The Double Eclipse at the Downfall of Old Babylon
aa r X i v : . [ phy s i c s . h i s t - ph ] J u l I N ORIGINAL FORM PUBLISHED IN :Habilitation at the University of Heidelberg arXiv: 0000.00000 [physics.hist-ph]
Date: 14th July 2020
The Double Eclipse at the Downfall of Old Babylon
Emil Khalisi [email protected]
Abstract.
For many decades scholars converse how to correctly include the Old Babylonian Empire intothe absolute timeline of history. A cuneiform text from the series
Enuma Anu Enlil (EAE
Keywords:
Astronomical dating, Chronology, Solar eclipse, Earth’s rotation, Babylon, Akkad.
The Babylonian civilisation stands at the origin of schol-arship that has left traces right to the present. As thepeoples of Mesopotamia laid the foundation for our mod-ern science, its culture emanated to neighbours as well assuccessors. The history of Babylon can be divided into twomain phases: the Old Babylonian period at the beginning ofthe second millennium BCE and the Neo-Babylonian from626 to 539 BCE. The dating of the Old Babylonian Empirecauses much trouble. Depending on the field of study, itsabsolute time varies within a range of two centuries.In this paper, we briefly delineate the political mile-stones of Mesopotamian history and explore the timescalesin use that rest upon different methods of measurement.These timescales are the key element when taking a closerlook at the history of the ancient world. We will not provideanother new chronology, but try to fit an account on an ec-lipse pair into the existing ones. This account describes thedownfall of Babylon after a lunar and a solar eclipse occur-ring in the same month. Eclipses often recur pairwise, butfor the same location they are rare. As our results fall intovarious eras, we ponder over the most likely chronology.Finally, we will end up in Akkad as the probable solutionand make a suggestion for a slight restructure in the schemeof rulers there.Moreover, when analysing historical eclipses, we doconsider the irregular fluctuations in the earth’s rotation giv-ing rise to an error decisive for the eclipse to be visible ina certain region. Though a reliable answer is blurred by theerror, our findings will not alter the historical framework.
The first forms of a broader government along the tworivers Euphrates and Tigris go as far back as the thirdmillennium BCE when various city states rose with localdynasties. They were frequently at odds with each other,as inscriptions on steles and temple fragments prove. Aterritorial state probably formed in the second half of the3rd millennium. The history of Mesopotamia commenceswith the legendary king Sargon of Akkad who conquered aconsiderable realm in that region. Until now the geographiclocation of Akkad is not well known, probably it is to beplaced in the vicinity of today’s Baghdad. Other dynastiesexisted somewhat farther to the south in Uruk, Ur, Kish,and Lagash.Sargon is regarded as the first notable monarch in Meso-potamia. He paved the way for the later empires of Babylonand Assyria. Prior to him there were godlike kings reigningfor many thousands of years, some 30,000 or even 200,000years. Their lifetimes cannot be determined, in general.One of them is Gilgamesh, the king of Uruk, who is linkedto the legend of the Noachian Flood. That old time beforeBabylon is often bundled to the “Sumerian Era”.The Sumerians laid the scientific basis of the advancedcivilisation that propagated throughout the entire region.Mathematics was based on a sexagesimal system (ourclocks and angles still maintain that custom); writing wasperformed on clay tablets; astronomy compiled and namedstar constellations; and calendrical adjustments of lunarmonths were practised within the cycle of seasons. Venuswas adored as the star of the goddess Inanna („Mistressof Heavens“) and called “Ishtar” by the Akkadians. She . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon retained the role of being one of the most important heav-enly bodies much later when the power passed over to theBabylonians.The foundation of Babylon presumably dates back tosome time during the Sargonic dynasty. It existed as a smallvillage at about 2000 BCE till a late Sumerian king estab-lished the first dynasty there and raised it to be his capital.Thereupon it continued to grow to a cultural center.The most prominent king of the first Babylonian dyn-asty was Hammurabi. He lived in the first half of the secondmillennium. His regency lasted for 43 years, and he estab-lished the significance of the then puny city state. His ante-cessors expanded the region of influence, but Hammurabienforced domestic policies like building an infrastructure,installing ambassadors with the neighbours, construction ofreligious temples, and the famous code, an early collectionof legal rules valid for everyone. By skilful tactical man-oeuvres he conquered the former empires of Akkad, Uruk,and Ashur during his reign. Individual wealth increased,and the number of inhabitants in Babylon, too. Under hisrule it became the dominant power in Mesopotamia.The prosperity of Old Babylon lasted for ≈
300 yearstill it ceased after one dynasty only. The dynasty incorpor-ated 11 kings, of which Hammurabi was the sixth. Afterhim the empire lost influence quite rapidly. Ammi-saduqawas the fourth successor of Hammurabi, and thereafter hisson ascended the throne as 11th king before the kingdomcollapsed. Babylon was captured by the Hittite king MuršiliI. Written documents suspended afterwards.The Hittites were not able to sustain the city, becauseof its large distance from their heartland, and abandonedit. The Kassites, a tribe from the Zargos Mountains in theeast, occupied the power vacuum. Changing sovereigns fol-lowed, some of them unknown, and probably kingless years.Nevertheless, the Kassites also contributed to the later cul-ture of the Neo-Babylonian Empire that was to flourish inthe 7th century BCE. At about 1200 BCE, the Bronze Ageended with the collapse of the Hittite Empire, and the Kas-sites were overthrown at about 1160 BCE as well.
