According to a Babylonian chronicle, on 20 Ṭebētu (December-January) 681 Sennacherib, the king of Assyria, was murdered by one of his sons during a rebellion.Footnote 1 Accounts of the assassination and the accompanying rebellion were also recorded in the royal inscriptions of Sennacherib's son and successor to the throne, Esarhaddon, as well as Babylonian records, the Hebrew Bible, and aspects of the affair survived into later Near Eastern and Greek traditions.Footnote 2 Yet despite the range of sources there remain considerable gaps in our knowledge. For instance, all the surviving sources state that Sennacherib was assassinated in an act of patricide, but the number of sons involved varies across the accounts and we do not have all the details about the causes and nature of the rebellion. Amidst all this, a cuneiform document has been placed at the centre of historical reconstructions of the murder of Sennacherib. The text under consideration is best known as SAA 18 100 according to its most recent scholarly edition, and was known as ABL 1091 in older literature. The previous editions were based on a large fragment of the text in the British Museum (80-7-19, 028). The text has been interpreted to report on the court conspiracy that led to Sennacherib's death. Pleasingly, a new fragment of the tablet (K 21923) has come to light, which provides insights into some of the key details of the text. Such a discovery calls for a new examination of SAA 18 100. In the following pages we present a new edition of the text and offer a new evaluation of the historical information that may be gleaned from its contents.
The recent history of SAA 18 100
As stated, the surviving portion of the tablet comprises two clay fragments kept in the British Museum. The beginning and end of the tablet are completely broken off and as much as 50% of the original text might be missing since, according to its profile, the tablet is at its broadest at the point of the break and suggests only half of the original is present in the two fragments.Footnote 3 The right-hand edge is also broken away which affects the end of each line of text, possibly up to four signs may be missing in some lines. The surviving text of the tablet comprises reports to the Assyrian court, most likely heard in Babylonia. The provenance of the tablet is believed to be from Nineveh.Footnote 4 Until 2005 only the larger fragment (1880-7-19, 028) was known to the scholarly community. In December of that year, Jeanette C. Fincke identified a small fragment (K 21923) which joined the larger fragment perfectly.Footnote 5
The most recent scholarly edition of the larger fragment was published in 2003 by Frances Reynolds in the State Archives of Assyria series.Footnote 6 It is understood by the authors that Reynolds’ original edition of the larger fragment of the tablet omitted many of Parpola's reconstructions and offered some alternatives. However, in the final stages of production some of the old reconstructions were inserted into her edition after the proofs had been corrected and sent.Footnote 7 This later editorial phase was mentioned in the introduction to SAA 18, as noted by Stephanie Dalley in her review of the volume, published in ZA 96 (Reference Dalley2006: 142–43), and explained by Reynolds (pers. comm.) in more detail when the review had been published. The significance of this editorial decision is that the scholarly edition as published in SAA 18 buttresses an out-of-date understanding of the large fragment, although Reynolds’ alternative reconstructions, as listed on Oracc, have attempted to rectify the matter in this difficult text. In the final stages of preparation of this article Christopher Jones published a hand-copy and edition of the smaller fragment.Footnote 8 His edition is different to our own and Jones did not re-evaluate the entire text in light of the new fragment. Indeed, Jones’ edition of the smaller fragment bolsters the existing reading of the larger fragment. Our edition will correct this situation.
