ROLF FURULI: PERSIAN CHRONOLOGY AND THE LENGTH OF THE BABYLONIAN EXILE OF THE JEWS
© by Carl Olof Jonsson, Göteborg, Sweden, 2004
Babylonian Chronology and the Biblical "Seventy Years"
Persian Chronology and the Length of the Babylonian Exile of the Jews is the first of two volumes in which Rolf Furuli attempts to revise the traditional chronology for the Neo-Babylonian and Persian periods. Furuli states that the reason for this venture is that this chronology is in conflict with the Bible. He insists that the Bible "unambiguously," "explicitly," and "definitely" shows that Jerusalem and the land of Judah were desolate for 70 years, until the Jewish exiles in Babylon returned to Judah as a result of the decree Cyrus issued in his first regnal year, 538/37 BCE (pp. 17, 89, 91). This implies that the desolation of Jerusalem in Nebuchadnezzar's 18th regnal year took place 70 years earlier, in 607 BCE, contrary to modern historical research, which has fixed the 18th year of Nebuchadnezzar in 587/86 BCE, a date that also agrees with the chronology of the ancient kinglist known as "Ptolemy's Canon." Furuli does not explicitly mention the 607 BCE date in this volume, perhaps because a more detailed discussion of the Neo-Babylonian chronology is reserved for his not-yet-published second volume.
Most chapters in this first volume, therefore, contain a critical examination of the reigns of the Persian kings from Cyrus to Darius II. The principal claim of this discussion is that the first year of Artaxerxes I should be moved 10 years backward, from 464 to 474 BCE. Furuli does not mention that this is an old idea that can be traced back to the noted Jesuit theologian Denis Petau, better known as Dionysius Petavius, who first presented it in a work published in 1627. Petavius' revision had a theological basis, because, if the "seventy weeks [of years]," or 490 years, of Daniel 9:24-27 are to be counted from the 20th regnal year of Artaxerxes (Neh. 2:1ff.) to 36 CE (his date for the end of the period), Artaxerxes' 20th year must be moved from 445 back to 455 BCE. Furuli says nothing about this underlying motive for his proposed revision.
The hidden agenda
Furuli published this book at his own expense. Who is he? On the back cover of the book he presents himself this way:
Rolf Furuli is a lecturer in Semitic languages at the University of Oslo. He is working on a doctoral thesis which suggests a new understanding of the verbal system of Classical Hebrew. He has for many years worked with translation theory, and has published two books on Bible translation; he also has experience as a translator. The present volume is a result of his study of the chronology of the Ancient world for more than two decades.
What Furuli does not mention is that he is a Jehovah’s Witness, and that for a long time he has produced apologetic texts defending Watchtower exegesis against criticism. His two books on Bible translation are nothing more than defenses of the Witnesses’ New World Translation of the Bible. He fails to mention that for decades he has tried to defend Watchtower chronology and that his revised chronology is essentially a defense of the Watchtower Society's traditional chronology. He describes his chronology as "a new chronology," which he calls "the Oslo Chronology," (p. 14) when in fact the 607 BCE date for the destruction of Jerusalem is the chronological foundation for the claims and apocalyptic messages of the Watchtower organization, and the 455 BCE date for the 20th year of Artaxerxes I is its traditional starting point for its calculation of the "seventy weeks" of Daniel 9:24-27.
Despite these facts, Furuli nowhere mentions the Watchtower Society or its chronology. Nor does he mention my detailed refutation of this chronology in various editions of my book The Gentile Times Reconsidered (GTR; 3rd edition, Atlanta: Commentary Press, 1998; 1st ed. published in 1983), despite the fact that in circulated "organized collections of notes" he has tried to refute the conclusions presented in its earlier editions. (A fourth revised and updated edition of GTR has been prepared and will be published in 2004.) Furuli’s silence on GTR is noteworthy because he discusses R. E. Winkle’s 1987 study which presents mostly the same arguments and conclusions as are found in the first edition of GTR (1983). As a Jehovah’s Witness, Furuli is forbidden to interact with former members of his organization. If this is the reason for his feigned ignorance of my study, he is acting as a loyal Witness—not as a scholar.
Clearly, Furuli has an agenda, and he is hiding it.
