Comments on Chapter II:

"The Litmus Test of the Absolute Chronology"


II-A.  Using astronomical tablets for establishing absolute dates 


In this chapter, Furuli discusses using astronomical tablets to establish an absolute chronology. In view of the varied quality and state of preservation of the Babylonian astronomical tablets, not all are usable for chronological purposes. Accordingly, Furuli states that each tablet must meet "two fundamental requirements." What are they?


II-A-1:  Furuli's criteria for the chronological use of astronomical tablets  


The first requirement is the following:


"A.  The positions of the heavenly bodies must be observed by the eye of a scribe and written down at the same time; and they must not only represent backward calculations made at a  much later time."


This criterion is quite in order. The value of the next requirement, however, is dubious:


"B.  The name of the ruling king must have been written on the tablet at the time when the observations were made."


One problem with this criterion is that it is unrealistic. Furuli admits that:


"because clay hardly can be kept moist for 12 months, the observations must have been written down on quite a lot of smaller tablets, which were copied when the original was made." (p. 30)


Modern scholars who take notes on paper face a similar task of collating their notes. Suppose a scholar is reviewing a book, and on page one of his notes he records the name of the book. Then he scribbles various comments on items of interest. The notes run to many pages, but he does not record the title of the book on every page. When he is finished reading, perhaps months later, he collates and condenses the scribbled notes and writes a neat summary. Does the fact that he failed to write the book's title on every single page of the notes invalidate the summary? Of course not. In like manner, if the name of a ruling king is not written down on "smaller tablets, which were copied when the original was made," it certainly does not invalidate the observations transferred to the final tablet, which is subsequently viewed as the original. Furuli's criterion B, then, is absurd.

It is transparently obvious that Furuli invented criterion B to disqualify tablets that could otherwise be used to invalidate his Oslo Chronology. Usually, the royal name is given only at the beginning of each tablet. But if a tablet is damaged and the beginning part is missing, the date connected with each observation recorded is given as the regnal year, the month, the day, and perhaps the part of the night, with no royal name. However, the observations might well be so detailed that the observed events can still be identified and dated to particular Julian years. This is often enough to identify the ruler, even if his name is missing. A couple of examples serve to illustrate this.


II-A-2.  The planetary tablets No. 54 and No. 56


Two tablets that do not meet Furuli’s second requirement (B) are LBAT 1393 and LBAT 1387+1486+1388, published as Nos. 54 and 56 in Hunger's ADT, Vol. V. Both are planetary texts that unequivocally overthrow Furuli's alternative reigns for Darius I and Artaxerxes I. Furuli gratuitously dismisses both tablets (pp. 37, 118, 211, and 227) for erroneous, specious, and illusory reasons. I examine his statements in detail later in this review.


Text No. 54 records observations of Jupiter dated to several regnal years of a king whose name is not preserved. The preserved regnal year numbers are 23 on the obverse side and  8, 19, 20, 31, and 32 on the reverse side. The ruler whose reign is treated on the reverse side must have had a long reign because the last preserved regnal year is 32. The observations recorded for these five years can be safely dated to years 514, 503, 502, 491, and 490 BCE. The observations on the obverse side dated to year 23, however, are too badly damaged to be usable.


The second text, No. 56, records about 80 preserved positions of Venus, half of which are related to Normal Stars. The data are arranged in 8-year-groups and 8 columns. The positions are dated to about 20 different regnal years (most of them fully legible or identifiable as part of the overall arrangement) that can be fixed to specific Julian years within the 70-year period from 463/2 to 393/2 BCE. The first king in this period must have had a very long reign because the highest preserved regnal year for him is 39. The observations recorded for this year can be dated to 426/5 BCE. The reason the royal names are missing in both texts is that these parts of the tablets are broken.


II-A-3.  How tablets 54 and 56 make mincemeat of Furuli's Persian chronology


All Julian dates pointed to by tablets 54 and 56 fall within the reigns of Darius I, Artaxerxes I, Darius II, and Artaxerxes II, not only according to the traditional chronology but also according to Furuli's Oslo Chronology. These tablets, therefore, can be used to challenge his alternative chronology for these reigns. It turns out that Furuli's attempts to push the reign of Darius I one year forward and the reign of Artaxerxes I 10 years backward are effectively blocked by these two tablets. The Jupiter observations dated in year 32, for example, clearly belong to year 490 BCE, not year 489 as required by Furuli's revised chronology. In fact, none of the observations dated to specific months and days in the Babylonian luni-solar calendar can be moved forward or backward in the way Furuli's revisions require.