The tablet was brought into the scholarly spotlight by Simo Parpola in his influential paper delivered at the 26th Rencontre Assyriologique Internationale in 1979 and published the following year. Despite having access only to the larger fragment, Parpola's study significantly improved the understanding of the text and reading of the personal names.Footnote 9 Of particular interest was Parpola's astute recognition that the name written marad-dnin.líl is to be read Urdu-Mullissu whom he correctly identified with Adrammelech, the chief protagonist in the Biblical accounts of Sennacherib's assassination (II Kings 19:37 // Isaiah 37:38).Footnote 10 In addition to this identification, Parpola proposed a more speculative argument that the letter reported back after the assassination with the chief aim to identify two Assyrian officials, Nabû-šumu-iškun and Ṣillāia, as accomplices in a conspiracy to assassinate Sennacherib. According to Parpola, Urdu-Mullissu who had been displaced as crown prince, was already known as the culprit in Sennacherib's assassination when the letter was written. Indeed, he states that “it is clear that the letter itself cannot have been addressed to Sennacherib” (the addressee's name is entirely missing on the tablet). “Had the writer wanted to warn the king of a threatening assassination, he would have expressed himself differently.”Footnote 11 Parpola's thesis is hinged on two points. The first is an interpretation of the problematic l. 9ʹ as i-na pa-<an> né-er-ti (alternatively i-na pa-né-er-ti as a sandhi writing), “bef<ore> the murder”.Footnote 12 Such a reading requires the letter to have been written after the assassination. The other point is that this letter was supposedly supported by the Biblical accounts of the assassination. This connection guided Parpola's interpretation and his reconstructions of broken passages of the text. For four decades, many scholars have followed Parpola's interpretation of the tablet, including his restorations of damaged and missing sections as well as the identity of the addressee, and most have repeated his conclusions.Footnote 13 However, reading l. 9ʹ as it stands removes the chief textual basis for his interpretation. Therefore, we take up a different position to that of Parpola, which is outlined in the edition of the text and the historical evaluation below. It is also worth dealing with some of the issues raised here as a preliminary discussion.
There is no question that the attraction of Parpola's interpretation is its correlation with the Hebrew Bible's accounts of the assassination. A link between the historical narratives of any two cultures from the ancient world is a desideratum; and for many, this desire is greatly increased when an extra-Biblical source correlates with a biblical record. Parpola's identification of Urdu-Mullissu (marad-dnin.líl) as Adrammelech was superb. However, the Biblical accounts are most likely informed by the Assyrian court during the time of Esarhaddon's reign. Both II Chronicles 33: 10–13 and Esarhaddon's inscriptions state that Manasseh was a loyal, tribute paying vassal of Assyria who contributed to the Assyrian ruler's building programme.Footnote 14 The official and/or prophetic accounts of the assassination of Sennacherib from Esarhaddon's court most likely made their way into the Biblical narrative through the Chronicles of the Kings of Judah, the source to which the Biblical authors regularly directed readers following the account of each king's death.Footnote 15 Both imply that more than one son were involved in the assassination, and in the biblical accounts two sons, Adrammelech and Sharezer, are named as the culprits. In the case of the prophet Isaiah, it is just as likely that his account was also informed by the same Assyrian propaganda. It is well understood that the prophet's mimicking and mockery of Assyrian royal inscriptions is evidence that he was directly exposed to Assyrian propaganda.Footnote 16 It is also telling that the accounts in II Kings and Isaiah are identical. In this way, the Biblical accounts should not be viewed as an independent record of the events of 681; rather they are confirmation of Esarhaddon's version of the assassination.
When one distances the text of SAA 18 100 from the Biblical accounts, other difficulties with Parpola's interpretation emerge, particularly the date suggested for the composition of SAA 18 100 and to whom it was addressed. As noted above, the letter has ostensibly been dated to 681 following Parpola's thesis, and the point of the letter is to identify the co-conspirators who aided Urdu-Mullissu in his plot against his father. Urdu-Mullissu is certainly involved in the report, but the role attributed to him by Parpola can be re-assessed, for a straightforward reading of the text would see the warning that “Urdu-[Mullissu], your son, is going to murder you” (ll. r4–5) places the letter in the reign of Sennacherib rather than Esarhaddon. This matter is taken up in more detail below.
The uncertainty of the date of SAA 18 100 is emblematic of scholarly knowledge of the period of Esarhaddon's accession in general. Esarhaddon's inscriptions were not arranged chronologically and the current understanding of the order of events during his reign is dependent on Mesopotamian chronicles.Footnote 17 The chronological difficulties associated with Esarhaddon's Babylonian inscriptions have also been clarified recently by Jamie Novotny, who has demonstrated that the ambiguous dating formula, šanat rēš šarrūti “accession year,” is an ideological statement and the inscriptions from Babylon date much later in Esarhaddon's reign and can no longer be viewed as contemporary to the period under discussion.Footnote 18 Similarly, Esarhaddon's so-called “Apology” (Nineveh A), his only account of the period of the assassination, was also composed late in his reign, in 673.Footnote 19 Even the Zincirli Stele (Monument A / Esarhaddon 98) in which Esarhaddon called himself mutēr gimil abi ālidīšu, “the avenger of the father who begot him”, is dated to 671 at the earliest.Footnote 20 Thus, all textual sources from Esarhaddon's reign that provide an insight into assassination and the conditions at the time of his accession were composed much later in his reign.
