|A CRITICAL ANALYSIS OF THE EVIDENCE FROM
RALPH HAWKINS FOR A LATE-DATE EXODUS-CONQUEST
Rodger C. Young and BRYANT G. WOOD*
*Rodger C. Young, who wrote sec. I, is a retired systems analyst residing at 1115 Basswood Lane, St. Louis, MO 63132; Bryant G. Wood, who wrote secs. II and III, is Director of Research at the Associates for Biblical Research, P.O. Box 144, Akron, PA 17501.
A CRITICAL ANALYSIS OF THE EVIDENCE FROM
RALPH HAWKINS FOR A LATE-DATE EXODUS-CONQUEST
In a recent issue of JETS, Ralph Hawkins sidestepped the insurmountable problems associated with a late-date exodus-conquest1 and offered five arguments which he suggested “may open up the possibility of a renewed consideration of the Late Date Exodus-Conquest as a viable choice for evangelicals.”2 Three of the arguments are textual and two are archaeological. The present paper addresses these five issues.
I. HAWKINS’S TEXTUAL ARGUMENTS: THE 480 YEARS OF 1 KGS 6:1 ARE SYMBOLIC OR ARTIFICIAL
1. First wrong textual argument: the 480 years are inconsistent with the chronology of Judges. The 479 years of elapsed time indicated in 1 Kgs 6:13 are entirely consistent with the chronology of the book of Judges, as Paul Ray, Andrew Steinmann,4 and other authors have shown, whereas a thirteenth-century exodus cannot be reconciled with its timespans and sequences. The various pericopes of Judges can be divided into two classes, the sequenced and those that might be called unprovenanced, to use a term familiar to archaeologists. Sequenced stories are those that are connected to what immediately precedes or follows by a time-sequence phrase (some are connected at only one end). An example is Judg 10:1–2 (NIV), “After the time of Abimelech a man of Issachar, Tola son of Puah, the son of Dodo, rose to save Israel. . . . He led Israel twenty-three years.”
Unprovenanced pericopes are those which are not related by a sequence-expression to either what precedes or to what follows. Examples are the story of Samson (Judg 13–16), the story of Micah and the Danites (Judg 17–18), and the final three chapters of Judges. The only chronological marker in the history of Samson states that he judged Israel for twenty years in the days of the Philistines (Judg 15:20; 16:31). This could have overlapped a part of the judgeships of Jephthah, Ibzan, Elon, or Abdon, who also were active in the days of the wars against the Philistines.5 Judges 13–16, then, provides an example of a pericope which is not in strict chronological order with what precedes and follows, and the proper way to determine the chronology of Judges is to distinguish between these unprovenanced sections and those that are sequenced. Sequences of years can be constructed from the latter, and the interpreter must then seek the most reasonable time to assign to the unprovenanced passages. It is completely improper to just add all the numbers together without this consideration, as Hawkins does in order to discredit the testimony of Judges as a chronological witness.6
For Hawkins, it is essential that the credibility of the numbers in Judges be negated, because the numbers exceed the time that proponents of a late-date exodus can afford to give to the time of the judges. This is true even with a judicious approach to the chronological data instead of Hawkins’s and Hoffmeier’s artificial adding up of everything. The proper approach to Judges, then, is to carefully study which sections are sequenced and which are unprovenanced, taking note of the exact meaning of the various bridge passages and considering whatever extra information is available, such as the 300 years of Judg 11:26. Advocates of a thirteenth-century exodus cannot afford to take this approach, and so they must discredit the data. Or, in the case of Kitchen’s treatment of Judg 11:26, he defames poor Jephthah.7 But with the proper literal approach to the text, the pericopes in Judges are compatible with the 480th-year datum of 1 Kgs 6:1. They cannot be made compatible with an entry into the land in the late thirteenth or early twelfth (per Hawkins; see below, sec. II.2.a) century bc.
