in Polybius’, JRS 52 (1962) 1-12 = Selected papers: studies in Greek and Roman history and historiography (Cambridge 1985) 262-79

H.D. Westlake, ‘Hos eikos in Thucydides’, in Essays on the Greek historians and Greek history (Manchester 1969) 153-60

T.P. Wiseman, Clio’s cosmetics (Leicester 1979)

T.P. Wiseman, Remus. A Roman myth. (Cambridge 1995)

A.J. Woodman, Rhetoric in classical historiography (London 1988)

^[1] Thanks are due to those who read and commented on earlier versions of this paper: Dr C.S. Kraus, Prof. J. Marincola, Prof. J.L. Moles, Prof. T.P. Wiseman and Prof. A.J. Woodman. I owe special thanks to Tony Woodman for much help with the translations, which aim as far as possible to reproduce significant verbal relationships in the Greek, although, as usual, HISTOS readers with Greek will find it useful to have the Greek text before them. The presentation of the paper (for example, the employment of itemisation and tabulation) is adapted to publication on HISTOS. The HISTOS editor was John Moles. All scholarly references are given in full on their first appearance, thereafter by name and date. A consolidated bibliography for this paper is given at the end. All references of the form ‘1.63.1’ are to Dionysius, Antiquitates Romanae, unless otherwise specified; citations of Dionysius’ literary works
^[2] J. Marincola, Authority and tradition in ancient historiography (Cambridge 1997) = Marincola (1997a) 95-117.
^[3] E. Herkommer, Die Topoi in den Proömien der römischen Geschichtswerke (Diss. Tübingen 1968); A.J. Woodman, Rhetoric in classical historiography (London 1988) 5-30; Marincola (1997a) 133.
^[4] Marincola (1997a) 225-36 (esp. 234-6).
^[5] A basic polarity by which ancient historians orientated themselves in relation to their themes (for Dionysius cf. 1.2-4) and their predecessors: Woodman (1988) 40-44; cf. also n. [9].
^[6] Such formal (but deceptive) modesty on the part of the historian can be paralleled in the prefaces of Livy (J.L. Moles, ‘Livy’s preface’, PCPhS 39 (1993) 141-68) and Arrian (in the Anabasis [n. [19] below]) and is taken to an extreme by Xenophon in the Hellenica, who (it seems) does not have a preface at all.
^[7] For the Polybian quality see Pol. 12.25d. 1 (discussed in section 3); for history as a journey cf. 1.4.1 ‘I turned’ (etrapomen); 1.5.2 ‘I shall lead off’ (aphegesomai); 1.8.2 ‘I bring down’ (katabibazo). See further J.L. Moles, ‘Herodotus warns the Athenians’, in F. Cairns (ed.), Proceedings of the Leeds International Latin Seminar 9 (1996) 259-84, esp. 262-5; and J. Marincola, ‘Odysseus and the Historians’, HISTOS 1 (1997 = Marincola (1997b)); cf. also n. [14] below.
^[8] The imagery of ‘rendering account’ recalls the ‘account’ given by magistrates at the end of their period of office (this is Dionysius’ ‘account’ of his ‘office’ as historian of Roman Antiquities); it is also potentially financial and is elegantly turned at 1.6.5 (one of several ‘rings’ with 1.1.1), when Dionysius describes his work as ‘rendering grateful returns’ to Rome for the education and other good things that he has enjoyed during his residence in the city.
^[9] Cf. Pomp. 3.2 ff.; for the noble/ignoble – praise/blame polarity as applied to choice of historiographical theme cf. n. [5].
^[10] The care/pains of the conscientious historiographer: Marincola (1997a) 148-58.
^[11] Dionysius anticipates and answers criticisms from readers who, knowing nothing of the early period, assume his theme is unworthy or trivial. In some contrast, Livy (praef. 4) believes — or affects to believe — that his readers are simply less interested in early history: see Moles (1993) 146 with n. 24. Of course, most of Dionysius’ readers would be Greeks, Livy’s Romans, and that partly explains the authors’ somewhat different stance in regard to this issue.
^[12] See E. Gabba, Dionysius and The History of Archaic Rome. (Sather Classical Lectures, 56). (Berkeley 1991), ch. 4 and 194-200.
^[13] The end of Book 1 (the story of Numitor, Romulus and Remus) forms a transitional area: see my second paper, ‘From muthos to historia in Dionysius of Halicarnassus’.
^[14] If history is a journey [n. [7] above], then it is obviously possible to wander through it in error; cf. already Hdt. 1.95.
^[15] ‘This is a demonstration [apodexis] ... of deeds demonstrated [apodechthenta]’.
^[16] They are Hieronymus, Timaeus, Antigonus, Polybius, Silenus, Q. Fabius, and L. Cincius; for the importance of the point see on 1.7.3 (discussion in section 3).
^[17] This debate in ancient historiography is discussed by Marincola (1997a) 113-17.