In sum, while Parpola's interpretation of SAA 18 100 greatly improved our understanding of the text, one cannot uncritically accept his historical interpretation of the text, dating the text to Esarhaddon's reign, nor his reconstructions of the broken parts of the tablet in order to fit the Hebrew Bible's version of events. What follows is a new edition of the tablet and a fresh historical assessment of its contents that takes the new join into account.
Synopsis of the new edition of SAA 18 100
The actual tablet of SAA 18 100 appears to be an archival copy of a letter sent from Babylonia to Sennacherib at the Assyrian court reporting on information that had been gathered somewhere among the southern cities of the Assyrian empire. Reynolds points out that the two Winkelhaken in l. r12 are typically found in archival copies to indicate the beginning of a new line in the original document.Footnote 21 A number of questions about this tablet's context cannot be answered, not only because little is known about its findspot, but also because the vast majority of Sennacherib's archive is yet to be found. The tablet is just one of 681 so-far excavated Babylonian letters sent to the king or members of the royal family from the reigns of Sargon II to Ashurbanipal that were archived at Nineveh.Footnote 22 However, our reading of the text of the tablet places the original letter recorded in SAA 18 100 in the reign of Sennacherib, and not as it commonly accepted, the reign of Esarhaddon. While Sennacherib's reign is poorly attested in the corpus of archival documents, scholars now recognise that some letters traditionally dated to Sargon II and Esarhaddon might in fact be dated to Sennacherib.Footnote 23 We argue that SAA 18 100 would be another example of such a document.
As noted above, from what can be understood from the preserved parts of SAA 18 100, the tablet contains a report to the Assyrian royal court.Footnote 24 The report begins before the opening lines of the surviving text and intelligible text starts with the court officials, Ṣillāia and Nabû-šumu-iškun, receiving three brothers who are goldsmiths, perhaps at a temple in Babylonia. The goldsmiths claim to have overheard unnamed conspirators in Babylonia swearing loyalty oaths for treachery. One of those brothers, ‘Anon’ (perhaps Šarḫīa who is named later in the text), asks to address the king concerning the prince, Urdu-Mullissu. Ṣillāia and Nabû-šumu-iškun cover his face with a garment. They take ‘Anon’ before Urdu-Mullissu. ‘Anon’ thinks he is in the king's presence and claims that Urdu-Mullissu plans to kill the king, his father, Sennacherib. Urdu-Mullissu reveals himself and questions them further. The three brothers are named and the king is requested to check the gossip in Urdu-Mullissu's household. This may be understood as an attempt to incriminate Urdu-Mullissu falsely or to check the veracity of the report in order to exculpate him.
The final lines of the surviving text seem to consider this matter as a positive occurrence (da-ba-ba bab-ba-nu-ú). Unfortunately, little more can be ascertained as the text breaks off at this point.
New edition of SAA 18 100
80-7-19, 028 + 28-3-23 (K 21923): The present edition is based on a hand-copy of the tablet made by Stephanie Dalley on 10 March, 2016 (fig. 1), and high-resolution photographs supplied by Kim Wright in May 2015 (figs 2 and 3).
Fig. 1 Hand-copy of SAA 18 100 by Stephanie Dalley. Digitised by Luis Siddall and developed by Alyssa Coundouris.
Fig. 2 Photograph of the join (obverse), SAA 18 100. Original photograph by Dr Kim Wright, developed by Alyssa Coundouris.
Fig. 3 Photograph of the join (reverse), SAA 18 100. Original photograph by Dr Kim Wright, developed by Alyssa Coundouris.
Obv.
1′ ḪA[L? …]
2′ MU? ˹x˺[…]
3′ šá m˹d˺[…]
4′ šá da-˹x x x˺ […]
5′ 3 šešmeš-ni urutin.ti[rki …]
6′ lúkù.dim!meš i-na é […]
7′ a-de-e šá si-i-ḫi šá ˹x˺ […]
8′ ki-i iš-mu-ú 1+en i-n[a šà-bi-šú-nu]
9′ i-na pa-ni ir-ti a-mat lugal i[q ?-ta-bi]
10′ mdnà-mu-iš-kun ù mṣi[l-la ?-a ?]