2. Second wrong textual argument: the 480 years represent twelve generations. Hawkins repeats the familiar argument that the 480 years of 1 Kgs 6:1 are a symbolic representation of twelve generations of forty years each. He gives as his basis van Daalen’s comments in The Oxford Companion to the Bible.8 As supporting evidence for 40 years = one generation, van Daalen cites Exod 16:35; Num 14:33, 32:13; Ezek 4:6 and 29:11.9 The first three of these citations refer to the forty years that Israel wandered in the desert, while the Ezekiel passages refer to other forty-year periods that are irrelevant to the discussion. For the present purposes, Num 32:13 can be taken as representative of the texts that are sometimes used to support the equation of forty years with a generation (others that can be cited are Deut 2:14, Ps 95:10, and Heb 3:9–10). These texts relate that the Lord was angry with that generation (dôr in Hebrew, genea in Greek) for forty years while they wandered in the wilderness (in Deut 2:14, thirty-eight years, i.e. from the time of leaving Kadesh Barnea).
In the passages cited the word “generation” is not equated with forty years, nor is it equated to the thirty-eight years in Deut 2:14. Instead, the forty or thirty-eight years are given as the time necessary for that dôr to die, excepting those under twenty years of age (Num 14:29). The word dôr in these passages does not refer to a lapse of time, such as the time from the birth of a father to the birth of his son, although that is one of its meanings elsewhere. This could not be the meaning in the case of Israel in the wilderness because every parent who had children twenty years old or older died together with those children; this would have been two generations dying in the wilderness if the meaning were a time lapse between the birth of the parent and the birth of the child.
The lexicons recognize that there is another meaning of dôr, which is “simply ‘contemporaries.’”10 An example is Gen 6:9, where Noah was a righteous man among his dôr. In the NT, the genea that tempted God in the wilderness in Heb 3:10 and the genea that sought a sign in Matt 12:39 indicate the same meaning: a group of contemporaries, not a measure of elapsed time. By failing to recognize the specific meaning of “generation” in these passages and taking it to mean a period of elapsed time, rather than a group of people, van Daalen and others have reached an erroneous conclusion. Moreover, Hawkins fails to deal with the arguments previously presented showing the incorrectness of this conclusion.11
The reduction of the 480 years into twelve generations of forty years fails because of this wrong practice of equating the “generation” with a period of forty years. This does not mean that the number forty in general and a forty-year timespan in particular are not significant in the Scriptures. Nevertheless, there is no indication in the text of 1 Kgs 6 that the reader was supposed to derive a hidden meaning by dividing 480 by forty to get twelve generations. When 1 Kgs 6:1 states that Temple construction began in the 480th year of the exodus-era, the only conclusion that the reader was intended to draw was that 479 years had passed, and unfortunately many commentators and translators miss even this meaning of the verse.12
3. Third wrong textual argument: the 480 years are an artificial construct designed to put the Temple at the center of Israelite history.
a. The attempt of Burney to show that the 480 years are artificial. In a further attempt to discredit the 480 years of 1 Kgs 6:1 as unhistorical, Hawkins appeals to the work of Charles Burney, published in 1903.13 Burney repeated the notion of earlier authors that the author of 1 Kgs 6:1 artificially constructed the 480 years based on a “known” period from a later time in Israelite history: “[T]he author of our verse [1Kgs 6:1] . . . may thus have purposely approximated the length of the little-known period from the Exodus to the building of the Temple to the chronology of some subsequent period for the knowledge of which he possessed available sources.”14
The theory that the 480 years are derived from the regnal data contradicts Hawkins’s first argument that they symbolize twelve generations of forty years each, but he seems to want it both ways. Whichever derivation is chosen, the main point is that the 480 years cannot be trusted because according to Wellhausen, whom Burney and Hawkins are following here, the regnal data for Solomon and his successors were manipulated to produce a fictitious 480 years.15 Hawkins apparently agrees with Wellhausen’s assessment that the regnal data have been falsified, because he writes, “When the books of 1–2 Kings are viewed as a whole, therefore, it seems clear that its author(s) wanted to place the building of the Temple at the center of the biblical history.”16 The implication is that the regnal data of Kings are not genuine history and cannot be used to create a proper chronology of the kingdom period.
