Dio and the annalistic tradition

1Recent research has done much to remedy the neglect of Dio’s fragmentary books 1-351. More remains to be done, and in this chapter I consider one aspect which has so far received little attention, namely Dio’s use of annalistic organization in this part of his history2.

2Dio opted to write in the traditional format pioneered by the first Roman historian, Fabius Pictor, recounting the deeds of the Roman people at home and at war from Aeneas to his own day. Such Roman histories, and also those dealing with shorter periods, were customarily organized by the consular year. Fabius Pictor and his immediate successors will have used such organization at least for the recent past, and from Piso in the later second century on such histories will have given a year-by-year account from the foundation of the Republic3.

3Livy in Books 21-45, covering the years 218-167 BC, follows a standard pattern for his year-narratives. Each year typically opens with the consuls’ entry into office and events at Rome up to their departure for their provinces; Livy then passes to the provincial activity of the consuls and other commanders, and returns to Rome (with one or both consuls) to narrate the year end. The domestic sections include much detail on routine matters such as the allocation of provinces and armies and the expiation of prodigies, information which must derive ultimately from archival sources. It is commonly supposed that this pattern was characteristic of all Roman annalistic historiography, but this is a misconception4. Livy probably took it from Valerius Antias, and the annual narratives of most of his predecessors are likely to have been more varied and informal, as were those of Livy himself for the early Republic in his first decade. From the outbreak of the Social War in 91, Livy will have been obliged to make radical modifications to the structure of his annual narratives to fit the changed character of events.

4In his extant books, Dio’s material is organized by consular years throughout, but this annalistic structure is deployed with considerable flexibility. On the late Republic and the civil war period, Dio sometimes opts to narrate the activity of a prominent individual over several years as a continuous bloc, although this results in chronological backtracking elsewhere5. For Augustus and his Julio-Claudian successors, the events of each reign are still narrated by consular years, but, by a device which was surely Dio’s innovation (perhaps influenced by Suetonius’ biographies), these annalistic accounts are set in a biographical frame: Dio opens each reign with an account of the emperor’s accession and the character of his rule and closes it with a further summary assessment6. This arrangement probably continued in Dio’s later books: little trace of their structure is preserved in the epitomes and fragments on which we depend, but the survival of his full text for most of AD 217-218 demonstrates Dio’s continued use of consular dating, since here, as in his earlier books, he marks the new year by naming the consuls (79[78].26.8)7. However, for full understanding of Dio’s use of annalistic organization, we need to consider also how he will have deployed it in the lost early books which dealt with the Republic from its foundation.

Books 1-35: fragments and Zonaras

5Dio’s first 35 books are preserved for us mainly in verbatim fragments and through Zonaras’ Epitome of Histories. There are also a few passages in which statements by Dio are paraphrased by other authors, principally John Tzetzes.

6The fragments are conventionally numbered fr. 1-111. This numeration, established by Bekker and retained (despite some rejections and order changes) by Boissevain in his still standard edition, groups many of the fragments together by topic, so that, for example, the 27 fragments relating to the First Punic War are all assigned to fr. 43 and the 53 fragments relating to the Second Punic War to fr. 57. The verbatim remnants of Books 1-35 in fact comprise 389 separately cited fragments8.

7All but four of the verbatim fragments have been preserved (often with minor changes) as extracts in Byzantine anthologies or lexica, with some fragments being cited by two or more sources. The four exceptions comprise portions of Dio’s accounts of the years 207-206, 204, 201 and 200 surviving on parchment leaves re-used as backing for a Strabo manuscript now in Paris9. Since the extracts all relate just to single episodes, these four Paris fragments, despite their poor state of preservation, have special importance for the present enquiry: they provide our only direct evidence for Dio’s handling in his early books of year-narratives and transitions between regions and topics.