Chapter 1: Pages 17-37:
In Chapter 1, Furuli (1) claims that the Bible and the astronomical tablets VAT 4956 and Strm Kambys 400 "contradict each other" (pp. 17-28), and he therefore (2) questions the reliability of astronomical tablets by describing nine "potential sources of error." (pp. 28-37)
Chapter 2: Pages 38-46:
In Chapter 2, Furuli claims that the "most acute problem for making an absolute chronology based on astronomical tablets" is that many, "perhaps most positions of the heavenly bodies on such tablets, are calculated rather than observed." (p. 15)
Chapter 3: Pages 47-65:
In Chapter 3, Furuli (1) makes some general comments on the Sumerian, Akkadian, and Hebrew languages and (2) describes some "pitfalls" in reading and translating the ancient documents.
Chapter 4: Pages 66-92:
In Chapter 4, Furuli (1) presents his views on "the chronological accounts of Claudius Ptolemy" and of those of some other ancient authors (pp. 66-74), then (2) discusses the 70-year prophecy of Jeremiah. (pp. 75-92)
In the material that follows (Part One of this review; Parts Two and Three will be published at a later date), I critically examine the argumentation of these four chapters.
Acknowledgements are made to a number of scholars and knowledgeable colleagues for their assistance in preparing this review. I choose not to mention any names, as some of them, for various reasons, need to remain anonymous. I am indebted to all of them for their observations, suggestions, criticism, and, in particular, for the professional help given by two of them with proof-reading and polishing my English and grammar.
For some works often referred to in the discussion below the following abbreviations are used:
ADT Abraham J. Sachs and Hermann Hunger, Astronomical Diaries and Related Texts
from Babylonia (Wien: Verlag der Österreichischen Akademie der Wissenschaften.
Vol. I – 1988, II – 1989, III – 1996, V – 2001).
CBT Erle Leichty et al, Catalogue of the Babylonian Tablets in the British Museum, Vols. 6,
7, and 8 (1986, 1987, and 1988). These volumes list the tablets from Sippar held at BM.
GTR4 Carl Olof Jonsson, The Gentile Times Reconsidered, 4th ed. (
LBAT Abraham J. Sachs (ed.), Late Babylonian Astronomical and Related Texts. Copied by
T. G. Pinches and J. N. Strassmaier (Providence, Rhode Island: Brown University Press,
PD Richard A. Parker and Waldo H. Dubberstein, Babylonian Chronology 626 B.C.—A.D.
75 (Providence, Rhode Island: Brown University Press, 1956).
One of Furuli's main goals appears to be to convince his readers that there are only three principal sources on which the chronology of the Neo-Babylonian and Persian periods can be based. These three, he claims, "contradict each other":
"There are three principal sources with information regarding the chronology of the New Babylonian and Persian kings, namely, Strm Kambys 400, VAT 4956 and the Bible. The information in these three sources cannot be harmonized." (p. 21; cf. also pp. 15, 45)
"It will be shown in the course of the book that there exist just two such independent sources which can give absolute dates for the New Babylonian chronology, namely, VAT 4956 and Strm Kambys 400 which already have been mentioned. … the chronology that is based on these two diaries cannot be harmonized with the Bible, and this means that at least one of the three sources must give erroneous information." (p. 24)
These statements reveal a remarkable ignorance of a subject that Furuli claims to have studied "for more than two decades." The absolute chronology of the Neo-Babylonian and Persian eras is fixed by about 50 astronomical observational tablets (diaries, eclipse texts, and planetary texts). Almost all these tablets have been published in ADT volumes I and V. And the least reliable of them is probably Strm Kambys 400. (GTR4, ch. 2, last section). For example, there are about 25 diaries from the reign of Artaxerxes II (404-358 BCE), 11 of which have the royal name and regnal dates preserved. Most, if not all, of these appear to be, not later copies, but original compilations from Artaxerxes' reign. (Letter H. Hunger to C. O. Jonsson, dated January 26, 2001) Therefore, to fix the absolute chronology of the reign of Artaxerxes II or any other Persian king, Strm Kambys 400 is needless and irrelevant. Nor is it needed to fix the reign of Cambyses, which can be more securely fixed by other texts.
Additional comments about Strm Kambys 400 and the claim that some astronomical tablets contradict the Bible are discussed in Part Two of this review.
Furuli argues against the validity of the so-called Canon of Ptolemy and traditional chronology by using certain oddly dated cuneiform texts that seemingly conflict with them. However, he admits that a few errors in the ancient texts cannot be used to overthrow a chronology that is substantiated by many other texts:
"One or two contradictory finds do not necessarily destroy a chronology that has been substantiated by hundreds of independent finds." (p. 22)
On the same page he gives three examples:
(1) A tablet that, in 1878, T. G. Pinches said "would overthrow the perfect agreement of Mr. Boscawen's list with the Canon of Ptolemy," adding that "I did not intend to publish it at all." But Furuli fails to mention that this is a tablet that at first seemed to be dated to "year 11" of Cambyses—which contradicts not only the Canon of Ptolemy but also Furuli's Oslo Chronology. That is why Furuli, too, finds it necessary to reject it.