To demonstrate this supposed artificiality in the regnal data, Burney added the balance of the years of Solomon’s reign after the initiation of the construction of the Temple (37), to the lengths of reigns of the succeeding kings of Judah (393), to the duration of the exile (50, presumably from the fall of Jerusalem in 587 to the first return in 537), thus obtaining an interval of 480 years.17 In Burney’s summation there are three mistakes: First, the thirty-seven years assigned from Solomon’s fourth year to his fortieth year should be thirty-six. Second, six years are assigned to Athaliah, from 2 Kgs 11:3, which is the accession equivalent for the seven years of non-accession reckoning that is given to her in the next verse, and which should have been used in keeping with the non-accession system being used at that time in Judah.18 Third, fifty years are assigned from the exile to the return under Cyrus, instead of the proper forty-nine years (587 to 538 bc).
Moreover, we know from modern investigations that this whole procedure is fallacious, a fact Hawkins acknowledges.19 Timespans in Judahite history cannot be determined simply by adding the lengths of reigns of kings, due to coregencies and non-accession reckoning.20 The actual duration from Solomon’s fourth year (967 bc)21 to the end of Zedekiah’s reign (587 bc) was 380 years, not 430, and the total to the first return under Cyrus in 538 bc (Ezra 1:1) was 429 years, not the artificial 480 years calculated by Wellhausen and Burney and cited by Hawkins. It is Wellhausen and Burney who are playing games with the numbers, not the authors of 1 and 2 Kings.
The thesis of Wellhausen and Burney that Hawkins follows is premised on an exilic or post-exilic authorship of 1 Kgs 6:1, and depends upon the further presupposition that an author who was recording chronological data in the time of Solomon, or shortly thereafter, had no way to accurately compute long periods of time over the course of Israelite history. This presupposition and the presupposition of exilic or post-exilic authorship of 1 Kgs 6:1 are both false. This will be demonstrated in secs. c and d below.
b. The attempt by Barnes to show that the 480 years are artificial. Another attempt to demonstrate an artificial span of 480 years is given by William Barnes. By examining his approach, we can see how arbitrary these schemes are. Like Wellhausen and Hawkins, Barnes does not seem to be aware that the Hebrew text of 1 Kgs 6:1 means that 479 years, not 480, had passed between the exodus and Solomon’s fourth year. He writes:
As the reader will no doubt recall, the book of Kings ends on a rather quiet note: in the 37th year of the exiled king Jehoiachin, the Judahite monarch was freed from prison by the Babylonian king Evil-merodach (= Amel-marduk); every day, we are then told, he dined at the king’s table. Now, it is undoubtedly not coincidental that according to the Judahite regnal totals as extant in Kings, exactly 480 years separated this event from the original coronation of King David over Judah. (Once again, the actual historical situation need not concern us at this point, although it would seem that only some 449 years actually separated these two events.)22
Barnes shows how he gets this calculation of 480 years in a table on p. 145 of his book. Here, each reign length is reduced by one in order to conform to the non-accession reckoning of rabbinic scholarship. Thus David is given 39 years, Solomon 39 years, Rehoboam 16, down to 10 years for Jehoiakim. He then adds thirty-six years to get to the thirty-seventh year of the captivity of Jehoiachin. The sum is indeed 480 years, even though, as Barnes notes, this does not represent actual elapsed time. In his view, however, it shows that the numbers have been manipulated to give an artificial total. This approach is similar to that of Wellhausen cited earlier, as followed by Burney and Hawkins, even though the methods of Wellhausen and Barnes contradict each other: one uses accession years, the other non-accession years; one starts with the construction of the Temple, the other with David’s accession; one ends with the return from exile, the other with Jehoiachin’s release from prison. But both conclude to their own satisfaction that they have shown that either someone has manipulated the reign lengths so that they do not reflect historical reality, or that the 480-year figure of 1 Kgs 6:1 is contrived and artificial.