8307 fragments from these books are cited in the collections of historical extracts made for Constantine VII (four in two collections). Within these collections fragments are presented in the order in which they stood in the cited author. Three of the surviving collections contain Dio fragments, namely those dealing with embassies, virtue and vice, and γνῶµαι (“maxims”), now commonly known by their latinized titles as Excerpta de legationibus, de virtutibus et vitiis and de sententiis (hereafter EL, EV, ES). Table 1 gives totals for these fragments across the historical periods covered by Books 1-3510. The excerptor on embassies made comparatively little use of Dio, but found most material on the later periods in which diplomacy was more prominent, while the turbulence of the years from the Gracchi to Sulla provided rich pickings for the excerptor on virtue and vice11. The extracts on γνῶµαι, where they survive, provide a fuller selection, drawing on a wide range of passages. Unfortunately, this collection survives only in a mutilated Vatican palimpsest: there are two four-page gaps in its Dio extracts, and they break off altogether in late 216 BC12. However, its surviving selection is rich enough to give some indication of the pattern of Dio’s own coverage.

9Our other main sources for verbatim fragments of Dio’s early books are the so-called Florilegium of Maximus the Confessor and lexica citing short passages to illustrate word usage. 61 citations in the lexicon Περὶ συντάξεως and seven citations in the Συναγωγὴ χρησίµων λέξεων report the book in which the cited passage stood in Dio’s history, and so provide our evidence for Dio’s book division13.

10Zonaras used Dio as his principal source for Roman history down to 146 BC. He also drew on Plutarch’s lives of Romulus, Numa, Publicola and Camillus, but thereafter used Dio exclusively, except for a citation of Plutarch on Aemilius Paullus’ reception of Perseus14. Zonaras returned to Dio from the death of Caesar, and comparison with Dio’s extant text illustrates his practice. Although abridging throughout, he often reproduced Dio quite fully, borrowing much of his language. Elsewhere he omitted substantial amounts of material or summarized it thinly, sometimes with considerable distortion. Like the Constantinian excerptors, he invariably adhered to Dio’s order of events15^.

11Zonaras’ account of the early and middle Republic frequently takes an annalistic form: he quite often notes the transition to a new year, and mentions numerous consuls, either in connection with the year change or to report their activities. Where such features occur, Zonaras will have taken them over from Dio, thus providing evidence for Dio’s use of annalistic structure. When, however, Zonaras’ narrative lacks such dating indications, sometimes passing over extended periods altogether or with only a sketchy summary, such features may merely result from Zonaras’ epitomizing, but may, alternatively, reflect Dio’s original.

Book structure and coverage

12In his edition Boissevain arranged the fragments of Dio’s Books 1-29 and (for Books 1-21) the corresponding portions of Zonaras according to his own reconstruction of the book division, slightly modified from the earlier scheme of Gutschmid (1894, 554-562)16. This arrangement has been retained in subsequent editions and translations, and many scholars have taken it as firmly established, sometimes including its book allocations in their fragment citations. This has been a pernicious development, which overlooks the very limited extent of our information17. Boissevain himself (1895a, lvi) conceded that his division was no more than probable and that the precise points at which Dio started a new book could not always be established.

13As noted above, our evidence for the division of Books 1-35 is provided by the book attributions accompanying citations in the lexicon Περὶ συντάξεως and the Συναγωγή. For only a minority of these brief fragments can the context be securely identified, either from their own content or through verbal overlap with other citations or with Zonaras. Moreover, book numbers are particularly vulnerable to scribal error, and, as Boissevain himself recognized, this was particularly the case for the lexicon Περὶ συντάξεως: 21 of its 57 citations from Dio’s extant books give false book numbers18. Thus this evidence can only be used with complete confidence to establish the book in which Dio dealt with any particular set of events when two or more fragments with cited book numbers provide corroboration.

14Gutschmid’s and Boissevain’s schemes depend heavily on the unsupported evidence of single citations and on highly conjectural identifications of fragment context, and many of their identifications of terminal points for Dio’s books are mere guesses. Their book divisions thus go far beyond what the evidence warrants.

15For Books 22-35 we have virtually no usable evidence. The lexicon Περὶ συντάξεως cites ten fragments as from these books, but only one, attributed to Book 31, has an identifiable context: containing the name Mithridates, it clearly relates to the First Mithridatic War19.