As it happened, the odd date soon found an explanation. On the tablet, the figure for 1 had been written over the figure for 10. It was pointed out by A. Wiedemann (Geschichte Aegyptens, Leipzig, 1880, pp. 225f.) that this seemed to be a scribal correction of a mistaken "year 10," which the scribe had tried to change to "year 1," thus creating a date sign that easily could be misread as "year 11." This simple and natural explanation was subsequently accepted by all scholars. (See my Supplement to The Gentile Times Reconsidered, Danville: Odeon Books, 1989, page 8.) The date, then, was not odd after all.
tablet that "did not fit" PD's "chronological scheme" and
was rejected because "the month sign is shaded, and in view of known facts
this date cannot be acccepted." But Furuli does not inform the reader that
this tablet is Nabon. No. 1054 (BM 74972), which is dated in PD to Nbn
VIII/10/17 (month VIII, day 10, year 17)—nearly one month after the fall of
In 1990, I
asked Christopher Walker at the
"On the Nabonidus text no. 1054 mentioned by Parker and Dubberstein p. 13 and Kugler, SSB II 388, I have collated that tablet (BM 74972) and am satisfied that the year is 16, not 17. It has also been checked by Dr. G. Van Driel and Mr. Bongenaar, and they both agree with me." (Letter Walker to Jonsson, Nov. 13, 1990)
Thus, Furuli's first two tablets cannot be used as examples of "contradictory finds" that conflict with the established chronology. This cannot be said of his third tablet, however, which clearly contains a scribal error.
65494 dates itself to "Artaxerxes VI.4.50" (month VI, day 4, year
50), a date that all scholars, for strong reasons, have concluded is an error
On page 27, Furuli mentions another example of an oddly dated tablet—a double-dated text from the accession year of Artaxerxes' successor, Darius II. The tablet dates itself to "year 51, month XII, day 20, accession year of Darius, king of lands." Furuli refers to this and the earlier text dated to Artaxerxes' year 50 as examples of how scholars "have been reluctant to publish tablets that seemed to contradict the traditional chronology."
But the very opposite is true. The above-mentioned reluctance of T. G. Pinches to publish the text dated to Cambyses' 11th year was an exception. The typical scholarly reaction to dates that conflict with the traditional chronology is interest and attention, not suppression and reluctance to publish. When then-unpublished lunar eclipse tablets dated to the reign of Nebuchadnezzar II were brought up in an interview in 1968, Professor Abraham J. Sachs indicated how scholars would react to such oddly dated texts (they are now published in ADT V). Pointing out that these eclipse tablets all confirm the traditional chronology, he said:
"I mean if they didn't fit it would be worth publishing immediately. I mean dropping everything and saying this whole thing is a mess and there's something wrong here. But they do fit." (Transcript, p. 12, of an interview held with Professor A. J. Sachs at the Brown University, Providence, R. I., on June 24, 1968, by R. V. Franz and C. Ploeger, at that time members of the Watchtower headquarters' Writing Department in Brooklyn, New York; emphasis added.)
The tablet dated to year 50 of Artaxerxes I is listed by E. Leichty and A. K. Grayson in CBT VII, p. 153, and the tablet dated to his year 51 was published back in 1908 by A. T. Clay, in both cases evidently without any reluctance. As noted above, the latter text is doubled-dated. There are, in fact, 10 such texts with double dates, nine of which show that the accession year of Darius II corresponded to Artaxerxes' year 41. That year 51 on the above-mentioned text is an error for year 41, therefore, cannot be seriously questioned. The nine tablets are listed in this Internet article: http://user.tninet.se/~oof408u/fkf/english/artaxerxes.htm.
On pages 27 and 28, Furuli argues that, because there were three (actually four!) Persian kings named Artaxerxes, it is often difficult to know whether a tablet refers to king number I, II, or III. He claims that scholars, in trying to get the dates to tally with the traditional chronology, tend to give themselves up to circular reasoning.
This situation, though, is not as bad as Furuli paints it. This is shown in Part Three of this review, in which I discuss in detail the reign of Artaxerxes I.