Jupiter's period of revolution is close to 12 years, which means that on average its position among the stars changes about 30 degrees a year. However, the apparent movement among the stars displays stationary points and even reversals of motion. Tablet 54 illustrates this by saying that in year 31, month VI, on day 28, Jupiter "became stationary in [the constellation of] Gemini." This was exactly the position it held on October 4, 491 BCE, so this date corresponds to day 28 of month VI in the Babylonian calendar. A year later, Jupiter had moved about 30 degrees to a new position between the constellations Leo and Cancer. The recorded position, then, does not allow the 31st year of Darius I to be moved one year forward. The Jupiter phenomena do not repeat themselves at the same date within the lunar month for another 71 years, the fact of which the Babylonian astronomers were fully aware. Therefore, tablet 54 cannot be assigned to any reign other than that of Darius I. The Jupiter positions in tablet 54 dated to the other four regnal years just as inexorably block any attempt to change the absolute chronology established for Darius' 36-year reign.


Venus, with a period of revolution of only 224.7 days, returns to its position in relation to various stars and constellations in less than a year. However, it does not return to the same position at the same time of the year—not after one year or after 10 years. Such returns occur at 8-year intervals, after 13 revolutions (8x365.2422 = 13x224.7). The observations on tablet 56, then, which are dated to specific regnal years, months and days, cannot be fitted into a chronology for the reign of Artaxerxes I that is moved 10 years backward.


It might be argued that the observations on the two tablets could belong to kings whose reigns fell in entirely different centuries. But such alternatives are limited to kings whose reigns lasted at least 32 years (the highest preserved regnal year in the Jupiter text No. 54) and 39 years (the highest preserved regnal year in the Venus text No. 56).


Within the period to which all extant Babylonian observational astronomical cuneiform texts belong (except for the Venus tablet of Ammisaduqa)—i.e., from the middle of the 8th century BCE to the 1st century CE—only five kings are known to have ruled that long or longer: the Assyrian king Ashurbanipal (42 years), the Babylonian king Nebuchadnezzar (43 years), and the Persian kings Darius I (36 years), Artaxerxes I (41 years), and Artaxerxes II (46 years). Another possibility is that the regnal years could refer to years in the Seleucid era (counted from 312/11 BCE).


By using a modern astro-program (Chris Marriott’s SkyMap Pro 10), I have checked all the alternatives to the reigns of Darius I and Artaxerxes I—and also the alternative chronologies for these reigns suggested by Furuli's Oslo Chronology—and found them all to be impossible. The planetary observations combined with the regnal years and the dates in the Babylonian luni-solar calendar fit only the traditional chronologies established for the reigns of Darius I and Artaxerxes I.


II-A-4.   Attempts to invalidate tablets 54 and 56


Tablets 54 and 56 do not meet Furuli’s second requirement (B), but he attempts to undermine the strength of their evidence in other ways.


On page 37, Furuli refers to tablet No. 54 (LBAT 1393) and states that there "may be different factors, which contribute to the misreading of a tablet due to lack of capacity." He quotes a statement about tablet 54 by Hunger:


"The following reconstruction of the tablet was proposed by C.B.F. Walker, who notes that any discrepancies between the years attested on this tablet and the dates reported by A. Sachs in LBAT, p. xxix are to be explained by the fact that the tablet was not baked and cleaned until 1978." (ADT, Vol. V, p. 158)


Isolated from context, this seems to indicate that the translation of the tablet was a mere proposed reconstruction and that it might have been misread "due to lack of capacity." This seeming indication is wrong.


Walker’s reconstruction is not an attempted translation of the preserved part of the tablet. It is a suggested reconstruction of the chronological scheme of the original, undamaged tablet, which might have covered all the 48 regnal years from 536/5 to 489/8 BCE arranged as a series of 12-year cycles. The reconstruction is shown in a table on page 159 of ADT V. The actual transliteration and translation of the tablet, with its preserved dates, observations, etc., follows on pages 160-165, after the table.


The regnal years that Sachs had read on the tablet (LBAT, 1955, p. xxix) before it was baked and cleaned in 1978 were not misreadings that conflict with the dates read after the cleaning. The "discrepancies" referred to are additional dates that became legible after the cleaning, dates that increase the chronological value of the tablet. The way Furuli refers to this tablet is thoroughly misleading.