16For Books 1-21 we are better informed, as shown in Table 2. The second column lists the periods assigned to each book by Boissevain. The third column gives details of the fragments attributed to these books whose context can, in my view, be established with confidence. The remaining fragments are listed in the fourth column. The symbol ‘Σ’ indicates the citations in the Συναγωγή20. The table demonstrates the inadequate basis for Boissevain’s book divisions. However, our information does permit us to draw some conclusions about how Dio distributed his material across these books, and, despite its undue precision, the broad implications of Boissevain’s scheme remain plausible.

17Numerous citations show that Dio devoted his first two books to the pre-republican period and all of his third book to the Republic’s first year, but we have only scattered indications for the rest of the first decade.

18The First Punic War is attested as extending across Books 11-12, with, if the citations are accurate, Book 12 starting with the speech of the captive Regulus to the senate in 25121. The Suda in its biography of Dio states that his eighty books formed decades, and it is likely that the second decade opened with the origins and outbreak of the First Punic War: Zonaras (8.8.1) shows that Dio stressed this moment as the start of Rome’s overseas conflicts.

19Books 15-16 are well attested as covering the middle years of the Second Punic War. Zonaras and the fragments show that Dio gave both the origins and the course of the war extended coverage, and so it is likely that, as the citations suggest, Books 13-17 were devoted to the war.

20Fr. 71.2 attests an event in the first year of the Third Punic War as from Book 21, and it is plausible to suppose that Dio opened his third decade with the origins and outbreak of that war. It would then follow that Dio allotted only the three books 18-20 to the first half of the second century.

21The implications of these findings for Dio’s book distribution and annual coverage are set out in Table 3, with comparison with Livy22. Except for pre-republican times, Livy allots each period more books than Dio, but their relative coverage is broadly consistent, except on the early second century, to which Dio assigns a much smaller share of his space than Livy. Both writers are relatively full on the Second Punic War and on the troubles of the late Republic. Livy increased his coverage as he approached his own time, as most of his predecessors had done from Fabius Pictor on23. Dio too accorded most space to the civil wars of the late Republic and the transition to monarchy, but, like Appian, he reverted to briefer treatment for the imperial period.

22Comparison of Dio’s and Livy’s coverage must take account not only of numbers of books, but also of book size. Dio’s extant books are among the shortest of ancient prose books, and are on average around half the length of Livy’s, and this no doubt held good also for the lost books24. Thus for most periods Livy’s coverage was around four to five times more ample than Dio’s, while for the early second century it may have been as much as twelve times longer.

23Early in his history Dio announced his intention to be selective, claiming that he had read virtually everything written about the Romans, but had written up “not everything but what I selected” (fr. 1.2), and that he aimed to include “everything which had been done by the Romans in peace or war deserving of mention” (πάνθ’ ὅσα τοῖς Ῥωµαίοις καὶ εἰρηνοῦσι καὶ πολεµοῦσι ἀξίως µνήµης ἐπράχθη), omitting “nothing indispensable” (µηδὲν τῶν ἀναγκαίων) (fr. 1.1). Later in the work he not infrequently notes that he has omitted some information, retaining only what was “indispensable” or “worthy of mention”25. We must consider whether, for the periods of Republican history to which he accorded relatively brief coverage, Dio may at some points have abandoned year-by-year treatment altogether, on the grounds that the years in question had seen no events deserving to be recorded.

Books 3-10: 509-265 BC

24As we saw above, Dio probably devoted just eight of his comparatively short books to the long period from the foundation of the Republic to the eve of the First Punic War, an average of thirty years per book26. Marked disparities in his coverage can be traced within the period. It is unlikely that he could have covered it within this limited space, with the fuller treatment he considered necessary for numerous topics, without at some points achieving compression by dispensing with year-by-year narration.

25For this as for later periods, Dio’s main sources will have been earlier annalistic histories. Although his claim to have read virtually everything will have been a gross exaggeration, he is likely to have drawn on several accounts. For this period he probably consulted Livy, and possibly also Dionysius, but will certainly have used some earlier accounts as well, since his version of events is frequently in conflict with theirs27.

26As already noted, cited book numbers show that Dio devoted the whole of his third book to the first year of the Republic, evidently using it for an extended exploration, largely through speeches, of the Roman choice of δηµοκρατία (i.e. republican government), just as he was to do in Book 52 for their return to monarchy.