Furuli is well aware that the most damaging evidence against his Oslo Chronology is provided by the astronomical cuneiform tablets. For this reason, it is important that he tries to weaken the reliability of these texts. Thus, on pages 29-37, he describes nine "potential sources of error" that might undermine the trustworthiness of the astronomical tablets. Unfortunately, Furuli fails to draw a clear conclusion about these sources of error. Although it is true that errors exist with respect to various aspects of ancient tablets, Furuli fails to explain how these errors affect the accuracy of traditional Neo-Babylonian and Persian chronology as a whole. He simply leaves the reader vaguely to conclude that, in some unspecified way, the possibility of errors invalidates the whole of the chronology. This is akin to someone stating, "Scientists make errors," then implying but not actually stating that "all science is invalid because there are sources of error." Thus, even though a particular astronomical tablet might contain enough errors to be useless for chronological purposes, it does not follow that all astronomical tablets are useless.
But this is how Furuli generally argues. He uses errors in some tablets to cast aspersions on the reliability of tablets he does not like, such as VAT 4956. Inconsistently, he uses the tablet Strm Kambys 400 as a basis for his Oslo Chronology—obviously because the Watchtower Society uses it.
A good example of Furuli's false implication is his using the demonstrated errors in the ancient astronomical tablet known as the "Venus Tablet of Ammisaduqa" to imply that the tablet VAT 4956 is riddled with errors. Parts of the discussion on pages 29-37 of his book are based on an article by John D. Weir, "The Venus Tablets: A Fresh Approach," in Journal for the History of Astronomy, Vol. 13:1, 1982, pp. 23-49. What are these Venus Tablets?
Weir's article discusses the well-known and much-discussed Venus Tablet of Ammisaduqa. This tablet belongs to a particular series of some 70 tablets about celestial omens called Enuma Anu Enlil (EAE). The Venus Tablet is no. 63 in this series. It contains records of observations of the first and last visibilities of Venus made in the reign of Ammisaduqa, the penultimate king of the first dynasty of Babylon. This king probably reigned at least 1000 years before the Neo-Babylonian era. The fragmentary copies of the Venus tablet, found in Ashurbanipal's library in Nineveh (Kouyunjik), are very late. The earliest pieces date from the reign of Sargon II (721-705 BCE). ( H. Hunger & D. Pingree, Astral Sciences in Mesopotamia, Leiden, etc.: Brill, 1999, p. 32)
During the past hundred years, many attempts have been made to date the first dynasty of Babylon with the aid of the Venus Tablet, but no consensus has been formed. The reign of Ammisaduqa has been variously placed all the way from the late 3rd millennium down to the 7th century BCE. In 1929 and 1941, Professor Otto Neugebauer "demonstrated the impossibility of using the Venus Tablet to date the First Dynasty of Babylon." (Hunger & Pingree, op. cit., pp. 37, 38) One reason this is impossible is that the extant copies bristle with copying errors. "The data set is the worst I ever have encountered as a statistician," said Professor Peter Huber, explaining that "at least 20% to 40% of the dates must be grossly wrong." (Peter Huber et al, Astronomical Dating of Babylon I and Ur III [= Monographic Journals of the Near East, Occ. Papers 1/4], Malibu, 1982, p. 14)
Weir points to several sources of error connected with the attempts to date the fragmentary pieces of the Venus Tablet. But it would not be fair to presuppose that the same sources of error also apply to VAT 4956 and other important tablets on which the absolute chronology of the Neo-Babylonian and Persian eras is based. These later tablets belong to the archive of about 1300 astronomical observational texts found in the city of Babylon, texts that contain thousands of observations recorded from the period ca. 750 BCE—75 CE.
In the discussion below, the subtitles are taken from Furuli's summary of the nine supposed "sources for potential errors" listed in his Table 1 on page 37.