Furuli mentions tablet No. 56 in three places in his book, on pages 118, 211, and 227. One reason for this spread seems to be that the tablet consists of three pieces, LBAT 1387, 1388, and 1486 (also listed by Hunger as A, B, and C), which Furuli tends to deal with separately and in different places in his book. The first two pieces (A + B) contain much information, so much in fact, that Hunger’s translation of them covers 10 large pages in ADT, Vol. V. Almost all the observations preserved on the two pieces are dated to various regnal years of Artaxerxes I, the only exception being one dated to year 6 of his successor, Darius II. Piece C, on the other hand, is a very small fragment, and Hunger’s translation of it covers only half a page. No regnal year numbers are preserved on it. Hunger writes (ADT , p. 172) that the observations recorded on it probably refer to years 5 and 12 of Artaxerxes II (400 and 393 BCE).


Furuli focuses exclusively on piece C in his description of tablet 56 on page 211, implying that Hunger’s description of this little fragment applies to the whole text:


"The planetary text consisting of the three pieces LBAT 1387, 1486 and *1388 is supposed to list Venus data between -462/61 and -392/91. This text is quite fragmentary. One scholar made this comment: ’of C, the obverse probably refers to Artaxerxes II year 5, the reverse to year 12. The astronomical information preserved fits this date, especially a close encounter of Venus and a Leonis in month III of Art II year 5.’ 


These words are rather cautious, indicated by the adverb ’probably.’ As a matter of fact, neither Venus nor any other planet is mentioned on C, Obv. and Rev. An interpreter may feel there are clues for identifying Venus, but none are mentioned. So there are problems with this text in connection with the making of an absolute chronology."


Furuli does not talk about the extensive information in pieces A and B, leaving the reader with the impression that the entire Venus tablet is as fragmentary and problematic as piece C. In a discussion on page 118, he makes some comments about piece A (1387) but these, too, are aimed at undermining the strength of the text. He erroneously claims that on this tablet "years 15, 27, 35 are clearly visible, but no other years," whereas in fact eight regnal years are visible on the text, namely, years 7, 15, 23, 27, 31, 35, 39 (of Artaxerxes I), and year 6 (of his successor Darius II). For example, Furuli points out that in T. G. Pinches' copy of the tablet published by Sachs in 1955, "the number '7' is shaded and not clearly seen." But as Sachs explains (LBAT, p. vii), Pinches copied from tablets that usually had not been oven-fired, and that "it is to be expected that improved readings will result from oven firing." Hunger’s translation indicates that number 7 is now clearly seen on the tablet, which may be a result of this. The observations recorded for year 7 in months I, II, III, IV, V, and VI all fit the 7th year of Artaxerxes I, 458 BCE. Further, Furuli fails to mention that number 7 is required by the arrangement of the data in 8-year-groups. It is followed horizontally in the next columns by year numbers 15, 23, 31, 39, and year 6. The 8-year intervals, of course, refer to the periodicity of Venus positions.


About the same number of years (in the reign of Artaxerxes I) paired at 8-year intervals are visible in piece B (1388)—years 4, 5, 12, 13, 20, 21, 28 and 2[9]. On page 227, Furuli says that piece B is in conflict with the Oslo Chronology, but his only explanation is that "the regnal years written by the scribe need not be correct." This desperate theory is discussed in section II-C below.


Tablets 54 and 56 are disastrous for Furuli's revised Persian chronology, and he knows it. That is why he wants to get rid of them by every possible expedient. And that is also why he wants to undermine the trustworthiness of the astronomical tablets in general by indicating that they probably mainly contain calculations, not actual observations.


II-B.  Are most astronomical positions calculated rather than observed? 


The "most acute problem for making an absolute chronology based on astronomical tablets," Furuli claims, is that many, "perhaps most positions of the heavenly bodies on such tablets, are calculated rather than observed." (p. 15)


Is it possible that the Babylonian astronomers could retrocalculate all or most of the phenomena recorded on astronomical tablets? Are there indications in the recorded data that they did just that?


As discussed in GTR4, Ch. 4, Babylonian astronomers recognized the various cycles of the sun, the moon, and the five planets visible to the naked eye. It is clear that at an early stage they were able to predict or retrocalculate certain phenomena, such as the occurrences of lunar eclipses and certain planetary positions. Does this mean, then, that all or most of the phenomena recorded on the astronomical tablets might have been computed rather than observed, as Furuli claims?


II-B-1.  Phenomena that Babylonian astronomers were able to calculate


In support of the idea that most of the recorded positions of the heavenly bodies on the astronomical tablets might have been calculated rather than observed, Furuli presents on page 39 three isolated quotations. All but the first of the references and footnotes are confused, incomplete, or wrong.