^[18] Gabba (1991) ch. 3; Valerie Fromentin, ‘La definition de l’histoire comme “mélange” dans le prologue des Antiquités Romaines de Denys d’Halicarnasse’, Pallas 39 (1993) 177-92. (I hope to deal more fully elsewhere with the design and purpose of the preface to Dionysius’ work.)
^[19] For the technique we might compare the much-discussed ‘Second Preface’ of Arrian’s Anabasis [1.12.5].), where, although Arrian finally introduces himself, he goes one better than Dionysius by not actually naming himself. Marincola (1997a) 146.
^[20] C. Hedrick, J. Marincola, E. O’Gorman, J. Moles. ‘Exchange and reply’ [to Moles 1999] HISTOS 3 (1999) (June, 2000). See section 2.2. 19 December 2000.
<http://www.dur.ac.uk/Classics/histos/1999/molesexchange.html>
^[21] One might compare Livy’s negotiation with the reader in his preface: Moles (1993) 142-55..
^[22] Digressions were conventionally associated with the ‘pleasure’ principle: Woodman (1988) 180-5; Marincola (1997a) 118; for Tacitus’ subversions of that notion see J.L. Moles, ‘Cry Freedom: Tacitus Annals 4.32-35 ’, HISTOS 2 (1998). (October, 1998). See section 4. 19 December 2000.
<http://www.dur.ac.uk/Classics/histos/1998/moles.html>
^[23] Cf. Hdt. 1.120.3; 1.159.3; 8.87.3.
^[24] See N. Horsfall, ‘Some problems of titulature in Roman literary history’, BICS 28 (1981) 103-14.
^[25] See in general L.H. Feldman, ‘Hellenizations in Josephus’ Jewish Antiquities: The portrait of Abraham’, in L.H. Feldman and G. Hata (eds.), Josephus, Judaism and Christianity, (Leiden 1987), 133-53 and 134 for imitation in terms of title and number of books; also Gabba (1991) 214-16, and the works there cited in n. 58.
^[26] C.E. Schultze, ‘Dionysius of Halicarnassus and Roman chronology’, PCPhS 41 (1995) 192-214 at 199 and 214.
^[27] R.A. Laroche, ‘The Alban king-list in Dionysius I, 70-71: a numerical analysis’, Historia 31 (1982) 112-20. Laroche also shows in ‘Popular symbolic /mystical numbers in antiquity’, Latomus 54 (1995) 568-76, esp. 572-4, with nn. 12 and 13, that seventeen is a frequently used ‘significant’ number, often found in the definition of lengthy periods. Cf. also D. Fehling, Herodotus and his ‘sources’: citation, invention and narrative art, tr. J.G. Howie, (Leeds 1989) 219.
^[28] Fourteen because the period of the seven reigns (seven generations, total 244 years) is virtually of the same length as the period from the start of the republic (Olympiad 68.1) to the outbreak (arche) of the First Punic War (Olympiad 128.3): 243 years. This can scarcely be a coincidence, and made Dionysius’ choice of stopping-place additionally appropriate.
^[29] Censorinus, de die natali 21.1-4 Sallmann (cf. Varro, fr. 3 P): nunc veri id intervallum temporis tractabo, quod historikon Varro adpellat. hic enim tria discrimina temporum esse tradit, primum ab hominum principio ad cataclysmum priorem, quod propter ignorantiam vocatur adelon, secundum a cataclysmo priore ad olympiadem primam, quod, quia multa in eo fabulosa referuntur, muthikon nominatur, tertium a prima olympiade ad nos, quod dicitur historikon, quia res in eo gestae veris historiis continentur. (2) primum tempus, sive habuit initium seu semper fuit, certe quot annorum sit, non potest comprehendi. secundum non plane quidem scitur, sed tamen ad mille circiter et sescentos annos esse creditur: a priore scilicet cataclysmo, quem dicunt et Ogygii, ad Inachi regnum annos circiter quadringenti <computarunt, hinc ad excidium Troaie annos octingentos>, hinc ad olympiadem primam paulo plus quadringentos; quos solos, quamvis mythici temporis postremos, tamen, quia a memoria scriptorum proximos, quidam certius definire voluerunt. (3) et quidem Sosibius scripsit esse CCCXCV, Eratosthenes autem septem et quadringentos, Timaeus CCCCXVII, Aretes DXIIII, et praeterea multi diverse, quorum etiam ipsa dissensio incertum esse declarat. (4) de tertio autem tempore fuit quidem aliqua inter auctores dissensio in sex septemve tantum modo annis versata. sed hoc quodcumque caliginis Varro discussit, et pro cetera sua sagacitate nunc diversarum civitatium conferens tempora, nunc defectus eorumque intervalla retro dinumerans eruit verum lucemque ostendit, per quam numerus certus non annorum modo, sed et dierum perspici possit.
This temporal division, with its Greek terminology, may derive from Eratosthenes: Jacoby FGH 2. Teil B, Komm. ad 241 F1c, p. 709; F. Della Corte, ‘L’idea della preistoria in Varrone’, in Atti del congresso internazionale di studi varroniani Rieti settembre 1974 (Rieti 1976), vol.1.111-36 (esp. 130-6).