According to Furuli, one problem for the ancient Babylonian astronomers was the mountain range to the east of Babylon:
"To the east of Babylon there is a mountain range rising to about 12,000 feet above sea level, while the area to the west of the city is a flat desert. … it is obvious that the high mountains to the east of Babylon would prevent some observations." (p. 29)
Furuli then quotes Weir's discussion of the change of the arcus visionis caused by "hills, mountains, trees and so on." But the Zagros Mountains to the east of Babylon create no serious problems. The higher parts of the range begin about 230 kilometers east of Babylon with Kuh-e Varzarin at about 9500 feet above sea level. Mountains "about 12,000 feet above sea level" lie considerably farther away. Due to the distance and the curvature of the earth, they are not visible from Babylon, at least not from the ground, as can be testified by anyone who has been there. Professor Hermann Hunger, for example, says:
"I have been there [in Iraq], three years, of which two months were spent in Babylon. There are no mountains visible from Babylon." (Communication Hunger to Jonsson dated December 4, 2003)
It is possible, of course, that an observer atop the 90-meter-high Etemenanki ziggurat in Babylon (if the observations could have been made from there) could have seen a very thin, irregular line of mountains far to the east, although this, too, is doubtful. This might have affected the arcus visionis to some degree (the smallest angular distance of the sun below the horizon at the first or last visibility of a heavenly body above the horizon), which in turn could have changed the date of the first and last visibility of a heavenly body by a day or two. Parker and Dubberstein were well aware of this uncertainty, stating that "it is possible that a certain number of dates in our tables may be wrong by one day, but as they are purely for historical purposes, this uncertainty is unimportant." (PD, p. 25; emphasis added) PD's tables were based on Schoch's calculated values for the arcus visionis which, by an examination of 100 Venus observations dating from 462 to 74 BCE, Professor Peter Huber found to be "surprisingly accurate." (Weir, op. cit., pp. 25, 29)
Furthermore, this is a problem with astronomical texts that report only phenomena close to the horizon, as does the Venus Tablet. (Weir, pp. 25-47) Observations of lunar and planetary positions related to specific stars and constellations would not be affected. And it is these observations, which are usually higher in the sky and not in the horizon, that are the most useful for chronological purposes. As noted in GTR4, ch. 4, A-1, the astronomical tablet VAT 4956 records about 30 such lunar and planetary positions, dated to various days and months in the 37th year of Nebuchadnezzar, thus fixing that year as 568/67 BCE with absolute certainty.
Another problem Furuli mentions is related to the place of observation. He states that it "is assumed that the observations … were made in Babylon; if they were made in another locality this may influence the interpretation of the observations." (p. 32) He then quotes from Weir's discussion of the observations on the Venus Tablet of Ammisaduqa, which according to his calculations might have been made at "a latitude of 1½ degree north of Babylon." This would be about 170 kilometers north of Babylon.
Again, this problem applies to the Venus Tablet, the fragmentary copies of which were found in the ruins of Nineveh, but it does not apply to the archive of ca.1300 astronomical observational texts found in the city of Babylon. As shown by modern calculations, these observations must have been made in, or in the near vicinity of, Babylon. (Cf. Professor A. Aaboe, "Babylonian Mathematics, Astrology, and Astronomy," The Cambridge Ancient History, Vol. III:2, Cambridge: Cambridge University Press, 1991, pp. 276-292)
On page 32 Furuli mentions another potential source of error:
"One problem is the crudeness of the observations. Because the tablets probably were made for astrological reasons, it was enough to know the zodiacal sign in which the moon or a certain planet was found at a particular point of time. This does not give particularly accurate observations."
By this statement Furuli creates a false impression that the lunar and planetary positions recorded on the Babylonian astronomical tablets are given only in relation to zodiacal signs of 30 degrees each. He supports this by quoting a scholar, Curtis Wilson, who in a review of a book by R. R. Newton made such a claim, stating that, "The position of the planet is specified only within an interval of 30o." (C. Wilson in Journal of the History of Astronomy 15:1, 1984, p. 40)
Wilson further claims that this was the reason why Ptolemy, "when in need of earlier observations of these planets turns not to Babylonian observations but to those of the Alexandrians of the third century B.C., which give the planets' positions in relation to stars." (C. Wilson, "The Sources of Ptolemy's Parameters," Journal for the History of Astronomy, Vol. 15:1, 1984, pp. 40, 41)
But anyone with even a cursory acquaintance with the Babylonian astronomical tablets knows that Wilson's claim—repeated by Furuli—is false. Although it is true that many positions recorded on the tablets are given with reference to constellations along the zodiacal belt, the great majority of the positions, even in the earliest diaries, are given with reference to stars or planets. The division of the zodiacal belt into signs of 30 degrees each took place later, during the Persian era, and it is not until "toward the end of the 3rd century B.C." that "diaries begin to record the dates when a planet moved from one zodiacal sign to another." (H. Hunger in N. M. Swerdlow [ed.], Ancient Astronomy and Celestial Divination, London: The MIT Press, 1999, p. 77. Cf. B. L. Van der Waerden, "History of the Zodiak," Archiv für Orientforschung 16, 1952/1953, pp. 216-230) During the entire 800-year period from ca. 750 BCE to ca. 75 CE, the Babylonian astronomers used a number of stars close to the ecliptic as reference points. As Professor Hermann Hunger explains in a work also used by Furuli:
"In order to give the position of the moon and the planets a number of stars close to the ecliptic are used for reference. These have been called 'Normalsterne' [Normal Stars] by Epping, and the term has remained in use ever since." (H. Hunger in ADT, Vol. I, p. 17; emphasis added)
On pages 17-19, Hunger lists 32 such normal stars known from the tablets. Noel Swerdlow states: "By far the most numerous observations of planets in the Diaries are of their distances 'above' or 'below' and 'in front of' or 'behind' normal stars and each other, measured in cubits and fingers." (N. M. Swerdlow, The Babylonian Theory of the Planets, Princeton, New Jersey, 1998, p. 39)
Such detailed observations are shown by VAT 4956, in which about two-thirds of the lunar and planetary positions recorded are given in relation to normal stars and planets. And, in contrast to positions related to constellations, where the moon or a planet usually is just said to be "in front of," "behind," "above," "below," or "in" a certain constellation, the records of positions related to normal stars also give the distances to these stars in "cubits" (ca. 2–2.5 degrees) and "fingers" (1/24 of the cubit), as Swerdlow points out. Although the measurements are demonstrably not mathematically exact, they are considerably more precise than positions related only to constellations. As Swerdlow suggests, the measurements "may have been made with something as simple as a graduated rod held at arm's length." (Swerdlow, op. cit. p. 40)
By parsing all the astronomical diaries in the first two volumes of Sachs/Hunger's ADT, Professor Gerd Grasshoff "obtained descriptions of 3285 events, of which 2781 are complete without unreadable words or broken plates. Out of those are 1882 topographical events [i.e., positions related to stars and planets], 604 are lunar observations called Lunar Six … and 295 are locations of a celestial object in a constellation." (Gerd Grasshoff, "Normal Stars in Late Astronomical Babylonian Diaries," in Noel M. Swerdlow [ed.], Ancient Astronomy and Celestial Divination, London: The MIT Press, 1999, p. 107) Thus, two-thirds of the positions are related to stars or planets, whereas only about 10 percent are related to constellations.
In further support of his claim about the "crudeness of the observations" recorded on the Babylonian tablets, Furuli gives a lengthy quotation from B. L. van der Waerden. Unfortunately, Furuli has grossly misinterpreted van der Waerden's statements.
Van der Warden is discussing, not the crudeness of the observations, as Furuli claims, but the crudeness of the calculations that the Babylonian astrologers performed for the position of the moon at a point of time when the zodiacal sign in which the moon stood could not be determined by observation, either because of bad weather or because it was in daytime, when the stars are not seen. These calculated positions had to be deduced from observed lunar positions near such a point of time. The observation that van der Waerden quotes from VAT 4956 to show what was required for such calculations is exactly a lunar position related to a normal star, not just to a zodiacal sign:
"At the beginning of the night of the 5th the moon overtook by 1 cubit eastwards the northern star at the foot of the Lion [= Beta Virginis]." (B. L. van der Waerden, Science Awakening II, 1974, p. 185)
Furuli, then, has totally misunderstood van der Waerden's discussion, because (1) he is speaking about the crudeness of (astrological) calculations, not about observations, and (2) the kind of observations needed for such calculations (which he shows by reference to VAT 4956) is detailed because the lunar position is given in relation to a star, with both distance and direction specified. Although van der Waerden's example happens to contain a scribal error (see below under I-B-4), the information given is definitely not crude. It is specific and precise.
A further source of error, according to Furuli, is "the process of writing down the data." His discussion of this focuses on the astronomical tablet VAT 4956, the "diary" dated to the 37th year of the reign of Nebuchadnezzar. Furuli explains:
"The tablet itself is a copy made a long time after the original was made, but even the original was not made at the time the observations were made. The tablet covers a whole year, and 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." (pp. 30, 31)
Furuli describes the procedure correctly, and it is well known to Assyriologists. But Furuli adds in parentheses, "(provided that the data were not later calculated and there never was an 'original tablet'.)" This theory—that Babylonian scholars at a later time calculated the information recorded on the astronomical diary VAT 4956 and dated it to the 37th year of Nebuchadnezzar—is false because many of the phenomena reported on the tablet were impossible to retrocalculate.
Because Furuli repeats and elaborates this theory in Chapter 2, I will refute his claims in connection with my comments on that chapter. It is sufficient to point out that scholars agree that VAT 4956 is a faithful copy of the original, which is proven by modern computations of the positions recorded on the tablet. The copying errors are few and trivial, as pointed out in GTR4, ch. 4, A-1. (See further below under I-B-4.)