The first quotation, taken from Bertel L. van der Waerden's work, Science Awakening (Vol. II, 1974, pp. 281, 282), deals with the ability of the Babylonian astronomers to calculate the time that a planet entered a certain zodiacal sign or the position it held when it could not be observed because of clouds or because it was too near to the sun. These calculations presuppose that Babylonian astronomers had worked out theories for dating and locating certain recurring planetary phenomena and had tables at hand that listed planetary positions at regular intervals. Such lists, which were termed "ephemerides" by Professor Otto Neugebauer, are called "cardinal tables" by van der Waerden. All extant tables of this kind are late, almost all dating from the 3rd to the 1st centuries BCE.


The next quotation, erroneously ascribed to van der Waerden, is actually from Otto Neugebauer's three-volume work, Astronomical Cuneiform Texts (1955, Vol. II, p. 281). Neugebauer's work does not deal with the observational tablets but is exclusively devoted to the arithmetical astronomical texts (mainly the tables of ephemerides mentioned above) from the last few centuries BCE. It is in his discussion of such texts that Neugebauer points to "the minute role played by direct observation in the computation of the ephemerides," a statement that Furuli greatly stresses by repeating it in extra bold type in a box on the page. What does Neugebauer really mean?


To be able to work out theories about the regular occurrence of planetary phenomena, the Babylonians needed numerous observations of the planets over long periods. Such observations were provided by the astronomical archives available since the middle of the 8th century BCE. When planetary theories were finally worked out, planetary tables could be used for calculating planetary positions when direct observations were not possible. Astronomical observational tablets, therefore, such as diaries and planetary texts, contain observations as well as occasional calculations. This is pointed out by van der Waerden in Furuli's 3rd quotation.


In this quotation, van der Waerden speaks of the difficulty of deciding "whether text data were observed or calculated." Furuli does not explain that van der Waerden is discussing a text that Furuli, on page 128, claims to be "the tablet which is most important for Persian chronology, Strm Kambys 400." Van der Waerden's statement is particularly applicable to this text, which seems to contain mainly calculations. Some scholars even question whether it records any observations.


It is clear that Babylonian astronomers could calculate a number of astronomical phenomena. At an early stage, they were using the Saros cycle for calculating and predicting the occurrences of lunar eclipses. As shown by the later ephemeride tables, they also learned how to calculate and predict the occurrences of certain periodic planetary phenomena such as first and last visibilities, stationary points, and retrogradations. But does this mean that they were able to calculate or predict all the different astronomical phenomena reported on the observational tablets? 


II-B-2.  Phenomena the Babylonian astronomers were unable to calculate


Although the Babylonian astronomers were able to calculate and predict certain astronomical events, the observational texts—diaries, planetary texts, and eclipse texts—contain reports of several phenomena and circumstances connected with the observations that could not have been calculated. 


That the diaries usually record real observations is shown by their reports of climatological phenomena. For example, the scribes repeatedly report when bad weather prevented astronomical observations. We often find reports about "clouds and rain of various sorts, described in detail by numerous technical terms, as well as fog, mist, hail, thunder, lightning, winds from all directions, often cold, and frequent 'pisan dib', of unknown meaning but always associated with rain." (Professor N. M. Swerdlow, The Babylonian Theory of the Planets, Princeton University Press, 1998, p. 18) Other recorded phenomena were rainbows, solar halos and river levels. None of these could have been retrocalculated much later. What, then, about the astronomical phenomena?


Discussing the various planetary phenomena recorded in the texts, Swerdlow observes:


"Conjunctions of planets with the moon and other planets, with their distances, could neither be calculated by the ephemerides nor predicted by periodicities." (Swerdlow, op. cit., p. 23)


Swerdlow further explains:


"The distances of planets from normal stars could be predicted," but "there was no way of predicting distances of the moon from planets or of planets from each other." (ibid., p. 173)


Note that VAT 4956 records a number of such— for the Babylonian astronomers— unpredictable and incalculable phenomena.


What about the lunar eclipse reports? Could they have been computed at a later date? In referring to the 18-year texts (the "Saros cycle texts"), Furuli uses the term "Saros tablets," but he does not make it clear whether he is referring to all extant 18-year tablets (about a dozen) or to a particular group of such texts. On page 40, he mentions two sets of tablets that use the 18-year Saros scheme. The first, he says, covers the period from 747 to 315 BCE. His footnote 51 shows that the set consists of lunar eclipse tablets LBAT 1413–1417 and 1419 (= Nos. 1–4 in Hunger, ADT, Vol. V). The other "set" he mentions is actually just one tablet that scholars generally refer to as the "Saros Tablet," BM 34576 (= No. 34 in ADT, Vol. V). It covers the 468-year period from 567 to 99 BCE.