^[30] W. von Leyden, ‘Spatium historicum’, DurhamUJ n.s. 11 (1950) 89-104. On the Censorinus passage quoted in the preceding note, he comments (95, n. 32): ‘In a systematic arrangement of the spatium historicum, spatium mythicum, etc., attributed by Censorinus ... to Varro, and to my knowledge the first of this kind in Antiquity [but see n. [29]], the character of these periods is outlined according to the nature and extent of our knowledge of them’.
^[31] Cf. also Plutarch’s typically colourful adaptation of this scheme at Thes. 1.1-2.
^[32] Schultze (1995) 194-5.
^[33] On the myth-history relation see Marincola (1997a) 117-27.
^[34] Gabba (1991) 101-2.
^[35] For Dionysius as a sometimes acute reader of Thucydides see J.L. Moles, ‘Anathema kai ktema: the inscriptional inheritance of ancient historiography’, HISTOS 3 (1999). (October, 1999). See n. 17. 19 December 2000. <http://www.dur.ac.uk/Classics/histos/1999/moles.html>
^[36] K. Clarke, ‘Universal perspectives in historiography’, in C.S. Kraus (ed.), The limits of historiography. Genre and narrative in ancient historical texts (Leiden 1999), 249-279, contrasts (278) Livy and Dionysius with universal historians stricto sensu. While undoubtedly fair from a formal point of view, this does not accord adequate recognition to Dionysius’ subtly universalising claims.
^[37] Sallust shows something of the same concerns in the BJ: D. Levene, ‘Sallust’s Jugurtha: an ‘historical fragment’’, JRS 82 (1992) 53-70.
^[38] Cf. Pol. 3.6.3 ff. with F.W. Walbank, Historical commentary on Polybius (Oxford 1957-79) ad loc.
^[39] Pomp. 6, Usener-Radermacher 2.245.1-5.
^[40] References are collected by A. Andrèn, ‘Dionysius of Halicarnassus on Roman monuments’, Hommages à L. Herrmann (1960) 88-104; A. Dubourdieu, ‘Denys d’Halicarnasse et Lavinium’, Pallas 39 (1993) 71-82.
^[41] See Walbank, ad loc.; F.W. Walbank, Polybius (Sather Classical Lectures, 42) (Berkeley 1972), 71ff.; F.W. Walbank, ‘Polemic in Polybius’, JRS 52 (1962) 1-12 = Selected papers: studies in Greek and Roman history and historiography (Cambridge 1985) 262-79.
^[42] Cf. Herkommer (1968) 91.
^[43] This neologism aims to express Dionysius’ ingenious inversion, in relation to Polybius, of the normal role of historiographical ‘continuator’ of some distinguished predecessor. Marincola (1997a), Appendices VI and VII, usefully lists such continuators.
^[44] E.g. the Tuberones: see G.P. Goold, ‘A Greek professorial circle at Rome’, TAPhA 92 (1961) 168-92; G. Bowersock, Augustus and the Greek world (Oxford 1965) 129-133; also section 7.1 below.
^[45] Professor Marincola kindly drew my attention to this parallel (and contrast), which he discusses in Marincola (1997a) 244-5.
^[46] I owe this point to Dr Kraus.
^[47] Similarly Cassius Dio (53.19.6); F. Millar, A Study of Cassius Dio (Oxford 1964) 32-4.
^[48] Marincola (1997a), Appendix IV, discusses the chief modes of adducing variants.
^[49] This can readily be seen from C. Jacoby’s Teubner edition (index scriptorum, vol. 5, 1-4) where the authors and documents are listed, with context.
^[50] O. Tomasini, ‘Per l’individuazione di fonti storiografiche anonime latine in Dionisio d’Alicarnasso’, AFLT 1 (1964-5) 153-74, lists these (157 n. 13) and discusses some of them. See 161-5 on identifying anonymous accounts as Greek or as Latin.
^[51] Employed, however, by Polybius for the purpose of criticism: 8.9.5 and 13 ff., of Theopompus, with Walbank ad loc.; 12.25h.1 and 12.26a.2-4, of Timaeus. As part of his sustained intertextual debate with Polybius Dionysius ‘turns’ this device by deploying such quotations positively.
^[52] Jacoby’s Teubner text restores Ionic forms in 1.28.3 (as suggested by G. Cobet, Observationes criticae et palaeographicae ad Dionysii Halicarnassensis Antiquitates Romanas (Leiden 1877) 26).