I am aware of only one scholar who has tried to overcome the evidence provided by VAT 4956, namely, E. W. Faulstich, founder and director of the Chronology-History Research Institute in Spencer, Iowa, USA. Faulstich believes it is possible to establish an absolute Bible chronology without the aid of extra-Biblical sources, based solely on the cyclical phenomena of the Mosaic law (sabbath days, sabbath and jubilee years) and the cycle of the 24 sections of the levitical priesthood. One consequence of his theory is that the whole Neo-Babylonian period has to be moved backward one year. Because this conflicts with the absolute dating of the period based on the astronomical tablets, Faulstich argues that VAT 4956 contains information from two separate years mixed into one. This idea, however, is based on serious mistakes. I have thoroughly refuted Faulstich’s thesis in the unpublished article, "A critique of E.W. Faulstich’s Neo-Babylonian chronology" (1999), available from me upon request.
This "source of error" is related to the previous one. As Furuli points out, VAT 4956 is a later copy in which the copyist tried to modernize the archaic terminology of the original tablet. This procedure, Furuli states, "may very well cause errors."
Copying errors do exist, but they usually create few problems in tablets that are fairly well preserved and detailed enough to be useful for chronological purposes. As discussed in GTR4, ch. 4, A-1, the dated lunar and planetary positions recorded in VAT 4956 evidently contain a couple of scribal errors. These errors, however, are minor and easily detected by modern computations based on the recorded observations.
Thus, on the obverse (front) side, line 3 has day 9, which P.V. Neugebauer and E. F. Weidner pointed out in 1915 is a scribal error for day 8. Similarly, obverse, line 14 (the line quoted by van der Waerden above), has day 5, which is obviously an error for day 4. The remaining legible records of observed lunar and planetary positions, about 30, are correct, as is demonstrated by modern calculations. In their recent examination of VAT 4956, Professor F. R. Stephenson and Dr. D. M. Willis conclude:
"The observations analyzed here are sufficiently diverse and accurate to enable the accepted date of the tablet—i.e. 568-567 B.C.— to be confidently confirmed." (F. R. Stephenson & D. M. Willis in J. M. Steele & A. Imhausen (eds.), Under One Sky. Astronomy and Mathematics in the Ancient Near East, Munster: Ugarit-Verlag, 2002, pp. 423-428; emphasis added)
The next source of error in Furuli's list is "the unknown length of the month" in the Babylonian calendar:
"In some instances we do know which months of a particular year in the reign of a particular king had 30 and which had 29 days, in most cases we do not know this. … our Babylonian calculation can be up to one day wrong according to the Julian calendar." (p. 33)
As I pointed out earlier under I-B-1, this is unimportant for chronological purposes. Parker and Dubberstein were there quoted as stating that "it is possible that a certain number of dates in our tables may be wrong by one day, but as they are purely for historical purposes, this uncertainly is unimportant." (PD, p. 25)
Often, when there is an uncertainty of one day, the corresponding Julian day for a dated Babylonian position of the moon or an inner planet can be determined exactly by modern computations. This is particularly true of the moon because it moves 13 degrees a day along the ecliptic, which means that its position in the sky changes considerably in one day.
Further, as Professor Peter Huber points out, "the Late Babylonian astronomical texts consistently indicate the month-length by stating whether the moon became visible on 'day 30' or on 'day 1'." This practice of indicating whether the previous month had 30 or 29 days is also consistently used in VAT 4956. (P. J. Huber et al, Astronomical Dating of Babylon I and Ur III. Monographic Journals of the Near East, Occasional Papers 1/4, June 1982, p. 7)
Contradicting Furuli's claim, Gerd Grasshoff, after his careful analysis of the 2781 well-preserved observation reports in the diaries published in ADT, Vols. I and II (see above under I-B-2), concluded:
"After having completed the successful interpretation of the observation reports, the analysis shows that 90% of the beginnings of the months are correctly predicted with an arcus visionis model, the rest differs only by one day." (G. Grasshoff, op. cit., p. 109)
Another source of error, according to Furuli, is the gradual change in the speed of the earth's rotation. On page 33, he again quotes from Weir's article about the Old Babylonian Venus Tablet of Ammisaduqa. Weir, in turn, quotes Huber, who explains that extrapolating the known rotation rates from the Neo-Babylonian period to the present, back to the preceding 1000-year period, is "beyond safe ground."