But Furuli does not explain that the first of his two "sets" is a series of observational texts that record both observed and predicted lunar eclipses at 18-year intervals, whereas the Saros Tablet belongs to a small group of five theoretical texts that do not record any lunar eclipse observations at 18-year intervals but contain only tables of royal names and dates at 18-year intervals. (See John M. Steele in ADT, Vol. V, pp. 390-393.) The Saros Tablet does show some traces of a possible eclipse report, but this appears at the bottom of the reverse side after the 18-year table. It is written at right angles to the main text  and is clearly separated from it.


Despite this basic difference between the observational and theoretical 18-year tablets, Furuli seems to regard all of them as "hypothetical tablets," which is incorrect. In addition, his use of the plural term "Saros tablets" is confusing, as he does not clearly explain which 18-year texts he is referring to apart from the Saros Tablet, BM 34576.


With respect to the eclipse observations reported on the lunar eclipse tablets, including the Saros cycle tablets (discussed in GTR4, Ch. 4, C), the Babylonian astronomers were certainly able to predict and retrocalculate the occurrences of lunar eclipses, but they were unable to predict or calculate a number of important details about them. This is discussed by Dr. John M. Steele in his work, Observations and Predictions of Eclipse Times by Early Astronomers (Dordrecht, Boston, and London: Kluwer Academic Publishers, 2000) and in the article, "Eclipse Prediction in Mesopotamia." (Archive for History of Exact Sciences, Vol. 54, 2000, pp. 421-454)


Commenting on the claim that the eclipse records on the lunar eclipse tablets might be retrocalculations by Babylonian astronomers in the Seleucid era, Steele explains:


"You were absolutely right when you argued that the Babylonians could not have retrocalculated the early eclipse records. The Saros cycle could have been used to determine the date of eclipses, even centuries earlier, but none of the Babylonian methods could have allowed them to calculate circumstances such as the direction of the eclipse shadow, the visibility of planets during the eclipse, and certainly not the direction of the wind during the eclipse, which we find in early reports (eg Text No 3 in Hunger's latest book states that the eclipse shadow crossed the moon's surface in a southerly direction during the eclipse in Bel-ibni's 1st year [Obv, I, 2-5], and Obv II, 1-7 says that the west wind blew during the eclipse of BC 686 Oct 15). Although the Babylonians could calculate the time of the eclipses, they could not do so to the same level of accuracy as they could observe—there is a clear difference of accuracy between eclipses they said were observed and those they say were predicted (this is discussed in my book), which proves that the 'observed' eclipses really were observed.


It is true that the Saros Canon texts published most recently by Aaboe et al in 1991 are retrocalculated—but they are theoretical texts and should be considered separately from the observational material of the Diaries and the eclipse texts in Hunger's book. The observational material alone is enough to confirm Parker & Dubberstein's chronology, with only very minor, and non-cumulative, corrections."  (Communication Steele to Jonsson, March 27, 2003)


II-B-3.  Most of the contents of the observational texts are observations


Although the observational texts, due to particular circumstances such as bad weather, occasionally contain calculated events, most of the entries are demonstrably based on actual observations. That this is the case with the Diaries is directly indicated by the Akkadian name engraved at the end and on the edges of these tablets: natsaru sha ginę, which means "regular watching." (Sachs/Hunger, ADT, Vol. I, p. 11)


Scholars who have examined these tablets in detail agree that they contain mostly genuine observations. Professor Hermann Hunger gives the following description of the various kinds of astronomical data recorded in the Diaries:


"Lunar Six [i.e., the time differences between the settings and risings of the sun and the moon just before and after opposition]; planetary phases, like first and last visibility … conjunctions between planets and the so-called Normal Stars … eclipses; solstices and equinoxes; phenomena of Sirius. Toward the end of the 3rd century B.C., Diaries begin to record the dates when a planet moved from one zodiacal sign into another. The rest of the Diaries' contents is non-astronomical."


