The solar omen of Mursili II

by Peter J. Huber
The solar omen of Mursili II
Peter J. Huber
Journal of the American Oriental Society
Start Page: 
End Page: 
Select license: 
Select License

The fragmentary text KUB 14.4, a prayer of the Hittite king Mursili II, is reexamined with regard to chronological implications. The king recounts various misdeeds of the old queen, his stepmother. In particular, he tells how she tried to leverage a solar omen into an attempt to demote him. From the reconstructed sequence of events and actions of the queen and the king, I argue that this omen--presumably a noticeable, but not necessarily a total, solar eclipse--likely occurred shortly before, or in the early phase of, Mursili's campaign to Azzi in his tenth year. The most convincing candidates for such an eclipse are those of -1339 January 8 (probably total) or -1334 March 13 (possibly annular), while that of -1311 June 24 (possibly total) would seem to fall too late in the year.
In one of his prayers, the Hittite king Mursili II tells a story of intrigues by the queen (that is, by his step-mother, the Babylonian-born last wife and widow of Suppiluliuma, alternatively referred to as SAL.LUGAL or as "Tawannanna"). The story mentions the death of Mursili's wife, supposedly caused by the Tawannanna through voodoo, and an omen by the Sun-god. All this happened just before and during Mursili's second campaign to Azzi-Hayasa, that is, in his ninth and tenth year, and the omen clearly may be relevant for astronomical dating. Unfortunately, crucial passages of the tablet are badly damaged (see KUB 14.4 for the cuneiform text, and de Martino [1998] for a recent edition of entirety).
For someone concerned with astronomical dating, tantalizing questions are whether the omen refers to a solar eclipse, and if so, whether the eclipse was total. If it was an eclipse, we may ask where and when it was observed: prior to the campaign; during the campaign, in the capital Hattusa, or in Azzi? After perusing some of the earlier literature, in particular Emil Forrer's Forschungen (1926), and subsequent articles by Goetze and Forrer (1930), it seems to me that their interpretations were made cum ira et studio, and were influenced more by preconceptions and mutual animosities than by the actual textual evidence. Also the identifications of the omen by various authors with particular eclipses (-1339 Jan. 8, -1334 March 13, or -1311 June 24) seem rash and arbitrary.
First and foremost, I believe it is necessary to free oneself from preconditioning by previous analyses. I have therefore tried to ignore the existing literature and to check it only afterwards. What follows is my own analysis; it is presented with great hesitation, for I am not a Hittitologist.
The crucial passage occurs near the end of the text, in col. iv:
22. [x x x l]e-e e-ep-ti nu SAL.LUGAL am-mu-uk DAM-IA DUMU-IA A-NA [is.sup.d]-ha-ra
23. [pi-ra-an hur-za-a]k-ki-it (1) nu-un-na-as-kan an-da si-pa-an-za-ki-it nu-kan DAM-IA a-pe-el-la-az BA.US
24. [ma-a-an-ma (2) I-NA KUR [a.sup.URU]]z-zi ma i-ia-ah-ha-at nu [UTU.sup.d]-us sa-ki-ia-ah-ta SAL.LUGAL-ma
25. [x x x x x x x x] x me-mi-il-ki-it e-ni-wa ku-it [UTU.sup.d]-us sa-k[i-i]a-ah-ta
26. [x x x x x i-si-i]a-ah-ta U-UL-wa SA LUGAL-pat i-si-ia-ah-ta nu-wa ma-a-an
27. [x x x x x LU.MES GAL.]MES [ha.sup.URU]-at-ti-ma-wa-za AS-SUM BE-LU-UT-TI ta-ma-a-in
28. [x x x x x x x x]-wa-za [am.sup.f]-mi-in-na-ia [U.sup.l] SA [am.sup.f]-mi-in-na-ia-ia
29. [x x x x x x x x x] x-an-zi [UTU.sup.d]-SI-ma-za I-NA KUR [ha.sup.URU]-ia-sa
30. [x x x x x x x x x x] x tup-pf-az EGIR-an-da ha-at-ra-a-es
31. [x x x x x x x x x x] x wa-tar-na-ah-ta ma-a-an [ha.sup.URU]-ia-sa-az-ma
32. [x x x x x x x x x x x x x x] hu-u-da-a-ak U-UL me-mi-is-ta
33. [x x x x x x x x x x x x ma-ah]-ha-an ku-u-un me-mi-an is-ta-ma-as-sa-an-zi
34. [x x x x x x x x x x x x x x x p]u-nu-us-su-un ku-u-un-wa ku-in
35. [x x x x x x x x x x x x x x x] x-ia U-UL sa-an-ni-es-ta
36. [x x x x x x x x x x x x x x x x]-ma-ia [me]-mi-is-ta
37. [x x x x x x x x x x x x x x x x me-m]i-is-ta
In this transliteration I have refrained from making speculative restorations, not only because of my insufficient knowledge of idiomatic Hittite, but also to give an unbiased view of the textual evidence. In the following translation, I have, however, felt free to present speculative restorations:
(22.) Now the queen repeatedly [curs]ed me, my wife (and) my son [in front of] the goddess Ishara,
(23.) and sacrificed against(?) us. Then my wife died because of this.