27Zonaras and the fragments suggest that Dio continued to provide fairly full treatment down to the mid fifth century, including minor events as well as the major episodes28. Boissevain’s conjecture that Dio devoted all of Books 4 and 5 to the years 508-449 is thus not implausible. Political conflict and constitutional innovation continued to dominate his account, and Dio inserted excursuses on individual magistracies when reporting their creation, perhaps drawing on a constitutional treatise29.

28At some points over this period Zonaras narrates events over successive consular years. Naming one or both, he reports the consuls of 503-502 as engaged with Sabine wars and those of 501-500 dealing with slave plots, and later recounts the mixed fortunes of the consuls of 481-480 against the Etruscans30. Zonaras passes from the consulship of Valerius and Horatius which ended the Second Secession to the institution of the consular tribunate and censorship, naming the consuls in each case31. He wrongly represents these events as happening in successive years (the consular years were in fact 449, 445 and 443), but this is probably his rather than Dio’s error. These signs of consular dating, together with the relative fullness of coverage, suggest that Dio may have given a year-by-year account for most, and perhaps all, of the period from the establishment of the Republic to the institution of the censorship (in the conventional reckoning, 509-443).

29After that point Dio’s account may have become more selective. For the next half-century Zonaras recounts only three episodes: the suppression of Sp. Maelius, the mutiny against Postumius (dated by Livy to 414), and Camillus’ wars against Veii and the Falisci (the former prompting another institutional excursus, on the triumph)32. Dio no doubt had more to report33. However, he may well not have provided a year-by-year narrative for this relatively uneventful period, for which just naming the chief magistrates would for many years involve listing multiple consular tribunes. Dio perhaps used the institution of the consular tribunate as an initial justification for passing over some magistrate years.

30For the fourth century what survives of Dio’s account also suggests selective treatment. Zonaras (7.23-26; 8.1.1) confines himself almost exclusively to a few famous episodes like the Gallic Sack, the Licinio-Sextian agitation and the Caudine Forks disaster. The fragments are mostly concerned with the same episodes, but add a few further incidents. Dio will have reported a good deal more, but it seems likely that here, as for the late fifth century, he did not attempt a year-by-year narrative.

31Dio will, however, have resumed annalistic coverage for the Third Samnite War. Zonaras reports the activities of one or both consuls for 295, the year of Sentinum, and for the following years up to 29234. Fragments cover these years and also 296 and 291-290. Dio probably gave a year-by-year account of this war from its outbreak in 298 to the peace settlement in 290.

32The only event mentioned for the immediately following years in Zonaras and the fragments is the debt crisis of 287, and Dio may have left some of these apparently uneventful years unmentioned, passing straight on to the resumption of widespread warfare c. 283. From that point it is clear that Dio gave a year-by-year narrative. Copious fragments and Zonaras show that Dio gave a full account of the Roman war with the Tarentines and Pyrrhus from its origins in 282 to Pyrrhus’ withdrawal to Sicily in 27835. Dio’s treatment of the following years will have been less ample, but, although Zonaras only mentions some of the consuls and some of his chronological indications are misleading, his account makes it clear that Dio gave a year-by-year narrative for 277-265, mentioning some domestic events and covering Roman campaigns including the final defeat of Pyrrhus in 275 and the completion of the conquest of southern Italy36.

33Thus it seems likely that Dio gave a year-by-year narrative for most, if not all, of the years from the foundation of the Republic to the institution of the censorship (509-443) and again from the outbreak of the Third Samnite War in 298 to 265, but that for the intervening period his treatment was more selective, concentrating on the episodes he regarded as worthy of record and passing over many years unnarrated. This pattern was no doubt reflected in his book distribution, with the years 442-299 perhaps covered in just Books 6-7 and the first half of Book 8 (an average of 55 years per book)37.

Books 11-12: 264-219 BC

34Dio, as we have seen, probably devoted Books 11-12 to the First Punic War (264-241) and the following years up to the eve of the Second Punic War, with Book 12 starting with Regulus’ speech in 251. As before, Dio’s sources will have been earlier annalistic historians, and the loss of Livy means that what survives of Dio’s account provides our fullest representative of the Roman tradition38.