c. The integrity of the chronological data of 1 and 2 Kings shows they are authentic, not artificial. A problem with these schemes is that they are just too clever. The late-date deuteronomists that these scholars posit as the authors of Kings would lack any motive to put together a scheme like this, since it took until the nineteenth century ad (for Wellhausen’s scheme) or the twentieth century (for Barnes’s scheme) for someone to figure it out. And it is not because no one was trying to find numerical schemes in the Scriptures. An example of such searching for patterns is found in the Seder ‘Olam (second century ad), where Rabbi Yose calculated 850 years from the exodus to the exile. He did this by starting with the 479 years from the exodus to Solomon’s fourth year, and then adding, in a non-accession sense, all Judean reign lengths from that time to the last year of Zedekiah. The sum comes to 851, but Rabbi Yose adjusted this to the round number of 850, which he interpreted as seventeen times fifty.23 Perhaps he got rid of the extra year by taking six years for Athaliah instead of seven, as Burney did.
The Seder ‘Olam’s 850-year figure is accepted as authoritative at several places in the Talmud, where no explanation is given for how it was derived; indeed the Seder ‘Olam itself does not explain the derivation, nor does Guggenheimer in his recent translation and commentary.24 It shows that Rabbi Yose was looking for patterns to impose on the Scripture, but he failed to see the scheme of either Wellhausen or Barnes that covers the same time period. If schemes like this were inherent in the text, why did the Seder ‘Olam, the most extensive and detailed document from antiquity devoted to OT chronology, fail to recognize them?
But the main, indeed insuperable, obstacle that confronts the idea that the regnal data of Kings and Chronicles are artificial and late is the fact that these data have all been successfully integrated into a chronology that has every indication of reflecting the actual history of the times. This is more than can be said for the chronologies of Wellhausen and Barnes; their chronological schemes (which are different between the two scholars) have not found any wide acceptance among historians, whereas the Thiele/McFall chronology that accepts these data as authentic is the most widely accepted of any chronology of the divided kingdom.25 In particular, Thiele’s date of 931 bc for the beginning of the divided monarchy is accepted by the majority of scholars who are most influential in this field, including Jack Finegan, Kenneth Kitchen, T. C. Mitchell, Gershon Galil, Leslie McFall, and Eugene Merrill.26 This date was derived by accepting the chronological data of Kings and Chronicles as genuine history, not the manipulations of a late-date deuteronomist and his kin to come up with an artificial and obscure numerological puzzle. Furthermore, it is additionally established by two independent methods: the agreement with the Sabbatical and Jubilee data and the evidence of the Tyrian king list.27
Granted, then, that the reign length data of Kings and Chronicles are historically accurate, could it be that some ancient editor was astute enough to add up the numbers and derive a 480-year figure in a fashion something like that of Wellhausen or Barnes, and then project this 480-year figure back into the time between the exodus and the start of Temple construction? In other words, those who are seeking ways to show that the Bible is not to be trusted in historical matters could say that the 480 years were deduced somehow from the regnal data, which can be accepted as historically correct. They would claim that the editor decided to stop counting either after the thirty-seventh year of Jehoiachin’s captivity (Barnes) or after the return under Cyrus (Wellhausen, Burney, and Hawkins). Then this late-date redactor, once he or she had discovered a 480-year sum in the regnal data, imposed it on the time between the exodus and the start of construction of Solomon’s Temple. One would wonder what purpose this might serve, since the pattern had to wait to modern times to be discovered. It would also imply that this editor knew nothing about the proper methods of interpreting the dates, but merely added numbers from various starting and stopping places until a nice sum was found. But let us, for now, consider this option as a possibility: namely, that the 480 years of 1 Kgs 6:1 were extracted somehow from the regnal data.