When it comes to a more exact dating, any of the historicalstages above turn out uncertain on the absolute timescale. Itwill be essential to reconcile them with other events, rulers,and dynasties of neighbouring countries. Taken the regen-cies alone, they prove inconsistent. There exist dozens oflists winding up crosswise and imprecise, patchy, and, mostof all, contradictory. Some texts incidentally contain anastronomical hint, like the sighting of a planet or a moonphase, and one tries to conflate that into a plot.In the course of research different methods of reckon-ing have been suggested, and the outcome led to a systemreferred to as “choice of chronology”. The chronologiesare an attempt to arrange some cornerstones of the Baby-lonian history with the timeline of the neighbours as theEgypts and Hittites. They are no more than a time mesh, and their emergence is a complex issue beyond the scope ofthis paper. The origin is unfurled by Weir [27] and Fother-ingham [19] in more detail. Without discussing the concept,though virtually important, a few basics are necessary to besummed up now.There are four variants for the chronology of the NearEast: long, middle, short, and ultra-short. Each defines afew fixing points like the year of accession of Hammurabior the sack of Babylon. None of the proposed chronologiesis perfect. The long one preferably agrees with astronom-ical data. For historians it causes headaches, for they do notlike it in view of their king lists. They favour the middlechronology, although it is subject to the strongest criticism.The short chronology adjusts to astronomy second best, butthe congruence of historical documents with the backreck-oning turns out of moderate use. The fourth, ultra-short onecame into being upon archaeological studies of pottery andceramics in 1998. It does not fit to astronomy at all. Tak-ing the sack of Babylon as an example, the chronologies fixthis at the years 1651, 1595, 1531, or 1499 BCE, respect-ively, see Table 1. Apart from the main chronologies thereare several other suggestions of lesser significance.The downfall of Babylon is equivalent with the datingof the so-called “Venus-Tablet” (EAE . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Table 1: Various methods on dating the fall of the Babylon-I-Dynasty. The four main chronologies are highlighted. Yearsare given in BCE. For Ammi-saduqa’s year 1 raise the year in the first column by 51.
Year Evidence or method Author see Ref.1926 Venus tablet (EAE 63) Kugler (1912) [19, 27]1870 Venus tablet (EAE 63) Fotheringham (1928) [19, 25]1733 Assyrian king list Landsberger (1954) [8]1665 Assyrian building inscriptions Eder (2004) [8]Statistics of astronomical data Huber (1982/2000) [15]1651 Venus visibility + lunar months Weir (1982) [28]1596 ( ±
7) Radiocarbon / Kültepe Manning etal (2001) [21]Excavations in Alalakh/Turkey Smith/Ungnad (1940) [27]1595 Generation count of kings Rowton II (1958) [24]Political history MacQueen (1964) [20]1587 Solar eclipse of Shamshi-Adad De Jong (2013) [6]1587 Volcano eruption on Thera/Santorini De Jong (2010) [5]1560 ( ± etal (1998) [11]1499 Eponym lists from Mesop. + Egypt Gertoux (2013) [12]1467 (suitablet chron. for EAE 63) (statistics) [15]1384 ( ± An astrologic portent on the cuneiform tablet from theseries
Enuma Anu Enlil (EAE) is our starting point to de-termine the destruction of Babylon. The series of tabletscontains hundreds of prophecies based on weather phe-nomena, shape of clouds, halos, planets, day numbers, andeclipses, of course. For example, the dark side of the ob-scured disk (West, South, etc.) was related to the directionwhat kingdom was affected by the prophecy. It would suf-fer from famine, rebellion, deluge, or the like. The oldestEAE omens go back to the Akkadian and old Babyloniantimes, most originate from the 7th century, while the young-est ones are estimated at 195 BCE [6]. It is anything butobvious to which period of history the texts apply.Tablets shurinnu was not obscured, but disappeared.On the 28th [day] you observe [the moongod] and an eclipse is close by; it begins andmakes full [its time]; it (the shurinnu ) willshow you the eclipse. Observe his eclipse,[that of] the god who in his eclipse became . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Figure 1: Reverse of the EAE Tablet visible and disappeared, and bear in mindthe north wind. — The prediction is givenfor Babylon: the destruction of Babylon isnear. The king to whom Enlil said “yes”, hispeople will be scattered. His reign will end . . .