The first problem appears when one collects the PC eponyms. Obviously, since all Neo-Assyrian eponyms down to 649 are known from the canon, any eponym from a Neo-Assyrian text that is not in the canon is post-canonical.[[3]] However, 648 to 612 requires 37 eponyms while the number of attested PC eponyms is ≈50.[[4]] A number of these attestations are single occurrences, however, which might be discarded as mistakes, either ancient or modern; but many of them are quite clearly written or cannot be confused with anything else or occur in a strong archival context and so cannot simply be discarded.
Falkner attempted to solve this problem by positing a number of years with more than one eponym diachronically. That is, the year began normally but for some reason a new eponym was appointed at some point in the year. Falkner proposed this solution for no fewer than four years in her list, based on complementary distribution of occurrences through the year so that an eponym who was attested only at the beginning of the year was paired with one who was attested only in the latter part of the year. Unfortunately, with the increased documentation that is now available, most of these pairings can no longer be maintained.[[5]]
There are a number of other solutions that might be employed to pare down the number of attested PC eponyms to fit the 37 year span from 648 to 612. One is to simply throw away some of them, assuming them to be hopelessly garbled versions of canonical or already known post-canonical eponyms.[[6]] Another is to assume that there may be a few eponyms that are both post-canonical and post-empire.[[7]] These solutions may eliminate a few of the excess eponyms, but the most likely solution to account for the main body of them is to assume that there were a number of eponyms in use at different cities simultaneously. Unfortunately, direct evidence to support this assumption is not yet available,[[8]] nor is there an obvious spatial distribution pattern among the attested PC eponyms that bears it out.
Of the 40 eponyms used by Falkner in her final list, three can be removed:
This is a small gain, because since Falkner's study another seven or eight post-canonical eponyms have appeared. Most of these are single attestations and might not, therefore, have to be given much weight in a reconstruction, but several are multiply-attested or have good archival context, particularly Kanūnāyu, governor of Dūr-Šarrukēn,[[12]] and Pašî[[13]]. Further, a strong case can be made for there being a post-canonical eponym Bēl-šadûa.[[14]]
In her list, Falkner combined Adad-nādin-aḫi with Nabû-nādin-aḫi as a single entry under the former.[[15]] However, Nabû-nādin-aḫi is now multiply-attested (including a date on an Aššurbanipal prism) while Adad-nādin-aḫi, although clearly written on the tablet,[[16]] remains a hapax legomenon. Therefore, if the two are to be combined, Adad- must be considered an error for Nabû- rather than the other way around.
3 An exception is the text ND679.r. 12-14 (BaM 24 7), dated 22:i: mpa-qa-ḫa amēlšá-kìn āllibbi-āli, which from archival context belongs to the eighth century BC (cf. K. Deller and A. Fadhil, BaM 24 [1993] 266).
4 The exact number of PC eponyms depends on who is counting them. Specifically, opinions vary on which eponyms should be accepted unconditionally, which variant spellings represent the same eponym and which a different one, which writings are mistakes for other eponyms, which eponyms with the same name but different titles are the same name but different titles are the same etc.
5 In fact, only one of the pairings suggested by Falkner, Mannu-kī-aḫḫē and Sîn-šarru-uṣur, arkû, for *627, could be maintained at the present time; this may actually be a valid pairing since otherwise the title arkû, 'second', is unexplained. Falkner was able to use 209 Neo-Assyrian texts of all types (royal inscriptions, letters, administrative and legal texts) for her study, whereas, through the SAA database, we have been able to collect over 400 published and unpublished eponym-dated legal texts alone.
6 This is an especially attractive option for names that are completely unfamiliar and do not otherwise occur in the Neo-Assyrian onomasticon such as Nūr-ṣalam-kaspi and Šanta-dameqi.
7 That is, eponyms belonging to the years after 612 and used by the small remnant of Assyrians who escaped to the west after the fall of Nineveh. These eponyms would be attested only at western sites; a prime candidate is Nabû-mār-šarri-uṣur, attested only at Guzana (Tell Ḥalaf). This solution was already suggested by Falkner (loc. cit. 106), but not utilized in her final scheme.