Hunger adds:


"Almost all of these items are observations. Exceptions are the solstices, equinoxes, and Sirius data, which were computed according to a scheme ... furthermore, in many instances when Lunar Sixes, lunar or solar eclipses, or planetary phases could not be observed, a date or time is nevertheless given, marked as not observed. Expected passings of Normal Stars by the moon are sometimes recorded as missed because of bad weather, but never is a distance between moon and Normal Star given as computed." (Hermann Hunger in N. M. Swerdlow (ed.), Ancient Astronomy and Celestial Divination, London: The MIT Press, 1999, pp. 77, 78; emphasis added)


Steele similarly concludes:


"Most of the contents of the Diaries represent observations; however, where observations were unavailable, for example because of bad weather or because an event was expected to occur at a moment when the heavenly body was below the horizon, then predictions were entered in their place. In addition, some data recorded in the Diaries, such as solstices and equinoxes, were always predicted." (John M. Steele, "Eclipse Prediction in Mesopotamia," Archive for History of Exact Sciences, Vol. 54, 2000, p. 429; emphasis added)


Whether an entry is based on observation or calculation is often directly stated in the text itself. In the eclipse reports, this is usually indicated by the terminology. Steele explains:


"As a general rule, eclipse predictions can be distinguished from observations by the terminology used:  sin AN-KU10 denotes an observed eclipse of the moon, whereas the opposite order, AN-KU10 sin, refers to a predicted lunar eclipse (for solar eclipses sin is replaced by šamáš). Furthermore, predicted eclipses are usually described as being šá DIB meaning that they would be omitted when the luminary was below the horizon, or ki PAP NU IGI meaning 'watched for, but not seen' when the anticipated eclipse failed to appear."  (ibid., p. 429)


In summary, Furuli's claim that "perhaps most positions of the heavenly bodies on such tablets, are calculated rather than observed" is groundless. It is refuted by statements in the tablets themselves and by the fact that they contain data that the Babylonians were unable to calculate. These circumstances are diametrically opposed to the suggestion that the data in the astronomical diary VAT 4956 might have been calculated later so that possibly "there never was an 'original tablet'." (Furuli, p. 30)


Furuli elaborates on this mistaken idea on page 40. Pointing out that VAT 4956 and Strm Kambys 400 "have the characteristics of being copies," he then goes on to consider "possible ways that such copies could be made by a scribe in the 2nd century B.C.E." He imagines that a scribe could make up such tablets by using "three different schemes that were at his disposition:" 1) a scheme of 18-year Saros cycles; 2) a scheme of regnal years of consecutive kings going backward in time, and 3) a scheme of intercalary months. Then he states: "By a combination of these three schemes, no observation was necessary, but a sophisticated chronology could be made for hundreds of years backward in time."


As was demonstrated above, the theory that VAT 4956 and other observational texts could have been made up at a much later time is nothing but a wild imagining. The idea is just wishful thinking based on insufficient knowledge of the astronomical tablets.


II-C.  A theory of desperation  


If the entries on the observational tablets—diaries, and lunar and planetary tablets—record mostly demonstrably genuine observations, and if the Babylonian astronomers were unable to compute and retrocalculate many of the astronomical and other data reported, how, then, is it possible for anyone to wriggle out of the evidence provided by these tablets?


Because the tablets often contain so many detailed observations dated to specific regnal years that they can be safely fixed to particular Julian years, the only escape is to question the authenticity of the regnal year numbers found on the tablets.


This is what Furuli does. He imagines that "a scribe could sit down in the 2nd century and make a tablet partly of some phenomena covering many years, partly on the basis of theory (the three schemes) and partly on the basis of tablets from a library" that might show real observations. Then, upon discovery that the dates on the library tablets conflicted with the theoretical data, "these erroneous data could be used to 'correct' the correct data of his library tablet, to the effect that the tablet he was making would contain wrong data of regnal years." (Furuli, p. 41)


Furuli indicates that not only the dates on the lunar and planetary tablets but also the dates on the diaries might have been tampered with by the Seleucid scholars in the same way. Referring again to the fact that the earliest extant diaries are copies, he says:


"But what about the regnal year(s) of a king that are written on such tablets? Have they been calibrated to fit an incorrect theoretical chronological scheme, or have they been copied correctly?" (Furuli, p. 42)


Furuli realizes, of course, that his Oslo Chronology is thoroughly contradicted by the Babylonian astronomical tablets. That is the reason he proposes, as a last resort, the theory that these tablets might have been redated by Seleucid sholars to bring them into agreement with their own supposed theoretical chronology for earlier times. Is this scenario likely? What does it imply?