^[53] The same problem arises in relation to the obscure works cited by Dionysius when discussing the archaioi suggrapheis in Thuc. 5. This is a much disputed passage: according to Jacoby, Atthis: the local chronicles of ancient Athens (Oxford 1949) 86 and 178-85 (with 354, n. 13), Dionysius is following Theophrastus’ incorrect early dating of a number of local chroniclers (horographers, in Jacoby’s classification of the sub-genres of history); Jacoby’s implication is that Dionysius had not read (many /most of) these works himself. C.W. Fornara, The nature of history in ancient Greece and Rome (Berkeley 1983) 16-20, holds that Dionysius is arguing from the premise ‘simple style’ to the conclusion ‘therefore early date’, and that, while Dionysius’ argument is at fault, he seemingly was acquainted with the actual works. W.K. Pritchett, Dionysius of Halicarnassus: On Thucydides (Berkeley 1975), ad loc. and ad Thuc. 23 offers some counter-considerations to Jacoby’s criticisms; he accepts that Dionysius has personal knowledge of the authors mentioned. D. Toye, ‘Dionysius of Halicarnassus on the first Greek historians’, AJP 116 (1995) 279-302, holds that Jacoby was mistaken to classify as horographers the writers whom Dionysius was discussing : they were, rather, writers who dealt with genealogies, city by city. See also S. Gozzoli, ‘Un teoria antica sull’origine della storiografia greca’, SCO 19-20 (1970-1) 158-211 and L. Trojani, ‘Contributo alle problematica dei rapporti fra storiografia greca e storiografia vicino-orientale’, Athenaeum 61 (1983) 427-38 for a less sceptical view of the existence of local traditions and records mentioned by Dionysius as available to these archaioi suggrapheis. There is further material relevant to this debate in C. Joyce, ‘Was Hellanikos the first chronicler of Athens?’ HISTOS 3 (1999) (July 1999). 19 December 2000.
<http://www.dur.ac.uk/Classics/histos/1999/joyce.html> and in the subsequent exchange: L. Porciani and C.J. Joyce, ‘Exchange’, HISTOS 3 (1999) (February, 2000). 19 December 2000.
<http://www.dur.ac.uk/Classics/histos/1999/porcianijoyce.html>
On the generic questions involved see further Joyce’s paper and J. Marincola, ‘Genre, convention and innovation in Greco-Roman historiography’, in C.S. Kraus (ed.), The limits of historiography. Genre and narrative in ancient historical texts (Leiden 1999), 281-324.
On the debate concerning the Antiquitates, cf. e.g. D. Musti, Tendenze nella storiografia romana e greca su Roma arcaica. (Quaderni Urbinati di Cultura Classica 10) (Rome 1970) 11, 18ff., 26, drawing attention to Dionysius’ use of phrases denoting personal conviction; E. Gabba, ‘Dionigi e la “storia di Roma arcaica”’, Actes 9 Congrès Association G. Budé (1975) 218-29, esp. 222; Gabba (1991) 118 n. 54: ‘Philologically, his direct or indirect acquaintance with the sources cannot be demonstrated, as the diversity of critical opinion shows. Even the individual examination of each particular case does not lead to any firm conclusions; the very methods of Dionysius, whereby the information of his sources is arranged to follow his theories, obliged him to choose and incorporate those passages most pertinent to his own ideas. For my part I accept the sincerity of 1.6-7. At any event, there are no major contradictions between the views we find in Dionysius and our other evidence about the same authors.’ Others are more sceptical: Jacoby suspected some tralatician references: e.g. Satyrus in 1.68.2, perhaps from the work of Domitius Callistratus (see Jacoby ad FGH 20 F 1 and 433 F 10); Fehling (1989) 157 is, suo more, wholly disbelieving; contra, W.K. Pritchett, The liar school of Herodotus (Amsterdam 1993).
^[54] Allusions to Athens in Dionysius have a distinctly second-hand flavour. Veii is once (2.54.3) and Rome twice (4.13.5, 9.68.2) compared to Athens in respect of size; no attention is drawn to any distinctive features. In the literary-critical works, there are very few references to places in Athens; such as there are evidently derive from the biographers and Atthidographers whom Dionysius used. A reference to ‘the mysteries’ is no more than a metaphor (CV 25, Usener-Radermacher 2.124.2-3). R. Blum, Kallimachos. The Alexandrian Library and the origins of bibliography (Madison 1991) 199, suggests that Dionysius used the library at Pergamon for the work on Deinarchos (generally held to be his last: G. Marenghi, Dinarco (Milan n.d. [1970]), ad Din. 1; 66): this entails a return to Halicarnassus at a late stage in Dionysius’ career.
^[55] A.J. Marshall, ‘Library resources and creative writing at Rome’, Phoenix 30 (1976) 252-64; E. Rawson, Intellectual life in the late Roman republic (London 1985) 39-53; N. Horsfall, ‘Empty shelves on the Palatine’, G&R 40 (1993) 58-67; see also N. Horsfall, ‘Rome without spectacles’, G&R 42 (1995) 49-56, esp. n. 4 and the works by G. Cavallo (non vidi) cited there; V. Strocka, ‘Römische Bibliotheken’, Gymnasium 88 (1981) 298-329. See also DS 1.4.3.
^[56] Except for the wholly trivial omission of hoi before Plakienoi.
^[57] But identified as Korthonia in 1.26.1.
^[58] D. Briquel, Les Pelasges en Italie. Recherches sur l'histoire de la légende (Rome 1984), 101-40 (with full discussion of the many earlier views). See especially 126-7 for his textual conclusion, and 157-8 for the crucial role played by Cortona in Dionysius’ argument. Contra D. Asheri, La Lidia e la Persia. Erodoto. Le storie. Libro I. Testo e commento (Milan 1988), ad Hdt. 1.57.3.