35Except for one phase, Dio evidently gave an annalistic narrative of the First Punic War. Zonaras and the fragments show that he treated the origins and the opening year of the war in considerable detail39. Zonaras then gives a year-by-year account, reporting the consuls’ activity for each year down to 247 and naming them (sometimes inaccurately) for all but two years40.

36The years before the Romans’ launch of a new fleet in 242 and their decisive victory at the Aegates Islands in 241 were a quiescent phase. After reporting the activity of the consuls of 247, Zonaras (8.16.7) tells us that “after this various men held the consulship, but achieved nothing worthy of record” (ἔκτοτε δὲ διάφοροι µὲν ὑπάτευσαν, οὐδὲν δὲ ἱστορίας ἔπραξαν ἄξιον), and then goes on to remark that the Romans suffered from their practice of changing commanders annually, since they lost their post just when they were learning generalship. Then, after reporting unrest among the Carthaginian mercenaries, he passes to the activities of the commanders of 242-241 and the end of the war. Zonaras must have taken his observations on the lack of Roman achievement in 246-243 from Dio: this was Zonaras’ way, and, as we have seen, it was characteristic of Dio to justify omitting events by declaring them unworthy of record. Thus this passage shows that Dio dispensed with year-by-year narration for these years, although, unlike Zonaras, he may still have named the consuls.

37Between the First and Second Punic Wars the Romans were occupied with wars in northern Italy, Sardinia, Corsica and Illyria. Zonaras (8.18.2-20.13) gives (mostly brief) reports of this warfare over the years 238-229 and 225-219, naming the consuls for each year except 224. Dio will thus have given a year-by-year narrative at least for these years, but may have passed over some of the other years for which no warfare is reported. Zonaras and the fragments preserve nothing for 240-239, and for 228-226 mention just the Romans’ live burial of Greeks and Gauls (dated to 228 by Tzetzes, following Dio) and an undated Sardinian revolt41.

Books 13-17: the Second Punic War (218-201 BC)

38Zonaras and the fragments show that Dio gave the Second Punic War extended coverage by comparison with other Republican periods, and, as we saw above, his treatment probably occupied all of Books 13-17.

39The availability of Livy and much of Polybius sheds some light on Dio’s sources for the war42. Livy made extensive use of Polybius, and his other major sources for the war included Coelius Antipater’s monograph on the war and Valerius Antias43. Dio often follows versions of events which conflict with Livy or which Livy noticed only as variants. He probably used both Coelius and Antias, but must have drawn also on Livy himself.

40A good deal survives of Dio’s extended account of the origins and outbreak of the war, which must have occupied much of Book 13 (though surely not all, as Boissevain supposed)44. For the course of the war we have the (for him) lengthy narrative of Zonaras (8.23-9.14) and numerous fragments. Zonaras’ account makes it clear that Dio narrated the war by consular years, moving within the year-narratives between the various theatres of war. Zonaras sometimes mentions the election of consuls, and for most years he names them and reports their activity, along with that of other notable commanders. Only once does his account become heavily compressed, when a very brief account of events in other theatres in 213-212 is inserted between two Sicilian sections (9.5.1-2).

41The treatment of events in some overseas theatres presented problems for Dio as for his predecessors, as to how long a sequence to include as a single section and under which year to place it. On Sicilian and Greek events Dio’s solutions are similar to Livy’s, and for some of the Greek sections he must be following Livy: under 208 and 207 Livy gave reports of the Roman warfare against Philip V based on Polybius’ accounts for the years 209 and 208, and Dio gave briefer versions on the same lines and with the same chronological displacement45. On Scipio’s command in Spain, however, Dio departed radically from Livy. He narrated all the events from the capture of New Carthage to the battle of Ilipa as a single section under the year 209, so bringing together events which Livy reported in four separate sections and which in fact took place over four years (the duration is obscured by Zonaras, but no doubt Dio was clearer)46. A later section reports under 207 the end of Scipio’s command (in fact in 206), with striking divergences from Livy (the romance between Masinissa and Sophonisba dated to this early point, and a claim, unattested elsewhere, that Scipio was recalled out of jealousy)47.