[Ur] will take away from there [the hegemonyof] Babylon. Ur will take supremacy overBabylon. If [night elapses] while the god isin eclipse: [floods] will come [in the rivers],rains in the sky, the harvest will be a success,good fortune will occur.In clearer words, the text deals with a pair of eclipses inthe time interval of 14 days. On 14th and 28th day occurreda lunar, then a solar eclipse. Shortly thereafter Babylon wasdestroyed, and the power passed over to Ur. The lunar ec-lipse began in the second part of the night (“dawn watch”)and was in progress at sunrise. The meaning of shurinnu is not fully clear. There are three possible explanations[14]: as a partial eclipse, in general; or that the moon setwhile partially darkened before it reached totality; or that itsank before being entirely restored to its full light. Someconsider shurinnu as a technical term used in Babylon forthe “crescent shaped appearance of an eclipsed moon” [1].Today we would just say “partial”.The “14th day” should not be interpreted too tight, be-cause it may stem from a systematisation of several eventsalike. Depending on the start of the month, full moon canfall on the 13th or 15th day, too, and be eclipsed [14]. De- tails about the solar eclipse two weeks later are absent. The28th day corresponds to the last visibility of the waning cres-cent at dawn and would also be subject to systematisation.Fortunately, the name of the month is given: Shabattu is the11th month of the Babylonian calendar equal to our Januaryor February.An obstacle concerns the attribution of the 11th month.The Babylonian calendar began in spring close to the vernalequinox (month I). Taking precession into account, it fellon 7th April in the middle of the second millennium [4].We do not know how good the old calendar was aligned tothe seasons. Particularly, this concerns the month XII , thetwelfth month to be counted twice, to complete the season,and we can only guess which years were affected by theinsertion. So, the new year started at some day between20th March and 20th April. The author of this paper takesthe liberty to render Shabattu (XI) between the beginning ofJanuary and the end of March. In the generous time span from 1800 to 1300 BCE thereare only seven eclipse pairs, M + S, stored in the month ofShabattu. They are listed in Table 2. Penumbral eclipsesare excluded for their invisibility.When comparing the seven years with those of Table1 for the sack of Babylon, the years 1753 and 1713 BCEappear historically too early. They will only be feasible, if . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Table 2: Pairs of eclipses for Babylon (32.5 ◦ N, 44.4 ◦ E). Times are given in local time, LT = UT + 3h, corrected for theextrapolated ∆ T .
Eclipse pair MS: Chrono-Date Time [LT] Mag. logy RemarksM: 1753, Feb 28 1:07 0,378S: 1753, Mrc 14 15:07 0,336 too early?M: 1713, Jan 08 7:17 0,467S: 1713, Jan 22 11:14 0,626 [16]M: 1659, Feb 09 6:21 0,294S: 1659, Feb 23 12:23 0,748 (long?)M: 1602, Dec 31 0:11 0,375 „last watch“ incorrectS: 1601, Jan 14 15:50 0,914 (middle?) central + annular in BabylonM: 1547, Feb 01 1:56 0,264S: 1547, Feb 15 13:53 0,809 (short?) [2]M: 1416, Jan 23 3:56 1,196S: 1416, Feb 07 13:25 0,123 too late?M: 1362, Feb 25 6:47 1,023S: 1362, Mrc 12 10:24 0,587 [22] — historically too lateM: 2161, Feb 09 3:44 1,278S: 2161, Feb 25 17:17 (0,4?) long [14]: Akkad?M: 2049, Feb 02 0:06 1,213 „last watch“ incorrectS: 2049, Feb 16 8:10 (0,7?) (short?) ([16]: alternatively 2041?)even longer chronologies are chosen than the proposed ones.At the beginning of the 20th century, when the EAE textswere deciphered correctly, scholars were struggling withfive different ways of reckoning, at least. They implied afixing point for the capture of Babylon between 1977 and1750 BCE [19]. Today those models are outdated and notused anymore, since they do not run conform to crosslinkswith Assyria or Egypt.At the other tail of the options, the pair of 1416 BCEappears too late. Wayne Mitchell prefers even the latersolution of 1362 BCE [22]. His calculation is said toprovide excellent agreement from the astronomical point ofview, but the synchronisation with archaeological evidencecollapses completely. Neither the king lists nor dendro-chronology nor the connections to Egypt harmonise withit. For example, the Hittite conqueror of Babylon, MuršiliI, would be contemporary to the Egyptian pharaoh Akhen-aten. This is absolutely incompatible with historical rami-fications. The author of this study draws upon a theory ofthree solar eclipses, in the 14th century, being responsiblefor Akhenatens’s admiration of the sun such that his life-time sets a lower limit to the choices for the Babylonian era[18]. Mitchell, however, untightens the problem ostensiblyby introducing a “super-short” chronology with all dataextremely compressed. Many consider that too fancy.The results for 1659 and 1602 BCE in Table 2 are dif-ficult to justify, for the double eclipse occurs 8 or 7 yearsbefore its respective fix point. Such large a discrepancybetween the date and the supposed sack of Babylon canhardly be explained. The former interval harboured 11 lunareclipses (8 partial and 3 total), though not in Shabattu, butthey weaken the relationship with the prophecy given on thetablet. The astrologer would rather relate the “fulfilment” of the omen to a closer eclipse and keep that for the recordrather than an event in the distance of eight years.We are left with the pair of 1547 BCE, however, it doesnot cover the scheme of the prevalent chronologies. BorisBanjevic outlined an “Upper Short Chronology” such thatthe system becomes more complicated [2]. He states thatthis solution would be consistent with the Venus Tablet ofAmmi-saduqa, but calculations by both Wayne Mitchell andPeter Huber showed, a few years before, that the contrary istrue [22, 15]. The heliacal sightings of Venus would not co-incide with the lunar months, if Banjevic’s year was chosenas the year of downfall. Furthermore, Banjevic believes thatthe