^[59] H.H. Scullard, ‘Two Halicarnassians and a Lydian’, in E. Badian (ed.), Ancient society and institutions. Studies presented to Victor Ehrenberg (Oxford 1966) 225-31, esp. 226-7.
^[60] See also Hdt. 1.171.6; 4.45.3.
^[61] See Thuc. 5 and 23, with the modern discussions cited in n. [53] above.
^[62] Cf. e.g. Thuc. 10 and 13; see also W.K. Pritchett (1975) xvi-xvii; H. Veit Apfel, Literary Quotation and Allusion in Demetrius peri hermeneias and Longinus peri hupsous (Diss. Columbia, 1935), notes similar slight differences which indicate quotation from memory.
^[63] 7.72-3. 7.71.1, where Dionysius explains his grounds for using Fabius in this way, is discussed in section 7.1 below.
^[64] Cf. W. Rhys Roberts, Dionysius of Halicarnassus. On Literary Composition (London 1910) 51, on Dionysius’ skilful use of quotation. See also Strabo 1.2.3-9.
^[65] 1.28.3; 1.48.3. Shorter: 1.12.3; 1.13.1; 1.25.3; 1.28.2; 1.73.4.
^[66] 1.12.2; 25.4; 48.2; 49.2.
^[67] E. Gabba, ‘Mirsilo di Metimna, Dionigi e i Tirreni’, RAL 30 (1975) 35-49; Briquel (1984) 278-82.
^[68] Musti (1970), esp. ch. 1.
^[69] In this paper cf. such things as Dionysius’ credulous acceptance of the historicity of Cephalon of Gergis (n. 90], his judgement of Xanthus of Lydia [n. 95), his misconceptions about Graeco-Roman contacts in the third century and earlier, about Pictor’s history and about the nature of religious practice (n. 98] and the silent omissions re Terminus and Juventas, and death of Remus (n. 102].
^[70] Macer: 1.7.3; 2.52.4; 4.6.4; 5.47.3; 5.74.4; 6.11.2; 7.1.4. Antias: 1.7.3, 2.13.2. Tubero: 1.7.3; 1.80.1.
^[71] Cf. R. Jumeau, ‘Un aspect significatif de l’exposé livien’, in M. Renard and R. Schilling (eds.) Hommages à Jean Bayet. Collection Latomus 70 (Brussels-Berchem 1964) 309-33: ‘Tel est, en general, l’usage dans l’historiographie antique: on nomme l’auteur non de la version qu’on prefère, mais de celle qu’on n’a pas adoptée’ (326-7).
^[72] One of the best of the nineteenth-century Quellenforscher is O. Bocksch, ‘de fontibus libri V et VI Antiquitatum Romanarum Dionysii Halicarnassensis quaestiones variae’, Leipziger Studien (17) 165-274; see later A. Klotz, Livius und seine Vorgänger (Leipzig-Berlin 1940-1) and T.P. Wiseman, Clio’s cosmetics (Leicester 1979) esp. ch. 9.
^[73] Tomasini (1964-5) attempts various attributions but is over-inclined to regard factual similarity to a predecessor as indicating direct use.
^[74] 1.76.3, 1.77.1-2; 1.79.2 (2nd variant): Tomasini (1964-5) 165-9, based however on the assumed certainty of identification of 1.79.4 ff. and of Plut. Rom. 3 as Pictor.
^[75] 1.79.2 (1st variant): Tomasini (1964-5) 168-9; it might however be direct; see also 1.73.1, 1.34.4: Tomasini (1964-5) 169-70.
^[76] 1.31.1, 1.64.5: Tomasini (1964-5) 170-2; the latter case doubted by W.A. Schröder, M. Porcius Cato. Das erste Buch der Origines. Ausgabe und Erklärung der Fragmente (Meisenheim 1971) 116.
^[77] 1.56.3-4: Peter HRR 117; Tomasini (1964-5) 173-4; but L. Pepe, ‘L’annalista Q. Fabio Massimo Serviliano’, StudUrb 49.1 (1975) 95-108, denies direct dependence (98-101).
^[78] 5.35.2. The version reported by Pliny NH 34.29 (Piso fr. 20 P. = fr. 27 Forsythe) has the hostages dedicate the statue; Dionysius says the senate decreed it and the fathers of the hostages put it up. G. Forsythe, The historian L. Calpurnius Piso Frugi and the Roman annalistic tradition (Lanham 1994) 256 holds that these versions are not necessarily incompatible; cf. Tomasini (1964-5) 60 n. 21.
^[79] 1.55.2: E. Maass, ‘Tibullische Sagen’, Hermes 18 (1883) 322-42, esp. 334-6; J. Perret, Les origines de la légende troyenne de Rome (281-31) (Paris 1942) 603 ff.; direct use, however, is not certain.