11′ ki-i il-li-ku-nu i-šá-a-[lu-šú]
12′ um-ma a-mat lugal-ka a-na u[gu mi-i-ni]
13′ šu-ú ma-a ˹a˺-na ugu marad-d[nin.líl]
14′ i-na túg.kur.ra-šú pa-ni-šú ˹i˺-[ter ?-mu ?]
Rev.
1 i-na pa-ni marad-dnin.[líl-ma]
2 ul-te-zi-zi-šú um-ma a-mur ˹x˺[x x]
3 im-ma-gar ina pi-i-ka qi-[bi]
4 šu-ú iq-ta-bi um-ma marad-˹d˺[nin.líl]
5 dumu-ka i-dak-ka pa-ni-[šú]
6 ki-i ip-tu-ú marad-dni[n.líl]
7 ˹ki˺-i ú-sa-niq-šú a-na šá-˹a˺-[šú]
8 ˹ù˺ šešmeš-šú i-t[a-bak ?]
9 šu-˹mu ?˺ šá erimmeš mšar-ḫi-ia m˹d+˺[PN]
10 ù mdnà-šeš-apin-eš ki-˹i˺ [lugal]
11 la i-qi-pi erimmeš šá é mar[ad-dnin.líl]
12 lugal ˹li˺-šá-al-la : en-n[a a-du-ú ?]
13 da-ba-ba bab-ba-nu-ú šá [a-na lugal?]
14 ki-i [i]m ?-ḫur-ru a-na lu[gal en-i-ni ?]
15 ki-i iš-pu-ru lu[gal en-a-ni ?]
16 […] ˹x x x˺ […]
(Rest broken away)
Translation
(1´–3´) Too fragmentary for translation.
(4´–9´) […] who [heard ? a report ? …] three Babylonian brothers […] goldsmiths in the temple […] When they heard about the loyalty oaths of rebellion of […] one [among them said] at the first meeting: “It is a matter for the king.”
(10´–13´) When he came, Nabû-šumu-iškun and Ṣil[lāya] questioned [him], saying: “[What is] your appeal to the king about?” He said: “(It is) about Urdu-[Mullissu.”] (14´–r10b) [They covered] his face with his cloak and made him stand in front of Urdu-Mullissu saying: “Behold, [your appeal] is granted, speak with your own voice”. So he said: “Your son, Urdu-[Mullissu], is going to murder you”. When they uncovered his face (and) when Urdu-Mullissu had interrogated him, he (Urdu-Mullissu) l[ed] him and his brothers [away]. The names of the men are Šarḫīa, [PN], and Nabû-aḫu-ēreš. (r10b–r12a) If [the king] does not believe (the report), the king should question the men of Urdu-[Mullissu's] household.
(r12b–r15) Now [then], it is an excellent report that he appealed […] (and) wrote to the king, [our lord.] The king, [our lord, …]
The rest of the tablet is broken away.
Textual notes
4ʹ The new fragment shows ŠÁ- and DA-signs at this beginning of this line. Jones’ suggestion of “šá 4(bán)” does not fit the context of the letter. In our translation we have tentatively suggested “report” on the basis of the DA-sign. However, the traces of the two signs that follow do not allow for a more certain identification.
5ʹ Parpola's reconstruction of a ŠEŠ-sign at the beginning of this line was correct and the new join also shows that the text is concerned with three brothers located in Babylonia.
6ʹ lúkù.dim!meš, “goldsmiths,” replaces the earlier reconstruction of a personal name Šulmu-aḫḫē ([mdi-m]u- papmeš) and is clear on the newly joined fragment. The DIM-sign is written a little awkwardly, but the preserved wedges fit a Neo-Babylonian sign form reasonably well. Following Simo Parpola (Reference Parpola1988), the logogram should be read as kutīmu in Babylonian rather than the Assyrian ṣarrāpu as it does not use the Assyrian logogram lúsimug.kù.gi.
The É-sign most likely refers to a temple, rather than a house or estate, but there is no clue as to which it might be. One may tentatively suggest the E-sagil as goldsmiths held high-ranking positions from the Neo-Assyrian period down to at least Hellenistic times. On one occasion during the reign of Antiochus IV a šatammu was succeeded by his goldsmith brother who also became a zazakku (Boiy Reference Boiy2004: 161).
7ʹ The A-sign of adê has an extra vertical wedge which might be the result of a split reed stylus.
