« A.U.C. 564 = 190 B.C. »

The solar eclipse of 14 March 190 and the siege of Samë

An exact astronomical synchronism is given by Livy 37.4.4 which describes a total solar eclipse at Rome during the games of Apollo, on a.d. V Id. Quint. AUC. 564 = 14 March 190. This synchronism is now universally accepted, but it was not always so.

K. J. Beloch, Klio 15 (1918) 382 at 409ff. argued against this synchronism based on two inscriptions from Delphi.

Beloch argued that the minimum possible delay between the letters of Postumius and Livius must be 1-2 months. Since Postumius apparently wrote to the Delphians shortly after a.d. IV Non. Mai., it follows that Samë cannot have fallen before late Iun. A.U.C. 566, around three months into Livius' consulate. Since Fulvius was consul at the start of the siege, Samë also can't have fallen much later than about Id. Quint. Thus, the siege must have started in Intercalaris or early Martius A.U.C. 565.

After setting the siege in place, Fulvius had gone to the Peloponessus to settle a dispute between the Achaians and the Spartans (Livy 38.30) and then returned to Rome to conduct the elections for A.U.C. 566 (Livy 38.35). After the elections he returned to take command of the army in Greece. Hence, Beloch argued, in view of these events, the siege must have started no later than mid summer, around July. Hence he derives the equation that Mart. A.U.C. 566 = c. August 189.

But, on any chronology derived from the eclipse equation a.d. V Id. Quint. AUC. 564 = 14 March 190, Mart. A.U.C. 566 = c. Nov/Dec. 189. Hence, Beloch argued, this contemporary data, when coupled with Livy, shows that Livy's eclipse synchronism is invalid.

This analysis was countered by M. Holleaux, BCH 54 (1930) 1. Holleaux noted that Sp. Postumius was elected praetor at the end of A.U.C. 564 (Livy 37.47). Hence, if the senatus consultum of SIG 612 was that mentioned by Postumius, it must, at the latest, be dated to a.d. IV Non. Mai. A.U.C. 565 and therefore cannot be used to estimate the dates of events in A.U.C. 566. Further, an additional fragment of the senatus consultum of SIG 612 published after Beloch's article showed that the presiding officer was a certain "Octavius". There is no consul or praetor recorded by Livy with this name, who gives complete records, in the entire period A.U.C. 565-587. Holleaux concludes that the decree must be at least 20 years later than A.U.C. 565, and that the inscription it comes from was simply recording a chronological series of Roman decrees related to Delphi. However, SIG 611 does prove that the siege of Samë was ongoing after peace had been concluded with the Aetolians.

Holleaux then proceeded to examine Livy's and Polybius' internal chronology. Polybius 21.26.4 shows that the departure of Fulvius from Apollonia was after the start of summer, i.e., in his view, no earlier than the end of April or early May 189. He argues that the march to Ambracia, the siege of Ambracia, the negotiation between the Ambracians and the Aetolians, must have taken at least two months, hence not before the end of June. A war against the Galatians began and ended while the Aetolian war continued (Polybius 21.33) and that war was concluded in the autumn (Livy 38.27). One of the Aetolian envoys for negotiating peace was Nikandros (Polybius 21.30.15), who had been strategos that year. The key point is that the term of the strategos ended with the autumn equinox. Hence the Aetolian peace was concluded some time after September. Holleaux supposed that the siege of Samë only began after peace was concluded, i.e. some time around the beginning of October 189, and concluded four months later, roughly the end of January 188. He argued that this is consistent with Livy's account of the resumption of Achaian actions against Sparta "at the beginning of spring" (Livy 38.33). Hence Fulvius must have left for Rome to conduct the elections during the siege, some time in October or November 189, which is exactly what we would expect for Intercalaris A.U.C. 565 in a Roman civil year based on the eclipse equation a.d. V Id. Quint. AUC. 564 = 14 March 190.

V. M. Warrior, Chiron 18 (1988) 330, has reexamined this analysis and proposed a refined chronology that would have the siege of Samë starting in mid-late September 189. Similarly, Livy's account of Aetolian events in the preceding months (Livy 38.3) allows us to move Fulvius' departure from Apollonia back to March. However, none of these arguments affect the basic consistency that Holleaux established between the seasonal chronology of Polybius and Livy and the Roman civil chronology applied by the eclipse equation.