(24.) [When] I was marching [toward the land A]zzi-now the Sun-god (had?) made an omen-but the queen
(25.) [acted with malice and] repeatedly said: "That omen which the Sun-god made-
(26.) [did it con]cern [the king's wife?] Did it not rather concern the king himself? Now, if
(27.) [it is so,... the great] ones of Hatti. with regard to the lordship, [shall select] another
(28.) [man as king and shall gi]ve [him] Amminnaya or(?) of Amminnaya
(29.) [her daughter as his wife]." But My Sun (i.e., the king) in Hayasa
(30.) [sent PN (to investigate)]. He wrote back on a tablet,
(31.) [and] he informed [me of the whole matter]. If from Hayasa
(32.) [...] he (or she?) did not speak at once.
(33.) [... wh]en they hear of this matter,
(34.) [...] I asked: "This (ace.) [...] which [..
(35.) [...] (s)he did not conceal.
(36.) [...] said.
(37.) [...] said.
The text and, in particular, its gaps raise a number of questions:
When and where did the omen occur?
Is this solar omen an eclipse or something else?
Lines 25-29 contain quoted direct speech. Who is the speaker?
Is this speech to be interpreted as assertive or as ironic?
In line 26, who is the alternative person potentially affected by the omen-the king's wife, the queen, or someone else?
What is the meaning of BE-LU-UT-TI in line 27?
What is the role of Amminnaya in line 28?
Before one even tries to identify the omen with a particular eclipse, one ought to find answers to the above questions and reconstruct from the fragmentary text a story that is not necessarily true, but at least reasoned and coherent.
My argumentation behind the above translation is as follows. The sentence about the omen of the Sun-god in iv 24 may very well be an explanatory note with reference to a past event, just as in iii 23-24, a year before, where the announcement of the march to Kummanni is followed by a parenthetical note on what Mursili's father had neglected to do there. I believe that the emphasis of the text is on the actions of the queen and on her interpretation of the omen, not on the omen itself, and that she may not have spoken up until Mursili was away campaigning. In other words, the actual sequence of the text offers little direct evidence with regard to the sequence of events, and the timing of the omen and the place of observation must be inferred from an interpretation of the text as a whole and from external evidence. In particular, the omen did not necessarily occur after Mursili's departure for Azzi-Hayasa. Since the emphasis of the text is on the actions of the queen, who lived in the capital, the eclipse was presumably observed there.
Interestingly, the omen is not recorded in Mursili's annals. This may mean that it was not observed during the campaign in Azzi-Hayasa itself. Clearly, this is risky arguing e silentio, and I hesitate to give it much weight. But it definitely shows that in Mursili's view the omen did not affect the campaign. This should be contrasted to the story of the meteorite related in year 3 of the annals, which had caused the king of Arzawa to fall seriously ill.
A solar omen is a conspicuous solar event, used to foretell consequences on the basis of similar events observed in the past that had been followed by an extraordinary piece of good or bad fortune for the king or for the land. Not every solar omen is an eclipse, but eclipses are certainly the most prominent of solar omens. Even if I cannot prove it, I believe that an eclipse was involved here, Of particular relevance is an omen mentioned in SAA 8, No. 104 (Hunger 1992), referring to the partial eclipse of -656 April 15 (magnitude 0.85): "If there is an eclipse in Nisan on the 28th day: the king of that land will fall ill but recover; in his stead, a daughter of the king, an entu-priestess, will die." (This omen may go back to an actual event that took place in the fiftieth year of Sargon of Akkad-cf. Huber (1999/2000)-but this is speculation.) We learn two things from this omen. First, while in principle the person to receive the evil of a solar eclipse is the king of the land where the eclipse is observed, someone in his entourage may receive it and die in his stead. Second, the eclipse need not be total to have evil consequences. However, the interpretation of solar eclipse omens is far from straightforward. For example, another report (SAA 8, No. 384) finds that a solar eclipse on day II 28 has favorable consequences, while on day II 29 it is catastrophic!