9ʹ This line is the crux of the text and close collation of the tablet in person by Dalley and from photographs by both authors has yielded a provisional interpretation. We do not follow Simo Parpola's reconstruction: i-na pa-<an> né-er-ti “bef<ore> the murder”. Rather, we have sought to understand the text as it stands: i-na pa-ni ir-ti. This expression is not elsewhere attested, but since in this period ina irti (CAD I/J: s.v. irtu, p. 187b) can mean ‘to meet’, as in SAA 10 269 r9 and SAA 13 128:19 and 158 r13ʹ, and taking pa-ni as the ordinal number, pānī’u ‘first’ (not panu ‘front’ or ‘before’, see Hämeen-Anttila Reference Hämeen-Anttila2000: 85–86), one might translate: ‘At the first meeting he said: “It is a matter for the king (and nobody else)”. Therefore, he supposed he met the king when he had his face covered. No other occurrences of the phrase as a whole have yet been found. The word order for pānī’u preceding the noun irti is comparable to the common usage of šānī’u ‘second’. Indeed Woodington (Reference Woodington1982: 200) has shown that the placement of a syllabically written ordinal number before the referent is typical in Neo-Babylonian letters. If the use of pānī’u is correct, irtu in metaphorical use is masculine; cf. lišānu f. as physical ‘tongue’, m. in metaphorical use as ‘language’. Alternatively pan(i) with an extended prepositional use, but the attested meanings ‘before the meeting’ (CAD P A 5a) or ‘at the time of the meeting’ (CAD P A 5a 2′) seem less appropriate if our understanding of the situation, which implies the need for a second meeting, is correct. We thank the editor for his help with this point.
The reconstruction of iqtabi at the end of the line is speculative. The traces of the broken sign on the edge of the tablet do not match the IQ-sign well, but qabû is the expected verb.
10ʹ It is an attractive idea to identify the Nabû-šumu-iškun of this line with the chariot-driver of Sennacherib, and most likely as a different individual from the contemporary šakin ṭēmi from Babylonia, or a son of Merodach-baladan II (Marduk-apla-iddina) (Baker Reference Baker and Baker2001). Remarkably, the same name is also found in the story of Ahiqar as the man sent to kill Ahiqar after he fled having been implicated in a plot – sent by Esarhaddon in some versions, by Sennacherib in others (Dalley Reference Dalley2007: 40).
12ʹ On the phrase amat šarri qabû “to supplicate the king,” see Postgate Reference Postgate and Garelli1974: 424–425.
14ʹ The reading of túg.kur.ra in Neo-Babylonian Akkadian remains unclear (CAD S: 20), but the term seems to have been used in texts from this period as a generic term for garment and can stand for túgḫi.a (Zawadzki Reference Zawadzki, Michel and Nosch2010: 412).
r2 ul-te-zi-zi-šú contains an overhanging vowel (ul-te-ZI-zi-šú for ul-te-EZ-zi-šú). For other examples of CV for VC orthographies in first millennium texts, see Deller Reference Deller1962: 190–193 and George Reference George2003: I, 350 and 438). See also comments to ll. r11, r12, and r 14.
r3 Here there is a change in orthography from i-na to ina.
r7 We interpret the ana here as a further example of the use of a nota accusativi (see Hämeen-Anttila Reference Hämeen-Anttila2000: 77).
r8 The new fragment adds support for Parpola's reconstruction of an Ù-sign at the beginning of the line. However, collation of the traces of broken signs on the right-hand side of the tablet do not fit a reading of i-d[u-ku] (so Parpola), but i-t[a-x-(x)]. Indeed, Frances Reynolds’ published collation (Reference Reynolds2003:257, and Reference Reynolds2017) also reads the surviving traces of the last sign as a TA. We tentatively reconstruct i-t[a-bak ?] based on a recommendation from one of the anonymous referees as abāku is reasonably common in Neo-Babylonian letters in the state archives of Assyria.