8 The most probable scenario to account for this would be a widespread civil war that lasted for several years, with two (or more) contenders to the throne holding different cities and each appointing his own eponym. The most likely time for this to have happened would have been after the death of Aššurbanipal, but this is a period that is presently shrouded in obscurity and, although there is general agreement that the end of Aššurbanipal's reign was accompanied by some sort of power struggle, there is no hard evidence or clear-cut indication of the extent and duration of this struggle. Indeed, even the date of the death of Aššurbanipal and the accession dates of his successors are currently matters of contention.
9 Now identified with Sîn-ālik-pāni.
10 Also read variously as Adad-bēlu-naʾid, Adad-milki-naʾid and Šarru-naʾid. The reading of the name as Daddî and the identification with the eponym for *620 was established by O. Pedersén in OrSu 33-35 (1984-86) 313-15.
11 See above, p. 14, n. 36. Falkner was skeptical about this eponym and included it only with reservations (loc. cit. 118 n. 56).
12 The existence of a PC eponym Kanūnāyu was conclusively shown by S. M. Dalley and J. N. Postgate, TFS [1984] 5, 55 ad no. 6.2, and 63 ad no. 11. Some examples had hidden behind the canonical eponyms for 671 and 666, but it is attested at Assur (cf. K. Deller, BaM 15 [1984] 232 n. 31), Kalaḫ and Nineveh. The text from Nineveh (K441 = ADD 400) was dated to 688 (Iddin-aḫḫē) by G. Smith, Canon 90, but Johns when he copied the text could not justify the reading (cf. C. H. W. Johns, ADD IV [1923] 53 § 825). A fine convergence of evidence shows that this text must be dated to the eponymate of Kanūnāyuu (a) since the text belongs to the archive of Kakkullānu (see below, n. 19), it must be post-canonical and the original dating of the tablet by Smith can be ruled out; (b) the only likely reading for the first sign of the eponym's name is 'ba/ÌRI/ITI, or SUM, I suppose' (collation, J. N. Postgate); (c) the title, governor of Dūr-Šarrukēn, is attested for Kanūnāyu in TFS 11; (d) the eponym Kanūnāyu is associated with the eponym Aššur-mātu-taqqin, possibly being the year immediately preceding it, in TFS 6; (e) the archive of Kakkullānu contains three texts dated to the eponymate of Aššur-mātu-taqqin. Thus, lhe possible readings of the sign, the title, and the archival context all point to the PC eponym Kanūnāyu for this text.
13 First reported by Deller, BaM 15 (1984) 246; cf. O. Pedersén, ALA I 22 n. 9. It is thus far attested only at Assur, but a strong archival context from administrative texts (ALA N4:462-70) indicates that it should immediately follow the eponymat of Sîn-ālik-pāni.
14 This name is usually taken as an abbreviation for Bēl-Ḫarrān-šadûa (see RIA 2, 446) and in some cases it can be shown to be so, but there is growing evidence that Bēl-šadûa may have also been a PC eponym (cf. Dalley and Poslgate. TFS [1984] 5 and Deller, BaM 15 [1984] 232 n. 32). The difficulties lie in the fact that the eponymate of Bēl-Ḫarrān-šadûa is so close to the post-canonical period (650 BC), making it difficult to distinguish on prosopographical grounds, and that no title is ever associated either with Bēl-Ḫarrān-šadûa or Bēl-šadûa in the legal texts.
15 Loc. cit. 104 n. 17 and 118 (*634).
16 Collated by J. N. Postgate.
Robert Whiting
Robert Whiting, 'The Number of Post-canonical Eponyms', The Eponyms of the Assyrian Empire 910-612 BC, SAAS 2. Original publication: Helsinki, Helsinki University Press, 1994; online contents: SAAo/SAAS2 Project, a sub-project of MOCCI, 2023 [http://oracc.org/saao/saas2/postcanonicaleponyms/numberofpostcanonicaleponyms/]