Livy 30.11-30.12 describes the defeat of Syphax by Scipio and Masinissa at the battle of Cirta under the events of A.U.C 551 = 203, Polybius 14.2.1 says that the campaign began at the beginning of spring, while Ovid, Fasti 6.769 gives a precise date for this event.
Ovid gives the date of this battle as the day after the date of the battle of Lake Trasimene, hence it was either a.d. VII Kal. Quint. or a.d. IX Kal. Quint. However this uncertainty is not sufficient to affect the analysis of intercalation. For the sake of discussion, we will assume a.d. IX Kal. Quint.
The sequence of events given by Polybius and Livy is as follows:
Operations began at the very beginning of spring (Polybius 14.2.1)
- Scipio sets in motion the siege of Utica, breaks off negotiations with Syphax and Hasdrubal, and immediately sets fire to the Carthaginian and Numidian camps
- The Carthaginians rally their forces on the Great plain in 30 days (Polybius 14.7.9)
- Scipio sets out for the rallying point as soon as he hears of it (Polybius 14.8.1)
- On the fifth day thereafter he reached the rallying point (Polybius 14.8.2)
- On the fourth day thereafter Scipio defeated the enemy and Syphax fled to Numidia (Polybius 14.8.4)
- On the fifteenth day after that Roman forces arrive in Numidia (Livy 30.11) with orders not to give Syphax time to prepare to resist (Polybius 14.9.2)
- Syphax is defeated at Cirta on a.d. X or a.d. VIII Kal. Quint. (Ovid, Fasti 6.769)
Thus, the distance from the beginning of spring (c. March 1) to a.d. IX Kal. Quint. is 55 days plus the time covered by steps (2) and (8). The time from (7)-(8) was certainly short, c. 15 days seems sufficient. If the beginning of spring is reckoned c. March 1, the earliest possible date for a.d. IX Kal. Quint. A.U.C. 551 is c. 70 days later = c 8 May.
We can estimate the bounds on the Julian equivalent of a.d. IX Kal. Quint. A.U.C 551 assuming 0 to 3 intercalations between this year and A.U.C. 564 = 190, as follows:
Number of Intercalations Number of intercalated days Julian date of a.d. IX Kal. Quint. A.U.C. 551
A.U.C. 551-5640 0 7 July 203
1 22-23 14-15 June 203
2 44-46 22-24 May 203
3 66-69 29 April - 2 May 203Clearly there was at most 2 intercalations in the period from A.U.C. 551 = 203 to A.U.C. 564 = 190. Zero intercalations can be eliminated because of the analysis of the events of A.U.C. 557 = 197.
So much is universally agreed. The key issue for discriminating between one and two intercalations is the amount of time covered by step (2). If there was one intercalation, it covers a period of about 15 days. If there were two it covers nearly 40 days.
P. Marchetti, AC 43 (1973) 473, and J. Briscoe, A Commentary on Livy Books XXXIV-XXXVII 20 argued that Polybius' circumstantial account does not allow us to decide. However, P. S. Derow, Phoenix 30 (1976) 265 argues that a period of 40 days is simply too long, on the grounds that Polybius explicitly states that Scipio's preparations were already complete at the beginning of spring, so that it only remained to execute the plan; since all that was necessary was the charade of negotiation, 15 days is sufficient. Briscoe's comment that another 23 days "cannot be totally excluded" is literally true, but does not seem likely to me, and Derow's argument for a shorter period seems extremely reasonable. Indeed, both Marchetti and Briscoe do favour two intercalations based on their analyses of the intervening years, even though they select different solutions, so they implicitly accept that the early spring manoeuvering was in fact short.
P. S. Derow, Phoenix 30 (1976) 265 at 271ff, argues that this consular year was intercalary, on the grounds that there was only one intercalation between this year and A.U.C. 557 = 197, and that this year comes at the end of a period where intercalation seems to have been regular. Although intercalation during the Second Punic War was not as regular as Derow assumes, the basic argument is sound in that intercalation was more or less balanced. But the same reasoning applies equally well to the next year. Derow's argument for this year follows from a schematic assumption that intercalations occurred every alternate odd year in the period immediately before the gap in intercalations. However, the evidence for A.U.C. 552 = 202, although not conclusive, tends to favour the view that A.U.C. 551 was a regular year followed by an intercalary year.
We can therefore estimate the Julian equivalent of Roman civil dates in this year to a precision of +1 day. But W. Soltau, Römische Chronologie 186ff, followed by P. S. Derow, Phoenix 30 (1976) 265 at 268, have argued that we can do better than this -- that we can arrive at an exact solution. This analysis implies that Non. Iun. A.U.C. 551 fell between 5 May and 7 May 203. A solar eclipse occurred on 6 May 203, which these authors propose to identify as the eclipse of Ennius; a conclusion not accepted here.
Website © Chris Bennett, 2001-2007 -- All rights reserved