Quoted direct speech, indicated by the particle -Wa, begins in line 25 and extends to line 29. It seems to me that the speech is ironic and that in line 26 the speaker scathingly ridicules someone else's attempt to divert the evil impact of the omen away from the king. Incidentally, the non-recognition of possible irony by Forrer and Goetze is the main reason why I doubt the correctness of their interpretations. Note in particular the enclitic -pat focusing the reference on the king (somewhat inadequately rendered in English by words such as "precisely," "only," "rather," "himself," and the like), and the combination of the negation UL with -pat suggests an ironic question. Written Hittite does not mark questions as such, and an alternative interpretation of this line as an assertive statement ("It concerned X, it did not concern the king himself") is of course possible. I would then, however, have expected the opposite order and focus ("It did not concern the king; it concerned only X"). Who is the speaker h ere? The queen mentioned in line 24 could be either the subject or the object of a short sentence contained in the gap at the beginning of line 25. If she is the subject, then she would also have to be the subject of the direct speech, since there is hardly enough space in the gap for both a predicate and for introducing another subject. If she is the object, then there might be just enough space in the gap to finish the sentence and to introduce the subject of the subsequent direct speech (e.g., "The Sun-god made an omen and [it affected] the queen. [The oracle priest] repeatedly said..."). However, the use of the iterative/distributive -sk- seems better to fit a queen lobbying the Hittite barons than a priest reporting to his king. And who is X? There is hardly enough space at the beginning of line 26 to identify X by name, just enough for a reference of the type "the king's wife" or "a palace official."
In line 27, BE-LU-UT-TI (gen.) is not likely to mean "rulership (as queen)," as Forrer thought (" ... zur Herrschaft eine andere [Konigin ...]"). In Bogazkoy Akkadian, in a reference to queenship, one would expect the use of the word SAL.LUGAL-UT-TU = sarrattuttu, or possibly of *BE-EL-TU-UT-TU, even though the latter does not seem to be attested. Admittedly, the word occurs in quoted direct speech, and if the speaker is the queen (and only then), one could perhaps argue that a Babylonian-born woman would have used belute also for "queenship" (in the Babylonian dialect, the abstract noun is formed from the masculine, never from the feminine form). But if the word refers to lordship of Hatti rather than to queenship, then, rather startlingly, the queen's speech would have nothing to do with a relatively harmless attempt by her to foist an inconvenient new wife of her choice onto Mursili (as Forrer thought)--which wife, presumably, would later become her own successor as Tawannanna--but with an attempt to depos e him! If Hoffner's (1986) suggestion to read SA LUGAL US ("of the king's death") instead of -pat is correct, she might have suggested even more extreme means of disposal, but I think the emphasis of the text is on who is affected by the evil of the omen, rather than on how he is affected, so I prefer the reading -pat. Apart from that, Forrer's interpretation fails to establish a convincing connection between the omen and Amminnaya. Note that there is a subtle semantic distinction between sarrutu and belutu: The latter is very consistently used in connection with protecting the rule and succession of a king--see, for example, the letter KBo 1.10, obv. 14 and 31, and Friedrich, Staatsvertrage, in particular the Kupanta-[LAMMA.sup.d] treaty, passim.
Lines 27-28 may contain an implicit reference to the succession law described in the Telepinu text, If there is no male heir, the nobility should select a husband for the first-born daughter of the king and declare him king. For this explanation to hold, Amminnaya would have to be in the direct line of succession (i.e., to be a daughter of Suppiluliuma by one of his principal wives). The questionable, damaged sign in the middle of line 28 resembles a U-sign without the leading Winkelbaken. Perhaps it should be read NUMUN ("progeny"), or -pat over an erasure? All three readings would fit my interpretation, but a collation would be desirable. It is unfortunate that we have no other information about Amminnaya. She also appears in the fragmentary passages KBo 19.84:7 (probably belonging to the same text group as KUB 14.4) and KUB 18.42:6(?) (an oracle); see van den Hout (1998), 44. The name Amminnaya is also attested as a queen's name in Arrapha (ibid.).