r9–10 The new fragment offers considerable improvement to the interpretation of these lines. It was previously thought that there were four men listed, but close collation of the join and subsequent signs in these lines show: a) there is no numeral, rather the ŠÁ-sign is the determinative pronoun, and b) that the first sign of l. r10 is an Ù-sign, which connects the three named persons rather than Nabû-aḫu-ēreš being the recipient of the report, and therefore not to be connected with the Eponym of the same name for 681. It now seems that these lines preserve the names of the three Babylonian goldsmiths whence the report originated. Close collation of the tablet shows that the older (tentative) reading which saw two names, mTUR? and mDÙG-ia, is in fact one name, mšar-ḫi-ia. The name is not attested in other Neo-Assyrian documents currently known (so PNA and Prosobab). The name is based on the adjective meaning to be ‘proud’, ‘noble’, ‘splendid’, and ‘magnificent,’ is said of both gods and humans, and does appear in personal names (CAD Š/II: 62). Importantly, šarḫu is attested as an element in a Neo-Babylonian name from the state archives: Nabû-šarḫu-ubâša (SAA 16 50). Unfortunately, the second name has not survived. If this reconstruction is correct, it stands to reason that the first named of the brothers, Šarḫīa, might have been the fellow reportedly tricked into revealing the plot to Urdu-Mullissu.
r11 Again the scribe has included an overhanging vowel: i-qi-PI for i-qi-IP. While we have followed the common meaning of qiāpu, “to believe” or “to trust”, we note that CDA: 228 and AHw (919a s.v. qiāpu (3) state an additional meaning, “to verify,” which would make good sense in this context.
r12 As noted above, this line has two Winkelhaken to indicate the end of one line and the beginning of another in the original document (see also n. 22 above). There is also a further defective spelling with an extra CV-sign: li-šá-al-la for lišāl.
r13 The join confirms Reynolds’ reconstruction of the beginning of this line. Jones’ interpretation of dababa babbanūtu as “excellent nonsense” is influenced by his attempt to make it fit previous editions of the larger fragment.
r14 The break on the tablet's surface where the fragments join has caused difficulties for reading the first sign of the verb. We have elected for a Neo-Babylonian IM-sign, but acknowledge Jones’ reading of the traces as [ni]m-ḫur-ru. This is another occurrence of the scribe's non-standard orthography, this time a CVC + CV for CVCV, that is im-ḫur-ru for imḫuru.
r14–15 The new fragment has clarified the beginnings of these lines, revealing the scribe's consistent style by structuring his report with a series of temporal clauses.
Historical evaluation of the new edition of SAA 18 100
The new edition of SAA 18 100 provides a considerably different interpretation of the text and requires a re-evaluation of what it contributes to the history of the period. While we would agree with the existing thesis that SAA 18 100 concerns a plot against Sennacherib, we would argue that the present edition of the text undermines Simo Parpola's central idea that the letter reports back on the assassination at the hands of Urdu-Mullissu. Rather, the letter seems to inform the king that Urdu-Mullissu's household is blameless and prompts the king to investigate the matter for himself. This minimalist reconstruction has significant consequences, which point towards an older theory that Esarhaddon should be considered the main suspect in the assassination of his father.Footnote 25
It may be seen that the greatest difference between the present edition of SAA 18 100 and that of Parpola is the fate of the brothers who came before Urdu-Mullissu. As noted in the comments to l. r8, the traces do not support Parpola's reading of i-d[u-ku], rather i-t[a-x-(x)], for which we suggest the reconstruction i-t[a-bak]. This new reading is not trivial, for it considerably alters the tone of the report. There is nothing in the surviving portions to demonstrate that the informant and his brothers came to a sinister end. If we combine this reading with our evaluation of the crux interpretum of l. 9ˊ that here was a supposed plot (and not after the murder) at the time the letter was written, then the common interpretation of Urdu-Mullissu as the wronged and vengeful Assyrian prince may not be so straight-forward.