solar eclipse was merely forecasted on a short timescalesubsequent to the lunar eclipse. This would mean that thesolar eclipse does not need to be an immediate one. If it wasnot promptly observed, then it may have occurred in a latermonth at a memorable space of time from the lunar eclipse.That subverts the 14-day-distance in Shabattu (further con-siderations below).The archaeoastronomer Göran Henriksson disregardsthe omen on the double eclipse and refers to the allusionof a single solar eclipse on Tablet EAE ∆ T (see Sec. 9). The central zone was as wide as 25 km,and the darkening occurred at 11 a.m. local time. Henriks-son bases his arguments on the accompanying informationon that tablet describing the plundering raid of Babylon aswell as a devastating conflagration in a quarter of the city.The dating rests upon geostratigraphy at excavation whenthe tablet was discovered. Using various additional inform-ation he builds up a bridge to a new chronology similar tothe short one. . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Figure 2: Total solar eclipse of 11 September 1558 BCEafter Espenak [9].
The two eclipses in Shabattu do not provide a satisfactoryanswer. A way out could be perhaps an inverted sequenceof the eclipse types: SM instead of MS. For this optionfive candidates can be identified for the same time span, seeTable 3.As above, the cases of the 14th century BCE, that mightapproach the super-short chronology, must be rejected ashistorically too late. At first sight, the event of 1650 BCEseems to be the sole feasible candidate to meet the longchronology. However, the interpretation comprises a goodnumber of flaws. On 14 February 1650 BCE, the observerwould have seen a solar obscuration with magnitude of0.977 in Babylon. The southern border of the totality zonepassed by at a distance of 120 km (Fig. 3). Two weekslater, on 2 March, a partial eclipse of the moon followedwith a magnitude of 0.730 happening in the evening time atmoonrise. The time of day contradicts the text, as “the lastwatch” is specified. A scribal error must be excluded, forthe subsequent two lines speak of a simultaneous observa-tion of the sun and moon. Such a threefold reinforcementcannot be ignored. The information on time receives highrelevance, and the eclipsed moon must have set at dawn, inparticular, if the word shurinnu shall make sense.But there is more to it. These two eclipses in spring of1650 BCE happened one year after the presumed downfallof Babylon. This impairs the astrological role of a “predic-tion”, so the omen itself becomes obsolete.Even more important, the alteration of the order, SM,implies that the eclipses occurred in two different Babylo-nian months. Either the solar eclipse belongs to the previousmonth, or Shabattu is grossly wrong for the lunar eclipse. IfShabattu is erroneous, the search becomes hopeless.The last resort would be a split of the eclipses into twodifferent windows of time being six months apart, at least,with only the lunar eclipse belonging to Shabattu. Such aseparation will overstrain the arguments. The number ofpossibilities increases rapidly, especially, when picking uplesser obscurations. We give an example for the dilemmathat gets close to the date of 1651 BCE (not included in the
Figure 3: Two solar eclipses accountable for the long chro-nology on the “inverted” sequence having the solar eclipsebefore the lunar. table): the total eclipse of the moon on 23 March 1652 BCEwas rather small (mag = 0.238), and the maximum obscur-ation was reached at 9 p.m. local time — contrary to thetext. The next window of opportunity opened half a yearlater: on 30 September 1652 BCE. A partial solar eclipsecould probably have been visible in Babylon (Fig. 3). Itsmagnitude of 0.530 at 4 p.m. local time was not conspicu-ous, and it could have escaped attention, either. Solar ec-lipses with mag < . ∆ T (see Sec. 9 below) does not fundamentally improvethe boundary conditions for the benefit of its visibility.The circumstances of weather are an important un-known, anyway. This is often omitted when reasoningabout eclipses. It is taken for granted that the event wasspotted, as the maps of today present a convenient courseof the moon’s shadow over the surface of the earth. But theactual sighting of an eclipse is not self-evident at all! An expert on history may probably be puzzled about the endof the apodosis of that omen. It is said that Ur took victoryover Babylon. — Is that correct?As pointed out in the historical overview, Babylon wasconquered by the Hittite king Muršili I as it was alreadydebilitated. The takeover is badly transmitted, and the se-quence of Babylonian rulers proceeds with Kassite kings. A“dark period” of unknown length between the conquest andthe subsequent monarchs is widely assumed. In contrastto that, the third and final dynasty in the city of Ur endedtwo centuries prior to the rise of Babylon. Ur was not ex-tinguished yet, when the Babylonian Empire expanded, butit lost most of its political importance. If power was to betransferred from one city state to another, then vice versa:Babylon is the younger empire. . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Table 3: Inverted sequence of eclipse pairs in Shabattu visible in Babylon between 1800 and 1300 BCE.