^[80] Gabba (1991) 97-107 considers that Dionysius made extensive use of Varro, though rejecting some of his theoretical anthropological models. The issue of Pallas 39 (1993), entitled Denys d'Halicarnasse: historien des origines de Rome. Actes du colloque organisé à l’Université Paul Valéry (Montpellier III) 20-21 Mars 1992 includes several relevant papers: D. Briquel, ‘Denys d’Halicarnasse et la tradition antiquaire sur les Aborigènes’, 17-39 (DH reliant on V.); G. Capdeville, ‘Les institutions religieuses de Rome selon Denys d’Halicarnasse’, 152-72 (DH reliant on V.); J. Poucet, ‘Varron, Denys d’Halicarnasse, Macrobe et Lactance. L’oracle rendu à Dodone aux Pelasges’, 41-69 (DH not greatly influenced by V.); also J. Poucet, ‘Denys d’Halicarnasse et Varron: le cas des voyages d’Énée’, MEFRA 101 (1989) 63-95 (little influence). See also the works cited in Musti (1970), 26 n. 10; B. Cardauns, M. Terentius Varro. Antiquitates Rerum Divinarum (Mainz 1976), 127, 130; frr. 40-1 60, 73, 40-41, 205-7, 214, with commentary. Note, however, as a striking example of non-use of Varro, 2.18.2 on statues to gods: contrast Aug. CD 4.31 (fr. 18 Cardauns) on aniconic worship. P. Bourgeaud, ‘Quelques remarques sur la mythologie divine à Rome, à propos de Denys d’Halicarnasse (ant. Rom. 2.18-20)’, in F. Graf (ed.), Mythos in mythenloser Gesellschaft. Das Paradeigma Roms (Stuttgart 1993) 175-87, discusses Dionysius’ reaction against Varronian ideas about religion.
^[81] J. Poucet, ‘Fabius Pictor et Denys d’Halicarnasse: Les enfances de Romulus et Remus’, Historia 25 (1976) 201-16. For this whole complex tradition see especially C.J. Classen, ‘Die Herkunft der Sage von Romulus und Remus’, Historia 12 (1963) 447-57; T.J. Cornell, ‘Aeneas and the twins: the development of the Roman foundation legend’, PCPhS 21 (1975) 1-32; and, above all, T.P. Wiseman, Remus (1995), esp. chs. 4 and 7.
^[82] As at 11.1.
^[83] See P. Veyne, Did the Greeks believe in their myths?: an essay on the constitutive imagination, tr. P. Wissing (Chicago 1988) 5-6.
^[84] 1.11.1; 1.12.3; 1.13.1; 1.19.3; 1.28.2; 1.68.2; 1.72.3; 1.74.1; 2.38.3; 2.39.1; 2.49.1; 7.71.1; 12.4.2; 12.9.3; 20.10.2.
^[85] 1.11.1; 1.14.1; 1.25.4; 1.28.3; 1.41.3; 1.48.1 bis; 1.49.1; 1.53.4; 1.68.2; 1.72.1; 1.74.2; 2.21.2; 4.7.5; 4.15.5; 4.62. 6; 5.73.3; 12.9.3. See also 4.30.3 (probably Piso’s work); 8.56.1; 10.1.4; 11.62.2.
^[86] 1.12.3; 1.49.1; 1.61.5; 1.68.2; 1.72.5.
^[87] Thus 1.22.5 (Thucydides) and 1.29.3 (Herodotus).
^[88] Compare DS 4.56; Livy 1.44.2 (contrast 1.55.8-9), 2.18.5. Other references: R.M. Ogilvie, A commentary on Livy Books 1-5 (Oxford 1965), 7.
^[89] M. Chassignet’s recent edition of the fragments of the Roman annalists surveys the state of the question regarding the tablets: L’annalistique romaine. t. 1. Les annales des pontifes et l’annalistique ancienne (Paris 1996), xxiii-xlii; she does not believe that the traditions Dionysius proceeds to mention come from the pontifical chronicle as such (xxxix with n. 126). On the derivation of traditions from the hierai deltoi see E. Gabba, ‘Considerazioni sulla tradizione letteraria sulle origini della Repubblica’, in Les origines de la république romaine. (Entretiens Fondation Hardt 13 (Vandoeuvres-Genève 1967) 133-74 at 153. See also B.W. Frier, Libri annales pontificum maximorum. The origins of the annalistic tradition. Papers and Monographs of the American Academy in Rome 27. (Rome 1979) 109-10, 305-6. (The second edition (Ann Arbor 1999) includes a new introduction but the main text and pagination are unchanged except for minor corrections.)
^[90] Cephalon (FGH 45) is also called suggrapheus palaios at 1.72.1: Dionysius was not alone in accepting as reliable this ‘source’ invented by Hegesianax.
^[91] While the kai could be understood as merely epexegetical, there is a play on logou axioi as ‘worthy of repute, worthy of account’ and as ‘worthy of inclusion in my account’.
^[92] palaios: 1.48.1; palaiotatos: 1.68.2; 1.71.1; 5.17.3; archaios: 1.12.3; 1.13.1; 1.34.4. See also Jacoby ad FGH 391 T 2 (Komm. 189, Noten 123 n. 4). Cf. Livy 2.18.4-7; 3.23.7.