But Furuli's quotation is irrelevant because Weir and Huber are discussing the 1000-year period that preceded Neo-Babylonian times. Weir and Huber both know that the change in the speed of the earth's rotation has been established back to, and even somewhat beyond, the Neo-Babylonian period. This deviation (called Delta-T) has been known for a long time, although the value has been gradually refined. The best and most up-to-date examination of the deviation, based on hundreds of dated observations of lunar eclipses all the way back to the 8th century BCE, is that of Professor F. Richard Stephenson in Historical Eclipses and Earth's Rotation (Cambridge: Cambridge University Press, 1997). (See also GTR4, appendix for chapter 4, section 2.)
The rate of increase of the length of a day due to slowing down of the earth's rotation, back to the 8th century BCE, has been fixed at an average of 1.7 milliseconds per century (1.7 ms/c; Stephenson, op cit. pp. 513, 514; cf. New Scientist, 30 January 1999, pp. 30-33). For this period, therefore, we are on "safe ground." Furuli can hardly be unaware of this. Today, the gradual change in the rate of the earth's rotation is definitely not a significant source of error when using astronomical tablets from the Neo-Babylonian and Persian eras to calculate the chronology of these periods.
Arguing that the interpolation of intercalary months in the Babylonian luni-solar calendar might be another potential source of error, Furuli (p. 34) quotes Drs. Ben Zion Wacholder and David B. Weisberg, who say:
"As Professor Abraham Sachs pointed out in a communication to us, some of the readings of the intercalary months recorded in Parker and Dubberstein's tables may not be quite reliable, while a handful are admittedly hypothetical. But even assuming the essential correctness of Parker and Dubberstein's tables, Professor Sachs maintains, the supposition of a 19-year cycle prior to 386 B.C.E. may be reading into the evidence something which possibly is not there." (Ben Zion Wacholder, Essay on Jewish Chronology and Chronography, New York, 1976, p. 67)
Nothing in this statement is not also admitted by Parker and Dubberstein, as can be seen in Babylonian Chronology 626 B.C.—A.D. 75 (1956), pp. 1-9. As Wacholder and Weisberg further demonstrate in their work, the development of the 19-year standard scheme of intercalary months was a gradual process begun in the 7th century. The final stage took place in the 5th and early 4th centuries, when the seven intercalary months of the 19-year cycle were fixed in years 3, 6, 8, 11, 14, 17, and 19. This process is also clear in PD.
Furuli concludes: "This means that calculations based on the Julian calendar can be wrong as much as 44 days or even more if the intercalary months were not added regularly." (p. 35) This conclusion is based on the unlikely supposition that sometimes four years could pass before an intercalary month was added. But the weight of evidence, based on the economic and the astronomical texts, shows that this never happened after 564 BCE. (See the updated tables of documented intercalary months presented by Professor John P. Britton in J. M. Steele & A. Imhausen (eds.), Under One Sky, Münster: Ugarit-Verlag, 2002, pp. 34-35.)
On page 35, Furuli again uses Weir's discussion of the Venus Tablet of Ammisaduqa, this time as a basis for his claim that "a 'best fit' scheme is accepted." This is undoubtedly true of scholars who have used the Venus Tablet of Ammisaduqa in their attempts to date the Hammurapi dynasty, but to imply that such a best fit scheme is also used to fix the absolute chronology of the Neo-Babylonian and Persian periods by means of VAT 4956 and other astronomical tablets—as if this were a last resort—is dishonest because it is simply not true.
Furuli notes that different calendars were used in antiquity by different peoples at different times. This, of course, is true. But because the use of the Babylonian luni-solar calendar in the Neo-Babylonian and Persian eras is well known, it is difficult to see how these other calendars can be "sources of potential error" in the examination of the Babylonian astronomical tablets. Furuli's argument is a straw man.
Furuli mentions that the Egyptians "may have used two calendars" and states that this might be a problem in "connection with the Aramaic Elephantine Papyri." (p. 36) These papyri are not astronomical texts. But, interestingly, some of them are double-dated in the sense that dates are given both in the Babylonian calendar and the Egyptian civil calendar. Because these texts are dated to the reigns of Persian kings in the 5th century BCE, they are useful to determine the chronology of the period and are discussed in a later part of this review.
mentions "the human factor" that might cause "the misreading of
a tablet due to lack of capacity." (p. 37) This is clearly a potential
source of error. Many odd dates found in works about the tablets published
during the past 120 years are due to this factor. It is important, therefore,
when such odd dates are encountered in modern works, to have the original
tablet collated afresh. Strangely, Furuli uses many such dates uncritically and without collation. Some examples of this have already been given
above under I-A-2 and others are presented in later parts of this review.