Eclipse pairs SM: Chrono-Date Time [LT] Mag. logy RemarksS: 1704, Jan 12 9:05 0,944M: 1704, Jan 28 19:22 0,500 at moonriseS: 1650, Feb 14 11:02 0,977M: 1650, Mrc 02 17:49 0,730 (long) at moonriseS: 1389, Feb 09 17:33 0,945 central + annular at sunsetM: 1389, Feb 24 23:01 0,224 „last watch“ incorrectS: 1351, Dec 30 9:53 0,251M: 1350, Jan 15 16:05 1,268 at moonriseS: 1335, Mrc 13 14:24 0,763M: 1335, Mrc 28 22:45 0,523 „last watch“ incorrectThe statistician Peter Huber put forward a brilliant idea[14]: the omen needs to be shifted to 2161 BCE, and thelong chronology applied to Ur. Under this perspective, hesuspects that a mistake crept in as “Babylon” will be a laterinsertion for “Akkad” (see lower part of Table 2).Historically this is conceivable, indeed. The AkkadianEmpire of Sargon waned to a northern part of Mesopotamiaand was marked by social unrests. On missing order therulership passed over to a Gutian king. The Gutians adop-ted all titles of Akkad and accommodated themselves to theMesopotamian society in many ways. History appreciatesthem negatively, for their kings were obviously unfamiliarwith the urban economics such that supplies fell short. Thatcaused uproars among the population. Civil wars were ra-ging under all Gutian kings. According to the SumerianKing List four kings changed within a three year’s time. Inthe heartland, Gutium, their reign lasted 75 years or, pos-sibly, up to 100 till the king of Uruk, Utuhegal, drove themoff from several cities in Mesopotamia. “Gutium” itself isnot identified, either. It simply denotes a region, not a city,vaguely mapped to the northeast of Akkad. The last Gutianking was Tirigan, and he is said to have been in power for40 days before being defeated and the kingship was takento Uruk.Based on this concept, that eclipse omen would dealwith the king Utuhegal (Table 4). He would have cometo power in Uruk in 2161 BCE where he founded the fifthdynasty whose sole member he was. He installed new gov-ernors in the cities under his control. For example, his son-in-law was Ur-Nammu and administered Ur for the next tenyears. In the same year or the year after, Utuhegal mighthave led the revolt against the Gutians and, finally, attainedhegemony over Akkad.The circumstances of the upheaval in Akkad are scanty,unfortunately. They are open for speculations. According tosome sources the last regular Akkadian king was Sharkali-sharri, but he lost a number of cities in the area of Sumeron account of a drought, maybe Akkad itself. After himthe king list mentions four potentates who were vying tobe king. There is no evidence from this short interval, andwe do not know anything about the contenders to the throneexcept their names. Thoughts were raised whether they be- longed to the Gutians making inroads into Akkad. Ilulu,the last one of them, was almost certainly a Gutian, beforeDudu seized power over the city after those three years ofconfusion [20].King Dudu is described as Sharkalisharri’s successor,however, without a family kinship to him. After 21 years inoffice Dudu was replaced by his son Shu-Durul. The latterkept himself on the throne for 15 years, and he is deemedto be the final ruler of Akkad. According to classical teach-ing, it was only after Shu-Durul that the Gutians took overcontrol in Akkad.The chaotic period after Sharkalisharri is doubtful inboth length and persons involved. An intriguing questionconcerns the affiliation of Dudu and Shu-Durul. Were theyAkkadians? Were they really kings or rather governors, likeUr-Nammu, appointed by Utuhegal? — Dudu’s rule waslimited to a little more than the capital itself. There areno eponym years known from his time, and, in general, itseems unlikely that he could have reigned as long as 21years. It remains unclear what person was operating inwhich city in whose charge.Form another source we learn that Utuhegal, who wonvictory over the Gutians, perished tragically in the 7th yearof his regency, instead of the tenth year, while visiting a dike[15]. After his death, Ur-Nammu, the son-in-law, seized theopportunity to ascend the throne of Ur, which he had con-trolled before as governor. At the same time he might havetaken command of Uruk itself and, possibly, the orphanedcity of Akkad that would have been part of Utuhegal’s realm.It is quite certain that Ur-Nammu founded the third dynastyof Ur, and reigned there as king for a minimum of eightyears. It may be conjectured that his regency was countedfrom his governorship instead of accession as king, result-ing 18 years in total. The commencement would be put inthe year of the double eclipse of 2161 BCE.Assigning the omen . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Table 4: Suggestion for a timeline of the kings of Akkad, Uruk, and Ur. The eclipse omen EAE 20-XI is used as an anchorpoint for 2161 BCE.