^[93] Piso: ho timetikos (2.38.3; 2.39.1; 12.9.3); Cincius (1.74.1). On the other hand, Cato is not specified as a censor (possibly because so well known?); Fabius Pictor, Licinius Macer, Cn. Gellius, Aelius Tubero, C. Sempronius Tuditanus are of course also senators, though not so described by Dionysius.
^[94] For the identification of this man as L. Manlius, Sulla’s proquaestor in 84 BC, see T. Mommsen, ‘Mamilius Sura, Aemilius Sura, L. Manlius’, RhM 16 (1861) 282-7: his work was a ‘Reise- und Wunderbuch’; see also Varro, LL 5.31. J. Perret, Les origines de la légende troyenne de Rome (281-31) (Paris 1942) 578-617, thinks that Dionysius used Polyhistor (603-6) in addition to Varro (607-17). The passage at 1.19.3 is closely related in subject matter to 1.51.1: Poucet (1993) discusses the latter passage (78-81) as part of his attempt to refute the view that Dionysius used Varro extensively for the voyage of Aeneas; Poucet does not attempt to identify Manlius. See also F. Della Corte, La mappa dell’ Eneide (Florence 1972), 60-68; 121-36.
^[95] FGH 765; Musti (1970) 16-17; K. von Fritz, Die griechische Geschichtsschreibung (Berlin 1967) 1. Anmerkungen, 348-77.
^[96] Gabba (1991) 85-9.
^[97] 1.67.4; 2.49.4; 4.2.1; and esp. 8.56.4. On Greek attitudes to epichorioi see Marincola (1997a) 283-5; H. Verdin, ‘Notes sur l’attitude des historiens grecs à l’égard de la tradition locale’, AncSoc 1 (1970) 183-200.
^[98] J.-P. Thuillier, ‘Denys d’Halicarnasse et les jeux romains (Antiquités romaines VII 72-73)’, MEFRA 87 (1975) 563-81; Frier (1979) 242-3 summarises the elements contemporary with Pictor; Gabba (1991) 134-6.
^[99] E.J. Bickermann, ‘Origines gentium’, CPh 47 (1952) 65-81.
^[100] Musti (1970) 64 (with n. 95) thinks this may be a purely local tradition.
^[101] See also 4.21; 4.7.1; 7.1.4; 9.21.6. At 5.47.2 what Dionysius conjectures is supported by ‘many native accounts’.
^[102] The main difference is over the dating: under Priscus in AR 3.69.5, and probably in Varro (ap. Aug. CD 4.23 fr. 40 Cardauns); under Superbus according to Livy 1.55.3; Florus 1.1.7-9; Serv. Aen. 9.446; no date in Cato fr. 24P; Ov. F. 2.665ff. Of the deities involved, Terminus will be the original version, given the riddle element of its refusal to move; Livy has Terminus alone at 1.55.3-4 (but see L. 5.54.7 with Ogilvie ad loc.); Dionysius (like Florus) has Juventas too, no doubt from Varro; however, he (or his source?) has rejected Varro’s third deity, Mars. On the related traditional elements associated with the Tarquinii and the Capitoline, Dionysius’ versions of Attus Navius, the caput, and the Sibylline books again show considerable discrepancies of detail from other accounts, which he does not remark upon; on the other hand he does not assert, as he does at 3.69.3, that the tradition is unanimous.
^[103] Schultze (1995), 196, 199.
^[104] See especially Walbank (1962); (1972) 49-55.
^[105] Note, however, that O. de Cazenove, ‘La chronologie des Bacchiades et celle des rois étrusques de Rome’, MEFRA 100(2) (1988) 615-48, has demonstrated just what an historically inaccurate construct the Pisonian genealogical table is.
^[106] See Schultze (1995) 199-200 and the works there cited in n. 46.
^[107] Recognised as chronologically impossible well before Dionysius’ time: see Gabba (1991) 13.
^[108] 1.32.1, where the absence of a tomb of Pallas disproves the account of Polybius and others unnamed; 5.74.4, where Dionysius rejects Licinius’ account of the origin of the dictatorship, claiming that he himself is treating the substance and not merely the name.
^[109] On some occasions, stylistic variatio seems to account for the selection of a different expression: e.g. 1.10.1-3; 1.31.1; 4.15.1.
^[110] 1.14.1; 1.22.4; 1.22.5; 1.23.5; 1.23.5; 1.49.1; 1.79.4; 1.80.1; 2.21.2; 2.38.3; 2.40.3; 2.45.2;.2.48.4; 2.52.4; 2.72.2; 7.1.4; 20.10.2. Anonymous use is neutral or slightly disapproving: 1.52.4; 2.13.2; 2.31.1; 2.47.3; 2.59.1; 2.76.5; 3.62.1; 5.18.1; 5.73.1; 12.4.2. See also Tomasini (1964-5) 155.