II-C-1.  The scale of the supposed Seleucid chronological revisions


To what extent does Furuli's Oslo Chronology differ from the traditional chronology? In a chronological table on pages 219-225 covering the 208 years of the Persian era (539–331 BCE), Furuli shows, reign by reign, the difference between his chronology and the traditional one. It turns out that the only agreement between the two are the dating of the reigns of Cyrus and Cambyses—the period from the fall of Babylon (539 BCE) to 523/2 BCE, a period of 17 years. By giving Bardiya one full year of reign after Cambyses, Furuli moves the whole 36-year reign of Darius I one year forward, as mentioned earlier. Then he moves the reigns of Darius' successors Xerxes and Artaxerxes I 10 years backward by adding 10 years to the reign of the latter, creating a coregency of 11 years between Darius I and Xerxes.


But Furuli also assigns a one-year reign to the usurper Sogdianus between Artaxerxes I and Darius II. The effect of this is that the remaining reigns up to 331 BCE are all moved one year forward. The end result is that Furuli's Oslo Chronology is at variance with the traditional chronology for the Persian era for 191 of its 208 years, or for 92 percent of the period.


But this is not all. As mentioned in the introduction, Furuli wants to add 20 extra years to the Neo-Babylonian period somewhere after the reign of Nebuchadnezzar—between 562 and 539 BCE. The effect of this—what Furuli calls the "domino effect"—is that not only the reign of Nebuchadnezzar but all the reigns of his predecessors are moved backward 20 years.


Because the Babylonian astronomical archive starts with the reign of Nabonassar, 747-734 BCE, Furuli's Oslo Chronology is at variance with the traditional chronology for most, if not the whole, of the Babylonian era from 747 to 539 BCE. This means that the disagreement between the two runs to more than 90 percent of the 416-year period from 747 to 331 BCE. This also means that the Oslo Chronology is contradicted by more than 90 percent of the astronomical observational texts—diaries, eclipse texts, and planetary texts—dated to this period. Because these tablets record thousands of observations dated to particular regnal years, months, and days within this period, we begin to get some idea of the scale of the chronological revisions the Seleucid scholars must have engaged in—according to Furuli's theory. Yet, this is only a fraction of the full scope of the necessary revisions.


II-C-2.  The scope of the original astronomical archive 


It should be kept in mind that the archive of ca. 1300 nonmathematical and principally observational astronomical cuneiform tablets is only a fraction of the scope of the original archive available to the Seleucid scholars. In a lecture held at a conference in 1994, Professor Hunger explained:


"To give you an idea of how much was originally contained in that archive, and how much is still preserved, I made a few rough estimates. From well preserved Diaries, I found that in each month about 15 lunar and 5 planetary positions, both in relation to Normal Stars, are reported. Also, every month the so-called lunar Six are recorded. Each year will in addition contain 3 Sirius phases, 2 solstices and 2 equinoxes, at least 4 eclipse possibilities or eclipses, and about 25 planetary phases. Together, this results in about 350 astronomical observations per year. In 600 years, 210,000 observations are accumulated. Now I do not know whether the archive was ever complete to this extent. Sometimes copies of older Diaries indicate that things were missing in the original. But on the whole, this is the order of magnitude. By counting the number of reasonably (i.e., not completely, but more than half) preserved months, I arrived at ca. 400 months preserved in dated Diaries (undated fragments do not help for the purposes of this lecture). If we compare this to a duration of 600 years for the archive, we see that we have preserved ca. 5% of the months in Diaries." (H. Hunger, "Non-Mathematical Astronomical Texts and Their Relationships," in N. M. Swerdlow (ed.), Ancient Astronomy and Celestial Divination, London: The MIT Press, 1999, p. 82; emphasis added) 


If only five percent of the original Babylonian astronomical archive is preserved today, the scale of the chronological revisions Furuli thinks Seleucid copyists engaged in becomes apparent. To bring their whole archive into harmony with their supposed theoretical chronology, they would have had to redate thousands of tablets and tens of thousands of observations. Is it likely that they believed so strongly in a supposed theoretical chronology that they bothered to redate four centuries' worth of archives containing thousands of tablets? The idea is absurd.


We can also ask why the Seleucid scholars would work out a theoretical chronology for earlier centuries when a reliable chronology for the whole period back to the middle of the 8th century could easily be extracted from the extensive astronomical archive at their disposal. Is it not much more realistic to conclude that their chronology was exactly the one found in the inherited archive of tablets, an archive that had been studied and expanded by successive generations of scholars up to and including their own?


It should be noted that, to make any claims at all about dates in his Oslo chronology, Furuli must rely on the dating of the tablets that the Seleucids supposedly revised. But if one assumes that his chronology is valid, then so must be the dates recorded on the tablets—which destroys his claim that the Seleucids revised the tablets. Thus, Furuli's argument is internally inconsistent and cannot be correct.