Year [BCE] “Gutium” Akkad Uruk Ur EAE-Omen. . . (?)???? Sharkalisharri EAE 21-VIII?Gutian period Uruk-IV-Dynasty. . . (total of 21 kings . . . (?) Ur-Gigir (6 years). . . in ≈ Kuda (4 years)2171 in the land of Guti)
Puzur-ili (5 years)21702169 . . . (Ka-du?)2168 Yarlaganda (7 years)2167 Si’um (7 years) Puzur-ili2166 (4 kings in
Ur-Utu (6 years)2165
Si’um | Tirigan Elulu Ur-Utu | Tirigan
EAE 20-XI
Different from Table 4, there are other ways of schedulingthe kings in Akkad. The classic draft embodies a strict ad- . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon option of the regencies of those four short-time kings. If the3 years were true, the column for Akkad would have to besqueezed and moved upward in time. All columns are slid-able in vertical direction while preserving the succession ofkings.Currently, the end of Sharkalisharri (at the “????”) isestimated at about 2210 BCE on the long chronology and2190 BCE on the middle chronology [20]. However, thetime gap between Sharkalisharri and Ur-Nammu could bevery small, as the orientalist Claus Wilcke asserts, probablyone generation only, see quotation by Huber [14, p64]. Inany case, historians would have to view the Gutian periodin a new light. Eclipse years are able to jump upwards ordownwards at the Exeligmos cycle (54 years, see [17]) withonly minor difficulties.Still, Utuhegal’s victory over the Gutians is free to becombined with a completely different omen. For instance,the omen on tablet EAE The conjecture on a modified text on EAE γ = 1,0965). The clock-time error, called ∆ T ,is estimated to 50,397 ± ∆ T is an indicator for the decelerationof the rotation of Earth caused primarily by tidal friction. The quantity denotes the difference between the strictly uni-form timescale (TT), measured with atomic clocks, and theconstantly lengthened timescale for the day (UT), which isused for our civil information on time: ∆ T = TT − UT = − + c · t [s] , where the constant c ≈
32 s/(cy) and t is given in centuries(cy) before 1800. For the detailed geoscientific backgroundon the effect of deceleration see [17] or respective literature.The value of ∆ T is known quite reliably from Antiquity( ≈
700 BCE) till now, but for epochs further back in timeit is extrapolated. The extrapolation deploying the formulaabove considers the regular and systematic slow-down thataccounts for about 2 milliseconds per day. But it does notinclude the minute random fluctuations arising from theunanticipated behaviour. For example, a climatic melt ofthe polar ice would increase the ocean level, or a strongearthquake may lead to a displacement of continents. Botheffects alter the moment of inertia of the earth, and its rota-tion is accelerated (positive or negative with respect to theaverage) in an unpredictable way. These changes are tiny,but they do accumulate over centuries to a conspicuous er-ror. Thus, when the astronomical conditions for an eclipseare fulfilled, the earth presents a different surface to thecelestial actors. A precise backcalculation in terms of localtime cannot be guaranteed; the lunar eclipse will be shiftedto a different time zone.The geographical shift takes effect on solar eclipseseven more. In our example of 25 February 2161 BCE, theuncertainty amounts to 4430 s = 74 min, meaning that thevisibility track could pass a certain geographical longitudewithin that error of time earlier or later. The eclipse itselflasts for about 90 minutes, and statements on observationalconditions become impossible: perhaps the partial eclipsewas seen at a small magnitude in Mesopotamia, perhaps itbegan after the earth has turned away to the night side. Inthe latter case it would have taken place below the horizonand must be excluded from visibility.The average ∆ T for 25 February 2161 BCE yields amaximum phase near Iceland (mag ≈ ∆ T opt of ≈ ∆ T , the sun suffered an eclipse on 29 June 2159 BCE atnoon. The totality was even exceptionally long, as darknesslasted for about 6 minutes. The assumed Akkad ( ≈ Bagh-dad?) resided at the edge of the central zone, and there wasa magnitude of ≈ . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon Figure 4: Course of the solar eclipse of 29 June 2159 BCE.The sun symbol marks the spot of maximum totality. suspicion of a deliberate amendment makes the state of af-fairs more complicated: Who was to do so and for whatreason? Also, the space of two and a half years of idle timebetween the lunar and solar eclipse remains unattractive.On the other side, the systematisation of astrologicalomens tends to support Huber’s Akkad hypothesis. In thecourse of many generations the divinations would havereceived a gradual evolution. An overall assumption isthat any kind of celestial documentation began with ec-lipses once, and a first catalogue of various portents wasput to record. Some of them would have “come true”, e.g.the death of a regent or a social misery, so the incidentsstrengthened the belief in the doom prophesied by celestialsigns. The passage on the double eclipse of EAE
10 Conclusions
We analysed the omen on EAE Tablet one attempt to shedlight on the historical stage. There are manifold pieces fromother disciplines that would cast deviant solutions. Themiddle chronology is the one that bears the worst concord-ance with astronomy, but it is so deeply rooted in the mindsof historians that it will almost certainly continue to be used,though there is no supporting evidence for it. It serves ratheras a compromise to suit as much authors as possible. Thiscomes along “democratic”, but it is not scientific.Actually, astronomy is the field that provides the hard-est facts to chronology. The radiocarbon method anddendrochronology could also provide reliable time slots, but both are subject to uncertainties of their own. It con-cerns the material available: the origin of the sample is notknown in many cases, or its attribution to the culture fails,or it might be contaminated. In addition, a much largererror in measurement (confidence interval) has to be accep-ted. Texts are sometimes subject to doubtful retrieval. Andthe interpretation itself often arises from a rather modernunderstanding than from knowledge of the ancients. Atthis point a combination of astronomical, archaeological, aswell as historical evidence will be of high value.
Acknowledgements
This paper is an excerpt from the Habilitation submitted tothe University of Heidelberg, Germany [17]. The unpaid re-search was accomplished under direful circumstances. Therelevant section is revised and now published via arXiv without assessment.
Appendix
Further texts (abbreviated) from the EAE tablets assignedin Table 4. Dates proposed for lunar eclipses after [14].
EAE 21-VIII: “If an eclipse occurs on the 14th day ofArahsamna (month VIII) . . . The prediction is given for theking of the world: Either the king will die, or a large armywill fall, or a large army will revolt.”
Possible context:
The decline of the Akkadian Empire setin with Sharkalisharri who lived about 150 years after Sar-gon. Control was lost over large parts of the country, whileneighbours increased political and military pressure. — Al-ternatively, that omen can be made conform with Utuhegal’sdeath, see below EAE 20-IV.
Lunar Eclipses:
EAE 20-IV: “If an eclipse occurs on the 14th of Dumuzu(IV) . . . The king who ruled will die. . . . The prediction isgiven for Ur. . . . The grandson, descendant of the king, willseize the throne. . . . The king together with his clan will bekilled.”
Possible context:
A skip from a king to his grandson isunknown for the entire dynasties of Akkad and Ur-III. Allkings were sons of their predecessors. At best, a transitionto a brother did take place. Arguable is also the death ofUtuhegal, as he was followed by his son-in-law Ur-Nammu.
Lunar eclipses:
EAE 21-I: “If an eclipse occurs on the 14th of Nisannu(I) . . . The prediction is given for the king of Akkad. Theking of Akkad will die. If the eclipse does not affect theking: There will be destruction and famine. The people will . Khalisi (2020) : The Double Eclipse at the Downfall of Old Babylon send their children out to the market (to be sold). The greatcountry will go to the small country for food.” Possible context:
This could be understood as the end ofthe Akkad dynasty. The two final kings were Dudu and hisson Shu-Durul. However, their regencies do not fill well tothe timeline, see Sec. 7.