If the actions of Urdu-Mullissu and his men were not sinister, then we need to also review another assumption about the account of face covering in ll. 14´–r6. This incident of face covering for a commoner in the presence of the king has been considered evidence that this letter was dated to Esarhaddon's reign. Karen Radner has considered this line to be confirmation that Esarhaddon's ill health and noticeable skin disease needed to be protected from public view, lest the king's subjects worry his afflictions were signs of divine abandonment.Footnote 26 However, if Parpola is correct that it was a rare privilege to see the face of the king, and l. 10ʹ then refers to a general practice of veiling before the king (Sennacherib on this occasion), then this text is not proof of the concealment of Esarhaddon on account of his illness.Footnote 27
To return to the immediate issue of the relationship of SAA 18 100 to the assassination, the careers of the so-called co-conspirators following the accession of Esarhaddon is also telling. If Nabû-šumu-iškun and Ṣillāia were acting against Esarhaddon, how did they survive the purge of the defeated faction that would have occurred in the months following Esarhaddon succession? Indeed, we know from the large number of denunciations during Esarhaddon's reign and the purge of the officialdom in 670, that the new Assyrian king had the stomach and temperament for decisive action. The careers of the co-conspirators fared far better than that of Urdu-Mullissu and his brother(s) who, according to Esarhaddon's Apology and the Hebrew Bible, fled north when the new king entered Nineveh. Nabû-šumu-iškun and Ṣillāia continued to enjoy prominent careers during the reign of Esarhaddon, despite the latter having a number of complaints made against his conduct throughout.Footnote 28 It beggars belief that two men who were revealed to be involved in the conspiracy not only survived retribution, but were rewarded with prominent positions in the Assyrian empire. Rather, this evidence would suggest that if Nabû-šumu-iškun and Ṣillāia were a part of the factionalism of the era, they were on Esarhaddon's side, not Urdu-Mullissu's. Indeed, the earliest version of the Tale of Ahiqar found at Elephantine includes an episode where Ahiqar, a great sage under Sennacherib, had fled to Egypt during Esarhaddon's reign. Esarhaddon sent Nabû-šumu-iškun to kill Ahiqar, but he was unsuccessful in his mission.Footnote 29 This apocryphal tale combined with the attestations of these officials’ careers during Esarhaddon's reign have serious implications for Esarhaddon's Apology and the Biblical accounts of the assassination as it isolates their shared account as the only source that places the blame on multiple assassins, namely Esarhaddon's brothers.
As previously noted, the near contemporary Babylonian sources consistently stated that a single son was the assailant. The Babylonian chronicle cited at the beginning of this article states that the assassination occurred during a rebellion. The more partisan account of Nabonidus who interpreting the patricide as the gods’ revenge for the destruction of Babylon also stated that Sennacherib was murdered by one of his sons.Footnote 30 As does the later account in Berossus’ Babyloniaca who most likely used the Babylonian chronicles among his sources.Footnote 31 Unfortunately the anonymity of the assassin is maintained in each case. The important point here is that the Babylonian chroniclers were far more neutral than those operating in the Assyrian royal court and certainly their distance from the power of the Assyrian king meant that their version was not subject to his approval. In this light, if we were to take the Babylonian accounts of a single son as the culprit, then our focus should return to the one son who benefitted from the assassination.
In 1991, A. K. Grayson stated that while there was much in favour of Parpola's theory, the broken state of the letter meant that scholars should be wary that it cannot be proven.Footnote 32 Much of what has been said here points the blame away from Urdu-Mullissu and towards Esarhaddon. Both the Babylonian chronicle and Esarhaddon's Apology state that there was a period of rebellion at the end of Sennacherib's reign, when the princes “butted heads for power”.Footnote 33 During this period Esarhaddon claims to have fled the capital for his own protection after a falling out with Sennacherib. Rather than after the assassination, it is worth considering if Esarhaddon's fallout with his father occurred as a result of an attempt to discredit his brother, Urdu-Mullissu. This is certainly in line with the most recent scholarly attempts to rethink Sennacherib's assassination. Stephanie Dalley, M. de Jong Ellis, and Andrew Knapp have advanced arguments against Esarhaddon after taking a more critical view of Simo Parpola's thesis and noting that at no point in the Apology did Esarhaddon state his brothers committed the murder.Footnote 34 Dalley has suggested that the origin of the rebellion might have stemmed from the ambition of Esarhaddon's mother, Naqi'a, who sought to frame the crown prince's half-brothers. Ellis and Knapp also drew on the shift in Esarhaddon's Babylonian policy from that of Sennacherib and suggested that this might have reflected an alliance with a power base in Babylon.Footnote 35 Certainly our edition of the tablet is better suited to these historical reconstructions.
The recent join has enabled a new reading and interpretation of SAA 18 100 that changes the understanding of the text and what it says about the lead up to Sennacherib's assassination. The idea that the original letter copied in SAA 18 100 reported back on the assassination can no longer be supported, nor can the certainty of Urdu-Mullissu's guilt. Instead, we have a text that might be a product of the pro-Esarhaddon machinations of the royal court and administration during the final years of Sennacherib's reign.
Acknowledgements
The authors would like to thank the two anonymous reviewers who made a number of important recommendations that improved this article. Any remaining errors are the responsibility of the authors.
