coins carry no indication of date, and it is possible that the calendar was in fact invented on this occasion. In 96 BC or shortly before, Mithridates VI began to strike coins in precious metals, and with the exception of a very small number of undated coins they are from the beginning dated according to the Bithyno-Pontic era, which must have been adopted in Pontos in the early part of his reign.18 His forefathers, on the other hand, had used the Seleucid era. Mithridates’ motives for adopting another state’s era are a mystery, as this would normally indicate a subordinate position. Furthermore, relations with Bithynia were not very amicable at the time, at best rather competitive. Only in the joint invasion of Paphlagonia in 108 BC did the two kings cooperate, and this event marks the most likely time for the changeover.19
With the Roman conquest, the Bithyno-Pontic era ceased to be used in Asia Minor, but continued to be used in the Bosporan Kingdom at least until the end of the fifth century AD. The era was certainly employed in inscriptions in the Hellenistic period, as evidenced by inscriptions from the Northern Black Sea area,20 but to my knowledge no dated inscriptions have turned up in either Bithynia or Pontos, where so far it is only known from coins. It has generally been assumed that the Bithyno-Pontic era was abandoned because it was associated with kingship, but against this speaks the fact that the Seleucid era remained the predominant calendar system in the East until the Arab invasion, and continued to be used sporadically in Asia Minor well into the imperial period.21 Furthermore, the Bithyno-Pontic era was not particularly associated with Pontos and Mithridates VI, the enemy of Rome, since it had only been introduced in Pontos a generation prior to his succession; and the Romans in general seemed disinterested in imposing new calendars in the conquered territories. Changing the calendar system was therefore a very conscious choice of connecting oneself to Rome, and the initiative undoubtedly came from local authorities. Had it been in the interest of the Roman administration to regulate the reckoning of time this could easily have been brought about at any time by implementing a common provincial era, a familiar phenomenon in other provinces. For the administration of the province, it could not have been practical to encounter at least six different calendar systems when travelling the relatively short distance from Herakleia to Polemonion. Many cities chose the year of the city’s inclusion in the Roman province of Bithynia and Pontus as the starting point for their new era (or in Galatia in the case of the inner parts of Paphlagonia and Pontos).
Fig. 3. Grave stele for Iulia Galatia erected by Antiochos in the year 174 of the local era (AD 171/72), now in Amasya Museum (author’s photo).
For some reason, however, no one seems to have used the initial creation of the province of Bithynia and Pontus in 63 BC, nor is there any solid evidence for the often proposed Pompeian era. Instead, the Lucullan era starting in 70/69 BC was chosen in Amastris and Abonouteichos. Amisos deviates from the rest in that this city seems to have used its grant of freedom in the year 32/31 BC as the starting point.22 The long use of the Seleucid and the Bithyno-Pontic calendars in Asia Minor may explain the unusual popularity of calendars with years numbered in succession as opposed to eponymous magistrates or the year of reign of the emperor.
It is one thing to calculate out the year from which a particular era was reckoned, quite another to figure out when the era was actually introduced. In a few instances the two events are definitely contemporary. In Amastris, for example, coins were struck in year one of the Lucullan era,23 but often we find a considerable gap between the starting point of the era and our earliest evidence for its use. Along the coast, the gaps generally tend to be short; inland, on the other hand, it is a question of centuries rather than decades: Neapolis/Neoklaudiopolis, 115 years; Pompeiopolis, 174 years; Kaisareia/ Hadrianopolis, 170 years; Gangra/Germanikopolis, 198 years – all according to the Paphlagonian era starting 6/5 BC.24 The proposition that this era was already used in the famous oath of the Paphlagonians to Augustus is false.25 The number three in the text does not refer to the local era but rather to the third year of the 12th consulship of Augustus. That the two different readings in fact yield the same date, 3/2 BC, is a mere coincidence. The oath’s close connection with the emperor can further be seen in the date chosen, the 6th of March, the anniversary of Augustus’ elevation as pontifex maximus, and the use of νωνών Μαρτὶων transliterated from Latin further suggests a nonnative dating system.
The question is whether the era was actually introduced at a later date or whether it appears so due to the insufficiency of our sources. It is quite possible that a city could maintain and employ a calendar system that was never revealed in any of the sources available to us, as these comprise only coins and monumental inscriptions on stone. I would suggest, however, that the introduction of eras related to the city’s incorporation into the Roman empire was part of a larger package that included new settlement patterns, the introduction of local coinage, new social structures, and new means of self-expression, both individually and for communities as a whole, the latter primarily visible through what has become known as “the epigraphic habit”. All these markers seem to coincide more or less chronologically – at a much later date than the creation of the province. The correlation between coinage and inscriptions is of particular interest, as these contain our most precisely datable evidence and can therefore provide a clue to the date of this transformation.
The epigraphic habit
The epigraphic habit, or the use of inscriptions in public and private contexts, was a fundamental feature of participation in the Graeco-Roman cultural sphere. Judged by this parameter, northern Asia Minor, apart from the coastal cities, was by no means Hellenized under the Pontic kings, as hardly any inscriptions exist from the Hellenistic period. The epigraphic habit was closely associated with the Greek language, and the use of Greek seems very restricted and a rather late phenomenon outside the old Greek colonies – with the exception of coin legends. Several literary sources remark on the linguistic talents of Mithridates VI and relate that the king spoke all the tongues and dialects of his domain: twenty languages or more.26
Since a surprisingly large proportion of inscriptions in northern Asia Minor can be dated accurately, we can determine with some degree of certainty when the epigraphic habit was introduced. Naturally, caution should be taken when drawing conclusions from epigraphic sources. The preserved epigraphic monuments are by no means an unbiased selection of what once existed. Most importantly, we essentially only possess inscriptions written on stone. In northern Asia Minor, hard limestone was in scarce supply; on the other hand, metal was abundant, and inscriptions on bronze may have been more common than we can perceive today. Painted inscriptions on wooden panels may also have existed in an area rich in wood and Sinopean red dye.27 This leads to the question of survival rates. In Herakleia Pontike, for example, only about seventy or eighty inscriptions have been preserved, and the earliest may well be a base for a statue of Claudius. By this time the city had been among the major cities in the Black Sea for nearly 600 years, and no one would hesitate to place it within the Greek cultural sphere. The destruction of the city by Cotta in 70 BC, the extensive reuse of inscribed stones as building material, and the destructive forces of modern town planning are the standard explanations given for the small number of preserved inscriptions.28 At the sites chosen for investigation here, however, the inscriptions do not seem to have been subject to such radical selection during the Roman period and probably represent a fairly random sample.
Table 1. Chronological distribution of the dated inscriptions from Amaseia (based on French 1996).
| Decade | Number of inscriptions |
| AD 50-59 | 0 |
| 60-69 | 0 |
| 70-79 | 0 |
| 80-89 | 0 |
| 90-99 | 3 |
| 100-109 | 1 |
| 110-119 | 0 |
| 120-129 | 3 |
| 130-139 |
|