42As noted above, the first three Paris fragments preserve, in a somewhat mutilated state, portions of Dio’s account of the later years of the war, and provide an opportunity, unique for the lost early books, to observe his handling of his year-narratives.

43The first Paris fragment covers the end of Dio’s account of the year 207 and the beginning of 206. The fragment opens in the misdated account of the conclusion of Scipio’s Spanish command (fr. 57.53-56), and then passes, with the linking words “in those same times” (ἐν δὲ τοῖς αὐτοῖς τούτοις χρόνοις) to events in Greece (fr. 57.57-59). This section closes with Philip’s separate peace with the Aetolians (an addition by Dio to the Livian section he had previously been following). Dio continues with a linking passage making the transition to the next consular year: “Nothing was achieved worthy of mention by them or any others, either then or in the next year in which Lucius Veturius and Caecilius Metellus were consuls, although many prodigies occurred for the Romans” (οὐ µέντοι καὶ ἐπράχθη τι µνήµης ἄξιον οὔθ’ ὑπ’ ἐκείνων οὔτε ὑπ’ ἄλλων τινῶν, οὔτε τότε οὔτε ἐν τῷ ὑστέρῳ ἔτει ἐν ᾧ Λούκιός τε Οὐετούριος καὶ Καικίλιος Μέτελλος ὑπάτευσαν, καίπερ σηµείων πολλῶν καὶ δυσχερῶν τοῖς Ῥωµαίοις γενοµένων). A list of the prodigies follows, in the course of which the fragment breaks off (fr. 57.60). As Zonaras shows, Dio will have gone on to report the inactivity in this year of both Hannibal and the consuls. Here, as often in the extant books, Dio contrived a thematic link across transitional material, in this case by one of his characteristic references to a lack of memorable events48.

44The corresponding passage in Zonaras (9.11.3-5) summarizes the material on Spain and Greece, but does not name the new consuls or clearly indicate the year transition, and omits the prodigies. Zonaras includes only a few prodigy notices, and in general retains little routine domestic material49.

45Livy gives the same sequence of material, passing from Spain to Greece and then to the domestic section for the transition to 206, followed by that year’s inactivity in Italy (28.5.1-12.9). As usual, Livy’s domestic section includes a range of routine detail, but Dio retains only the prodigies. Since Dio includes two prodigies omitted by Livy (a swarm and a horned woman), he must have been following a common source, probably Antias50.

46The second Paris fragment comes from Dio’s account of the year 204. It opens with the concluding part of his account of the war in Africa in that year (fr. 57.63-69, epitomized by Zonaras 9.12.3-5). Then follow two topics omitted by Zonaras (fr. 57.70-71): “In Italy in the war against Hannibal nothing important was achieved” (ἐν δὲ τῇ Ἰταλίᾳ κατὰ µὲν τὸν πρὸς Ἀννίβαν πόλεµον οὐδὲν µέγα ἐπράχθη), with the consul P. Sempronius losing a minor engagement and winning another, but the censors Livius and Nero demanded registers from recalcitrant Latin colonies, imposed a new salt tax, and degraded each other. The fragment breaks off here, but Dio no doubt ended his account of the year at this point. Livy has the same order, following the war in Africa with the year’s remaining events (29.36.4-38.8), but includes further items – other military activity in Italy, elections, changes in priesthoods, activity of aediles – as well as a much fuller account of the censors’ activity. Dio once again contrives thematic continuity around his judgement of historical significance (Sempronius and Hannibal achieved nothing important, but the censors did), and he justifies mentioning the salt tax from its political motivation.

47The third Paris fragment, from the account of the year 201, deals with the peace settlement with Carthage (fr. 57.83-86), followed by strained relations with Philip (fr. 57.76), but is unfortunately poorly preserved.

48Thus these Paris fragments enable us to sample Dio’s handling of his year-narratives in a period of Republican history for which he provided a relatively full annalistic account. They show him passing fluently between topics within a year and across the year transition, and (unlike Zonaras) regularly including domestic information on matters like prodigies and censorships. However, there is also a marked difference between these narratives and Livy’s much fuller treatment, with its wealth of routine detail and formalized structuring around the consuls’ movements.