^[111] Numerous examples: e.g. legousi 1.43.1; 4.40.1; 9.22.1; legetai 1.53.2; 2.68.3; 9.13.1; phasin 1.53.4; 2.8.2; 5.13.4.
^[112] 1.27.1; 1.30.5; 1.33.1; 1.36.1; 1.42.2; 1.49.2; 1.54.1; 1.77.2; 2.60.5. See also 1.40.6.
^[113] 1.22.5; 1.26.2; 1.28.1; 1.28.4; 1.30.2; 1.48.3; 1.67.4; 1.72.6; 1.73.1; 1.74.2; 1.77.1; 1.79.1; 2.61.2; 2.66.5; 7.1.4 (disapprovingly). See also Tomasini (1964-5) 168 n. 53.
^[114] 1.23.5. 1.32.1. 1.72.3; 1.75.4; 1.86.2; 2.5.5; 2.8.3; 2.38.3; 2.76.5; 4.6.4; 4.7.5; 4.15.1; 4.15.5; 4.62.6; 5.2.1; 5.47.3; 5.73.3; 8.56.2. See also Tomasini (1964/5) 168 n. 55.
^[115] 1.11.1; 3.67.5; 4.64.3; 7.71.1.
^[116] 1.12.2; 1.13.1; cf. tekmairomai (‘prove’) and cognates, 1.53.1; 3.67.5; 7.71.1.
^[117] 1.30.4; 1.62.1-2; 1.63.3, 4.64.3; 5.23.3 (self), cf. 1.54.3; others ; 1.22.5; 1.41.3.
^[118] 1.13:4; 1.35.3; 1.48.1; 1.75.4; (cf. 1.79.3); 1.87.4; 2.8.3; 2.45.2; 2.56.3; 8.79.1; (cf. 9.19.3; 9.22.1); 12.4.2. Cf. 2.31.1; 3.35.5.
^[119] 1.41.1; 5.31.2; 9.20.1; cf. 1.79.1 and 3; 3.35.5.
^[120] 1.39.1; 1.79.1. See also 1.48.1 and 4; 1.84.1.
^[121] As does Livy at the resumption of his work at 6.1.
^[122] M. Fox, ‘History and rhetoric in Dionysius of Halicarnassus’, JRS 83 (1993) 31-47 calls this ‘a bizarre rationalisation of myth’ (44).
^[123] Again, like Livy: Moles (1993) 148-9.
^[124] Note the play on the ambiguities of logos and muthos in 1.36.2.
^[125] It is possible that Dionysius has Dicaearchus, Bios Hellados fr. 49 Wehrli in mind. See Wiseman (1979) 49 with n. 45; contra, Gabba (1991) 78, 100-1. Cf. B. Gatz, Weltalter, goldene Zeit und sinnverwandte Vorstellungen (Hildesheim 1967) 123.
^[126] Cf. his attitude to Thucydides: Thuc. 2, esp. Usener-Radermacher 1.327.16ff. Livy’s attitude is similar: T.J. Luce, Livy: the composition of his history (Princeton 1977) 145.
^[127] 1.59.3; 1.77; 1.79.3; 2.31.3; 2.60.4-61.3; 2.72.2; 3.61.2; 6.1.4. Cf. Luc. quomodo hist. conscr. 60. See also S. Ek, Herodotismen in der Archäologie des Dionys von Halikarnass (Lund 1942) 5-6, for the important influence of Herodotus in Dionysius’ presentation of variants.
^[128] 1.45.4; 1.69.4; 1.90.2; 8.79.1; 9.21.6; cf. 1.75.4. See also 11.62.3.
^[129] It is noticeable that at 12.4.2-4 he likewise prefers a more formal and official version of a condemnation. For the Cassius story, see E. Gabba, ‘Studi su Dionigi da Alicarnasso. III. La proposta di legge agrarie di Sp. Cassio’, Athenaeum n.s. 42 (1964) 29-41; id., ‘Dionigi d'Alicarnasso sul processo di Spurio Cassio’, in La storia del diritto nel quadro delle scienze storiche. Atti del I. congresso internazionale della società italiana di storia del diritto (1966) 143-153; for Maelius, A. Valvo, ‘Le vicende del 44-43 a.C. nella tradizione di Livio e di Dionigi su Spurio Melio’, CISA 3 (1975) 157-83.
^[130] Livy 2.41.10-11 similarly accepts the popular trial version rather than the family concilium one.
^[131] Wiseman (1979) 48 ff. on to eikos (‘probability’); H.D. Westlake, ‘Hos eikos in Thucydides’, in Essays on the Greek historians and Greek history (Manchester 1969) 153-60. See also Marincola (1997a) 282-3 and Luce (1977) 141.
^[132] 1.30.1-2; 1.54 (plus monuments); 2.31.1; 2.64-5.
^[133] In contrast to some fifty in Books 1-4: 5.18.1 (implicit); 5.31.2; 5.74.4; 6.1.4; 6.4.1; 6.11.2; 7.1; 8.79; 9.19-22; 11.62; 12.4.2-4. Of these, the last four at least have been well worked over previously.
^[134] Wiseman (1979), Part 2.