The Crossing of the Hellespont and the Salian Dances of P. Cornelius Scipio Africanus

Turning to the end of the year, Polybius 21.13.10 provides data which is used by several scholars, most recently V. M. Warrior, Chiron 18 (1988) 330, to determine its length. Polybius notes that the Roman army was stuck for 30 days en route to fighting Antiochus III after crossing the Hellespont, because P. Cornelius Scipio Africanus, advisor to the commander, his brother the consul L. Cornelius Scipio (Asiagenus), was forbidden to move for 30 days in order to conduct his duties as a Salian priest, and therefore could not cross the Hellespont with the army. (Livy 37.33 is vaguer, but does suggest that the time was more than a few days.) The battle of Magnesia followed a few weeks after the resumption of movement, and the march to Magnesia is described as occurring before the onset of winter.

Priestly activities of the Salii are known to have occurred from the Regifugium (a.d. VI Mart.) to the Tubilustrium (a.d. X Kal. Apr.), from a passage of Festus and the Fasti Praenestini, thus covering a period of 28 days. These dates are consistent with the Scipios' movements as given by Polybius and Livy, and so approximately limit the civil dates of the military immobility. Warrior argues that this data shows that the year was not intercalary. She points out that if A.U.C 564 = 190 was not intercalary, the Regifugium fell on 29 (recte 20) October 190, so the march would have resumed on 27 (recte 18) November 190, i.e. before the onset of winter, but if A.U.C 564 = 190 was intercalary, the Regifugium fell on 20 or 21 (recte 11) November 190, so the march would have resumed on 19 or 20 (recte 10) December 190, i.e. about the time of the onset of winter. Therefore the year was not intercalary.

Even though, as noted, Warrior's date calculations are in error by 9 days, this error is not substantive in the context. J. Briscoe, A Commentary on Livy Books XXXIV-XXXVII 338, argues that the delay was for the month of Martius, citing I.I. XIII.2.417 to show that the first of the sacred Salian dances was actually held on Kal. Mart. With no intercalation, this corresponds to 25 October-24 November 190. While I think Warrior is right on this point, because Briscoe adduces no evidence to show that the Salians were immobile after the Tubilustrium, again the difference does not substantively affect the chronological argument. In fact Briscoe also supposes that A.U.C 564 = 190 is regular.

However, M. Passehl, in two postings to the Yahoo Romanfederation group on 3 and 12 November 2007 and in pers. comm., points out a number of problems with this apparently convincing analysis, which he proposes to resolve by relocating the Salian dances to a later point in Livy's account: the period when Africanus was delayed in Elaea, supposedly due to illness (Livy 37.37).

In short, removing the synchronism of the Salian dances with the Hellespont crossing gives the following advantages:

Detaching the Salian dances from the Hellespont crossing also detaches the chronology of the Anatolian campaign from the Roman calendar. The reconciliation of Aetolian events with the Roman calendar given above nevertheless indicates that the year was regular.

It is more likely that the Salian story is misplaced than that it is false. There clearly was an extended delay at the Hellespont crossing, and Livy accepts Polybius' explanation of its cause. Moreover, it would have been well known that Scipio was a Salian priest, and Polybius was a close personal friend of his adoptive grandson, P. Cornelius Scipio Aemilianus (Polybius 31.23ff.). We also know that Scipio was immobilised for a second time on this campaign, from around the time the army reached the Caicus, when he is said to have fallen ill at Elaea (Livy 37.37), which is at the mouth of the river. He did not leave there until he rejoined his brother in Sardis after the battle (Livy 37.45 -- "Ephesus" is in error, cf. the Latin). On the Julian chronology Passehl proposes, this is c. 20 October - 20 November 190, which corresponds almost exactly to a.d. VI Kal. Mart. A.U.C. 564 - a.d. X Kal. Apr. A.U.C. 565 if it is assumed that A.U.C. 564 was regular. It appears that an illness at the Hellespont was exchanged with a religious duty at Elaea. How and why this happened is unknown. Possibly "Elaea" was confused with a similarly named town "Elaeus" on the Gallipoli peninsula.

For the purposes of this study, it does not matter whether Passehl or Warrior is right, so long as one of them is, since both conclude the year was regular. Having initially been quite sceptical, I have come around to Passehl's point of view, after working it through for myself. So long as the Salian delay happened at some point in the Anatolian campaign, the effect of making this year intercalary is to push the battle of Magnesia deep(er) into winter. In any case, the independent evidence of the Aetolian campaigns supports the view that the year was regular.

Website © Chris Bennett, 2001-2011 -- All rights reserved