Books 18-20: 200-150 BC

49As Zonaras makes clear, Dio’s account of the first half of the second century was dominated by the three eastern wars, against Philip (200-196), Antiochus and his allies (191-188), and Perseus (171-168). Dio, as we have seen, devoted Books 18-20 to this period, and, despite the unreliability of his division, Boissevain was probably right to suppose that each of these wars formed the main focus of one of the books51.

50Livy, extant to 167, used Polybius as his main source for the eastern events of the period and Roman writers, especially Antias, for events elsewhere. On non-eastern events Dio again sometimes drew on accounts differing from Livy’s. On the east he generally followed the Polybian account as mediated by Livy rather than using Polybius directly, although he also gave some eastern information not found in Livy52.

51Zonaras (9.15.1-18.4) shows that down to 195 Dio continued to give a year-by-year account in which each year-narrative reported events in diverse sectors. For each year down to 196 Zonaras reports warfare both against Philip and in northern Italy. The account of the year 195 opens with Cato’s opposition to the repeal of the Oppian law and his campaign in Spain before returning to events in Greece. Portions of Dio’s account of the year 200 survive in the mutilated fourth Paris fragment. This includes some of the narrative of the war against Philip and in northern Italy (fr. 58.1-6), followed by a domestic section (once again passed over by Zonaras) reporting the disputed triumph granted to the praetor L. Furius and an embassy from the Numidian prince Vermina (fr. 57.81)53.

52Zonaras does not mention the consuls for 194-192. After sketching developments in Greece in 194-193 (9.18.5-6), he doubles back to outline Antiochus’ deteriorating relations with Rome from c. 197 (9.18.7-14), before proceeding with the events of 192 up to the outbreak of war (9.19.1-5). Dio’s narration of these years doubtless continued to take annalistic form, but with an interruption at the start of the year 192 to expound the origins of the war. Dio had done the same for the wars with Carthage and with Philip, but the diplomatic antecedents of the Antiochus war necessitated more extended chronological backtracking. This may have involved substantial remoulding of the material in Dio’s sources, since both Polybius and Livy covered these events in annual sections. An alternative possibility is that Dio here turned to Appian, whose non-annalistic account provided a connected narrative of Antiochus’ pre-war dealings with the Romans (Syr., 1,1-14,59).

53For the years 191-187 Zonaras once again shows that Dio gave a year-by-year narrative (9.19.6-21.5). Dio’s account, like Zonaras’, must have been largely taken up by the warfare in the East, but also included information on other sectors: Zonaras reports Scipio Nasica’s war against the Boii in 191 (9.19.6), and fragments show that Dio mentioned the insulting of a Carthaginian embassy in 188 (fr. 63), and the consuls’ roadbuilding in northern Italy in 187 (fr. 65.2). However, the trials of the Scipios seem to have been mentioned only in digressions54.

54Zonaras devotes only a few lines to the period between the wars with Antiochus and Perseus (9.21.5-22.3). For the years 187-179 he briefly notices developments in the Seleucid and Macedonian royal houses, and various events of 183 (the incursion of Transalpine Gauls into Italy and the deaths of Hannibal and Scipio). The inclusion of these events suggests that Dio kept up (probably brief) year-by-year coverage through the 180s. However, after the accession of Perseus (in 179), Zonaras merely states that “in the period following this some events took place, but were not so indispensable as to be thought worthy of record” (9.22.3: ἐν δὲ τοῖς µετὰ ταῦτα χρόνοις συνηνέχθησαν µέν τινα, οὐ µέντοι καὶ ἀναγκαῖα πάνυ ὥστε καὶ συγγραφῆς νοµίζεσθαι ἄξια). He then passes directly to the Roman rejection of Perseus’ final embassy and the resulting outbreak of war55. Zonaras must again be reproducing an observation by Dio himself, characteristically deeming omitted matter unworthy of record56. Thus, although Dio may not have omitted the events of 178-172 as completely as Zonaras, it does appear that he abandoned year-by-year narration for that period.