It makes a great difference to the interpretation, and in particular to the timing of the eclipse, whether the speech is assertive or ironic. A detailed analysis seems necessary. If the speech is assertive, then the speaker would have to be an expert from the king's party, trying to deflect the evil of the omen away from the king, presumably onto the queen mentioned in line 24. The omen would have occurred after Mursili's departure for Azzi, because otherwise the questions would have been discussed and settled before he had left. But it could not have happened too late in the campaign season, because otherwise the apparent excuse in line 32 would not have been needed. This would point to an eclipse between, say, March and September. The problem with this reconstruction is that it is difficult to fit the passages about BE-LU-UT-TU and Amminnaya into it--as we have mentioned above, in Bogazkoy Akkadian BE-LU-UT-TU cannot mean "queenship," and hence the passage must refer to the lordship of Hatti. In particular , it cannot be interpreted as a proposal (by someone of the king's party) to depose the queen and to replace her by Amminnaya. But then, how does the latter enter the argument? Moreover, the queen was afterwards deposed and exiled not because of the omen, but ostensibly because she had murdered Mursili's wife. In other words, an assertive interpretation does not seem to fit the text. Quite apart from that, the current consensus among Hittitologists seems to argue on purely philological grounds in favor of interpreting the speech as containing a negative rhetorical question (cf. Hoffner [1986], 90).
On the other hand, if the speech is ironic, then the speaker ridicules someone else's attempt to divert the evil of the omen away from the king. Thus, the speaker must be someone from the queen's party, most likely the queen herself. At the same time, it follows that the "official" interpretation of the omen by the king's priests must have been shaky and open to challenge. Now, if the omen occurred before the death of Mursili's wife, then it presumably would have been adequately fulfilled by her subsequent death (in close analogy to SAA 8, No. 104, quoted above), and the irony would fall flat. I therefore conclude that the omen occurred after her death. If it occurred just shortly afterwards, the king's party might have been tempted to bend the rules and claim that the omen had been fulfilled retroactively. If it occurred much later, the king's priests might have tried to deflect its impact by claiming on the basis of oracles that it affected someone else in the king's entourage. Solar eclipse omens seem to p ertain first and foremost to the ruling king, and the substitute person would presumably have to die to make sure that the omen was fulfilled. However this may be, the queen may have had solid reasons to delay her reaction, since she would have a much better chance to dispute the Hittite priests' claims if she could base her counter-claims on the opinion of acknowledged experts. For that, she would presumably have had to write home to Babylon, more than 1200 km away. She would get an answer at best about three months later. If she spoke up while Mursili was on his way to Azzi-Hayasa, that is in April (or so), this would point to an eclipse in January, plus or minus a month. If she spoke up later, (3) the eclipse might have occurred somewhat later.
If we cast Mursili's wife in the role of the unknown X supposed to receive the evil of the eclipse, we note that her death is mentioned before the paragraph stroke between lines 23 and 24. Hence she would have died before the beginning of the campaign, that is, not later than February or early March. The solar omen must have taken place shortly afterwards, that is, at the latest in March or early April.
In order to accommodate a later date for the eclipse, we must assume that X was someone else, presumably a person in the king's entourage who had either died from natural causes--in view of the situation described in Mursili's plague prayers there should have been enough such deaths--or who had been quietly sacrificed to absorb the evil of the eclipse. But unless X was Mursili's wife, whose death is mentioned in line 23, the text, despite its
highly elliptical style, would seem to require a preparatory reference to X in the following line 24, such as, "the Sun-god made an omen and so-and-so died," or: "... and so-and-so was found to be affected." Since there is a reference to the queen right at the expected place in line 24, it might be tempting to assume that X is the queen herself. However, this hardly makes sense, since the queen is still alive. But then who is the mysterious X, and what is the basis for the queen's irony? Perhaps, if X were some distant relative of the king who had died after the eclipse, the queen could somewhat lamely claim that X was too distant a relation to qualify as an alternate recipient of the evil of the eclipse. This is possible, but does not seem convincing to me.