Another problem is what became of the original pre-Seleucid tablets. A necessary consequence of Furuli's theory is that almost all extant tablets should reflect only the erroneous theoretical chronology of the Seleucid scholars, not what Furuli regards as the original and true chronology—the Oslo Chronology. In his view, therefore, all or almost all extant tablets can only be the late revised copies of the Seleucid scholars. Thus, on page 64, he claims: "As in the case of the astronomical diaries on clay tablets, we do not have the autographs of the Biblical books, but only copies." This is certainly true of the Biblical books, but is it true of the astronomical diaries? Is there any evidence to show that all the astronomical tablets preserved today are only copies from the Seleucid era?


II-C-3.  Are all extant tablets late copies from the Seleucid era?


It is certainly true that some of the earliest diaries, including VAT 4956, are later copies. They frequently reflect the struggle of the copyist to understand the ancient documents they were copying, some of which were broken or otherwise damaged. Twice in the text of VAT 4956, for example, the copyist added the comment "broken off," indicating he was unable to decipher some word in the original. Often the documents used archaic terminology that the copyists tried to modernize. What about diaries from later times?


As an example, there are about 25 diaries from the reign of Artaxerxes II (404-358 BCE), 11 of which not only preserve the dates (year, month, day) but also the name of the king. (Sachs/Hunger, ADT, Vol. I, pp. 66-141) Some of them are extensive and contain numerous observations (e.g., nos. –372 and –366). None of these tablets show any of the above-mentioned signs of being later copies. Is it likely, then, that they, or at least some of them, are originals?


This question was sent to Professor Hunger a few years ago. He answered:


"In my opinion, the diaries from the time of Artaxerxes II can all be from his reign. You know that the larger diaries are all copies in the sense that they are collections of smaller tablets which covered shorter periods. But that does not mean that they were copied much later. To me it would make most sense if after every half a year the notes were copied into one nice exemplar. I had a quick look through the edition and did not find any remarks like 'broken' which are an indication that the scribe copied an older original. So I would answer your question 'is it likely' by 'Yes'." (Hunger to Jonsson, January 26, 2001)


These tablets, therefore, do not reflect any "theoretical chronology" supposedly invented by the later Seleucid scholars. The tablets might very well be original documents. We cannot take it for granted that they are late copies from the Seleucid era. And the same holds true, not only for the diaries from the reign of Artaxerxes II but for most of the observational tablets dating from before the Seleucid era.


Even if some of the diaries and other tablets dated to the earliest centuries are later copies, it is not known how late these copies are, or whether they were copied in the Seleucid period or earlier. One interesting example is the lunar eclipse tablet LBAT 1420 (No. 6 in Hunger's ADT, Vol. V). This tablet contains annual records of lunar eclipses dated to the first 29 years of Nebuchadnezzar. (See GTR4, Ch. 4, C-3) Steele says of it that "this text was probably compiled not long after its final entry in –575 [= 576 BCE]." (Archive for History of Exact Sciences, Vol. 54, 2000, p. 432) But even if the compilation was made in the mid-6th century BCE, the question still is whether the tablet is a copy or not. If it is a copy, how late is it? Steele explains:


"In answer to your question, there is nothing conclusive in the text that points to a date of composition as the mid-sixth century. However, some of the terminology points to an early date, for example, the inclusion of US '(time-)degrees' after the timings is rare in late texts (the unit is usually just implied by the context), and the facts that the predicted eclipses have no times and the general lack of many details of the observed eclipses are also suggestive of an early date. There is no evidence for the modernizing of terminology, but because the observations are quite brief there are not many occasions where modernizing could have taken place (it is easier to spot in things like star names and the ways in which the moon and planets are said to be near certain stars, neither of which appear in this text). For these and other reasons, the text feels to me like it is contemporary with the material it contains.


Now that all refers to the date on which the text was composed, not the date of the tablet. We have no idea whether this is an original text or one copied in the Seleucid period. (The appearance of a 'variant' time in Obv. I, 4', which I failed to mention in my book, does not necessarily imply the text has been copied–it could just be that the scribe who compiled the text had reports of this eclipse from 2 different observers.) If it is a copy, then I think it is a straight copy, with no attempt to change or modify the text.


Because almost none of the diaries and other observational texts have colophons, we can never be sure whether texts are copies or originals."


In conclusion, the theory that Seleucid scholars worked out an erroneous hypothetical chronology for earlier times that they systematically embodied into the astronomical tablets they were copying cannot be supported by the available facts. It is not based on historical reality and is a desperate attempt to save cherished but false dates.