55Dio reverted to the annalistic mode for the years 171-167 to recount the war with Perseus and Paullus’ triumph, as Zonaras shows (9.22.3-24.5). However, for the period between the defeat of Perseus and the outbreak of the Third Punic War in 149, Dio appears to have abandoned not only annalistic structure, but even his Rome-centred focus. From Perseus’ own fate, Zonaras (9.24.6-7) passes to the Roman treatment of eastern states whose stance during the war had made them suspect and then to the complex affairs of the Cappadocian, Ptolemaic and Seleucid monarchies over the 160s and 150s (9.24.8-25.8). These three narratives each extend over a lengthy chronological period and focus on the kingdom itself, rather than the Romans’ interventions. Zonaras then reports just one other event, the Roman war against the Dalmatians of 156-155 (9.25.9), before passing to the Third Punic War with the comment that “other events also took place at that time, but not deserving of mention or record” (καὶ ἄλλα δὲ κατ’ ἐκείνους συνέβη τοὺς χρόνους, οὐ µνήµης µέντοι οὐδ’ ἱστορίας ἐπάξια). Zonaras must again have taken both the comment and the narrative structure from Dio. The exceptionally low Roman military activity in the period after Perseus’ defeat appears to have led Dio to give it distinctive treatment. If his source for this section was Polybius, direct or through Livy, he will have had to reshape the annually organized material for non-annalistic narration.

56Thus, while Dio continued with annual narratives comparable to those for the Second Punic War down to the war with Antiochus, he achieved a more compressed account thereafter by dropping annual narration from c. 179, and by privileging eastern affairs in his selection of material. Like Zonaras, he may have omitted significant developments elsewhere, for example in Spain (not mentioned by Zonaras after 195).

Books 21-35: 149-70 BC

57For the major wars of the years 149-146 Dio provided extended treatment which probably occupied all of Book 21. As Zonaras shows, Dio narrated the Third Punic War by consular years, but treated the Macedonian and Achaean Wars in single blocs under respectively 149 and 14657. He also maintained his interest in the eastern monarchies, reporting under 149 the replacement of Prusias, king of Dio’s native Bithynia, by his son Nicomedes58.

58Dio’s account of the years following 146 was not available to Zonaras, who accordingly omitted the period altogether. However, the fragments make it clear that Dio must have continued year-by-year narration for the remainder of the lost books. For the years after 146 he gave detailed treatment to the wars in Spain against Viriathus and the Numantines, and also reported warfare elsewhere and domestic developments such as censorships59. From Tiberius Gracchus’ tribunate in 133 on, Dio’s account was full of political upheavals as well as ongoing warfare.

Conclusion

59Study of what survives of Books 1-35 in the fragments and Zonaras has enabled us to establish a good deal about how Dio allocated his space and structured his narrative by books and years. He accorded relatively detailed treatment and annalistic narration by the consular year to the early years of the Republic down to c. 443, to the period from the Third Samnite War to the war against Antiochus (298-187) and to the period from the outbreak of the Third Punic War in 149 on. The Paris fragments show some of Dio’s late third century year-narratives moving between sections on warfare in various sectors and on domestic events, linked by fluent transitions. Dio was, however, also ready to vary his annalistic techniques: thus he sometimes narrated events relating to individual regions over several years (as on Scipio’s Spanish command or the origins of the war with Antiochus); on the early Republic, he inserted institutional excursuses; and for a few of these years, for example 246-243, he deemed no events worthy of record.

60For other periods, however, Dio adopted a more selective mode of treatment, abandoning year-by-year treatment and narrating only such events as seemed to him to meet his threshold for inclusion. He appears to have handled the later fifth and the fourth century in this way (in Books 6-8) and to have reverted to this mode (in Book 20) for the years 179-150, during which he accorded detailed treatment only to the Third Macedonian War and then allowed his focus to switch from Rome to the eastern kingdoms. Adopting this more compressed, non-annalistic mode for these periods enabled Dio to complete the coverage of Roman history down to the mid-second century BC in the first twenty of his eighty books.

61Dio opted to write in the traditional format of annalistic Roman history. However, in his extant books he deployed that structure with notable flexibility, and, as we can now see, the same was also true of his lost early books60.