The final lines of the section (29-37), which are progressively more damaged, apparently deal with an interchange between the king in Hayasa and a person elsewhere, presumably in the capital. Curiously, line 30 emphasizes explicitly that reply by this person was by letter. Such emphasis may indicate that reply would not normally have been made in written form. Perhaps the king sent a personal envoy, who then wrote back instead of returning immediately. There is not enough space at the beginning of line 30 to spell out the contents of the king's inquiry, so implicitly it must refer to the actions of the queen described just before. Depending upon whether line 32 is interpreted as a negative rhetorical question, it would offer either an excuse or a complaint concerning delay. Line 34 may present a follow-up question posed by the king. The reconstructed sequence of events--eclipse, seditious conduct of the queen, receipt of news by the king in Hayasa dispatch of an envoy, report back, plus delays and follow-up q ueries--is so extensive that it is compatible only with an eclipse occurring during the early part of, or prior to, the Hayasa campaign.
If the repeated sayings of the queen (line 25) were indeed a well-founded attempt to persuade the Hittite nobility to fulfill the evil impact of the solar omen by deposing Mursili, then it would have been difficult for Mursili to defend himself directly against such an attack. Perhaps he decided to fight back indirectly, and to silence the Tawannanna by formally accusing her of the murder of his wife (which ultimately led to a conviction and her exile).
To summarize my admittedly speculative conclusions: I think that the omen was an impressively large, but not necessarily total solar eclipse, observed in Hattusa. If the quoted speech in lines 25-29 is ironic, and if the unknown person in line 26 is taken to be the king's wife, then all pieces of the puzzle fall into place. They are compatible with an eclipse in January through March, or perhaps April, at the end of Mursili's year 9 or at the beginning of year 10. I have not been able to find an equally convincing story-line fitting an eclipse occurring later in the year, although, a later occurrence cannot be absolutely excluded.
It remains to check this interpretation against calculation. During the second half of the fourteenth century the following solar eclipses were reasonably large in Bogazkoy: -1339 Jan. 8 (most probably total, shortly before noon); -1334 Mar. 13 (magn. at least 0.75, possibly annular with magn. 0.94, around noon); -1327 Oct. 17 (magn. 0.75, early afternoon); -1314 Aug. 26 (maga. 0.70, afternoon); -1311 Jun. 24 (magn. at least 0.99, perhaps total, in the early afternoon); -1307 April 13 (magn. 0.66, at sunrise). Note that solar eclipse magnitudes are relatively sensitive to the inaccurately known rotation of the earth, and that eclipses with a magnitude below 0.99 can easily pass unnoticed, unless they occur near sunset or sunrise. It is not the place here to discuss the thorny questions of Hittite chronology; but only the first and second possibilities, and perhaps the last (with a strong preference for the probably total eclipse of -1339), appear to fit all parts of the evidence. (Corresponding calculations for Trebizond, assuming that Azzi-Hayasa was situated somewhere between here and Hattusa, yielded comparable eclipse circumstances, apart from minor adjustments in the verbal qualifications of "possible" and "probable").
I thank Gernot Wilhelm, Calvert Watkins, and Gary Beckman for discussions and for help with the literature. However, the opinions expressed here are my own, and of course I absolve them from blame for any blunders I may have committed.
(1.) Restoration based on iii 19.
(2.) Restoration in analogy to iii 23.
(3.) G. Beckman has pointed out in a personal communication that the verb iya-in line 24 could be taken to mean while I was campaigning in A."
Forrer, E. 1926. Forschnungen. vol. 2/1. Berlin: Selbstverlag des Verfassers.
_____. 1930. Sakija(h) = "verfinstern"! Kleinasiatische Forschungen, 1: 273-85.
Goetze, A. 1930. Nochmals sakiiah(h)-. Kleinasiatische For-schungen, 1: 401-13.
Hoffner, H. A. 1983. A Prayer Qf Mursili II about his Stepmother. JAOS 103: 187-92.
_____. 1986. Studies in Hittite Grammar. In Kanissuwar: A Tribute to Hans G. Guterbock, ed. H. A. Hoffner and G. Beckman. Pp. 83-94. Chicago: The Oriental Institute.
van den Hout, T. 1998. The Purity of Kingship. Leiden: Brill.
Huber, P. J. 1999/2000. Astronomical Dating of Ur III and Akkad. Archiv far Orientforschung 46/47: 50-79.
Hunger, H. 1992. Astrological Reports to Assyrian Kings. State Archives of Assyria, vol. 8. Helsinki: Helsinki Univ. Press.
de Martino, S. 1998. Le accuse di Mursili II alla regina Tawananna secondo il testo KUB XIV 4. In Studi e Testi, vol. 1, ed. S. de Martino and F Imparati. Pp. 19-48. Florence: LoGisma.
COPYRIGHT 2001 American Oriental Society

  • Recommend Us