Lunar eclipses:
References [1]
Al-Rawi F.N.H. & George A.R. (2006) : “Tablets from theSippar Library XIII: Enuma Anu Ellil XX”,
Iraq 68 , 2006,p23–57; [2]
Banjevic, Boris (2006) : “Ancient eclipses and dating thefall of Babylon”,
Publications of the Astronomical Obser-vatory of Belgrade 80 , 2006, p251–257[3]
Boese, Johannes (2012) : “Shamshi-Adad I, Hammura-bi und die Chronologie. . . ”,
Mitteilungen der Dt. Orient-Gesellschaft zu Berlin 144 , 2012, p17–29 (in German)[4]
Cornelius, Friedrich (1969) : „Das Alter der babylonischenAstronomie “,
Zeitschrift der Deutschen MorgenländischenGesellschaft , Supplementa 1, ed. by Voigt, Franz SteinerVerlag, Wiesbaden, 1969, p170–173 (in German)[5] de Jong T. & Foertmeyer V. (2010) : “A new look at theVenus observations of Ammisaduqa”,
Jaarbericht „Ex Ori-ente Lux“ 42 , 2010 p141–157[6] de Jong, Teije (2013) : “Astronomical Fine-Tuning of theChronology of the Hammurabi Age”,
Jaarbericht van hetVooraziatisch-Egyptisch Genootschap „Ex Oriente Lux“ 44 ,2013, p147–167[7] de Jong, Teije (2013) : “Review of »Die Venustafelndes Ammi-saduqa. . . «”, Resume of Mebert’s publication,
Journal of the American Oriental Soc. 133 , 2013, p366–370[8]
Eder, Christian (2004) : “Assyrische Distanzangabenund die absolute Chronologie Vorderasiens”,
Altorien-talische Forschungen 31 , No. 2/2004, p191–236; doi:10.1524/aofo.2004.31.2.vii (in German)[9]
Espenak, Fred (2006) : “Predictions for Solar andLunar Eclipses”, Eclipse maps courtesy of Fred Espenak,NASA/Goddard Space Flight Center; [10]
Furlong, Pierce James (2007) : “Aspects of Near EasternChronology”, PhD thesis at University of Melbourne/AU,2007, esp. Tab. L, p300; http://hdl.handle.net/11343/39353 [11]
Gasche H., Armstrong J.A., Cole S.W., Gurzadyan V.G.(1998) : “Dating the Fall of Babylon”,
Mesopotamian His-tory and Environment , Ser. 2, Mem. 4, Ghent/Chicago, 1998[12]
Gertoux, Gérard (2013) : “Dating the fall of Babylonand Ur thanks to Astronomical Events”, eprint arXiv:1309.2758 [physics.hist-ph] [13]
Henriksson, Göran (2005) : “Astronomical Dating of theOld Babylonian Kingdom”, in: „Cosmic Catastrophies“ ed.by Koiva, Pustolnik & Vesik, Tartu/Estonia, 2005, p63–70;ISBN 9949-418-32-1[14]
Huber, Peter J. (1999) : “Astronomical Dating of Ur III andAkkad”, Archiv für Orientforschung 46/47, 1999, p50–79[15]
Huber, Peter J. (2000) : “Astronomy and Ancient Chrono-logy”,
Akkadica 119/120 , 2000, p159–176[16]
Kelley D.H. & Milone E.F. (2011) : “Exploring An-cient Skies”, Springer Science + Business Media, NewYork/USA, 2011; ISBN 0-387-95310-8[17]
Khalisi, Emil (2020) : “Das Buch der legendären Finster-nisse”, Habilitation submitted to the University of Heidel-berg, 2020, ch. 10.2 (in German)[18]
Khalisi, Emil (2020) : “The Solar Eclipses of thePharaoh Akhenaten”, eprint arXiv:2004.12952[physics:hist-ph] , 27 April 2020, 5 pages[19]
Langdon S., Fotheringham J.K., Schoch C. (1928) : “TheVenus Tablets of Ammizaduga”, Oxford University Press,London/UK, 1928[20]
MacQueen, James G. (1964) : “Babylon”, Robert HaleLtd., London/UK, 1964, p27[21]
Manning S.W., Kromer B., Kuniholm P.I., Newton M.W.(2001) : “Anatolian Tree Rings and a New Chronology forthe East Mediterranean Bronze-Iron Ages”,
Science 294 , 21Dec 2001, p2532–2535; doi: 10.1126/science.1066112[22]
Mitchell, Wayne A. (1990) : “Ancient Astronomical Obser-vations and Near Eastern Chronology”,
Journal of AncientChronology Forum 3 , 1990, p7–26[23]
Rowton, Michael B. (1952) : “The Date of the HittiteCapture of Babylon”,
Bulletin of the American Schoolsof Oriental Research 126 , Apr 1952, p20–24; doi:10.2307/3218987[24]
Rowton, Michael B. (1958) : “The Date of Hammurabi”,
Journal of Near Eastern Studies 17 , Apr 1958, p97–111[25]
Schoch, Carl (1925) : “Die erste Dynastie von Babylon”,
Klio 20 , 1925, p107–109; doi: 10.1524/klio.1926.20.20.107(in German)[26]
Waerden, Bartel L. van der (1968) : “Erwachende Wis-senschaft”, Band II, Springer Verlag, Basel/CH, 1968/2019;doi: 10.1007/978-3-0348-4127-6 (in German)[27]
Weir, John D. (1972) : “The Venus Tablets of Am-mizaduqa”,
Pihans 29 , Nederlands Instituut voor het NabijeOosten, Leiden/NL, 1972[28]
Weir, John D. (1982) : “The Venus Tablets — a Fresh Ap-proach”,
Journal for the History of Astronomy 13 , 1982,p23–49; doi: 10.1177/002182868201300103[29]
Wilhelm G. & Boese J. (1987) : “Absolute Chronologieund die hethitische Geschichte. . . ”, in:
International Col-loquium on Absolute Chronologie , ed. by Paul Aström,Gothenburg, 1987, p74–117, insb. p101 (in German), ed. by Paul Aström,Gothenburg, 1987, p74–117, insb. p101 (in German)