This idea cannot be right because it cannot be reconciled with what has just been established. Since the regnal data of Kings and Chronicles, covering a period of over four centuries, have been demonstrated by careful scholarship to have every mark of authenticity,28 then how could it be that when we come to 1 Kgs 6:1, the chronological data there are suddenly no longer historical, but contrived and mythical? For those who prefer redaction criticism, if we grant that the surrounding numerical figures, including the “fourth year” of Solomon, are to be taken literally, then could any judicious approach that deals with literary genre say that the 480 years in the same verse are to be taken as unhistorical? This is particularly pertinent if we accept Cassuto’s argument that the very form in which the number is written is meant to convey exactness.29
Some numbers in the Bible clearly are not to be taken in a strictly literal sense (the “seventy times seven” of Matt 18:22, for example). The context and literary convention being followed are usually plain enough in such cases, however, to show that a non-literal interpretation is intended. For 1 Kgs 6:1, similarly, the context and literary convention being followed dictate that the 480 years must be taken as literal in intention. There is no indication that ancient readers would have understood it in any other sense. To treat it as other than literal would open the door to the radical revisionism that no interpreter with a high view of the inspiration of Scripture could accept: the forty years of Israel in the desert would not be literal, nor the forty days of the temptation of Jesus, nor his three days in the tomb, and so on without end, so that we would no longer be able to understand the plain meaning of any factual statement in Scripture.
d. The Jubilee and Sabbatical cycles show that the 480 years are literal years. Redaction criticism, such as would seek to impose a non-literal 480 years in the midst of an otherwise historical account, has been shown by its practitioners to be a very subjective methodology. It can be, and has been, bent to favor propositions that fly in the face of archaeological or historical facts. Fortunately we do not need to use this unreliable method in order to investigate whether the 480 years of 1 Kgs 6:1 are authentic. A proper way of determining their validity is to examine their agreement with the Jubilee and Sabbatical cycles. Once we accept the small adjustment that Solomon died before Tishri 1 of 931 bc, instead of on or after Tishri 1 as Thiele assumed, then we not only have a correction for Thiele’s problems with the reigns of the Judean monarchs that he was never able to resolve,30 but also, by placing the start of Temple construction in 967 bc instead of 966 bc, the Sabbatical and Jubilee years all fall into place with precision and harmony. This precision and harmony cannot be explained as the interpolations of a late-date deuteronomist and his supposed daughters (dtr1, dtr2, etc.) who were interspersing into their account the various allusions to these events in order to fool readers into thinking that the Jubilee and Sabbatical cycles were observed in Israel’s past. Although interpolations by a “deuteronomist” are the standard wisdom of rationalist scholarship, it is clear that any deceiver who was interspersing allusions in this fashion could never have gotten all the dates right.
The principle of the Jubilee years, first presented in JETS in 2003,31 was cited in Wood’s “Rise and Fall” article (pp. 477, 488) and by Steinmann in the same issue of JETS32 as an important argument in favor of the early date for the exodus. It is also important in demonstrating the integrity of all the chronological data of Kings and Chronicles and in establishing the date for the composition of Leviticus. The argument, however, has never been addressed by advocates of a thirteenth-century exodus, even though there have been several expansions of the basic thesis and additional information in its support since the original presentation in JETS. These later articles have provided new evidence to show that Israel’s priests were keeping track of the Jubilee and Sabbatical cycles all the time that Israel was in its land, and that the start of counting must have been in 1406 bc.
Since these various later articles dealing with the Sabbatical and Jubilee cycles may not be readily available to all readers, a summary will be given here of their findings. This will be a brief summary only; for more complete information the articles referenced must be consulted. The reader may also wish to compare the dates that will be given with the dates for the kings of Judah given in Young’s “Tables of Reign Lengths” article.33 The simple thesis that Israel’s priests began counting for the Sabbatical and Jubilee cycles when they entered the land in Nisan of 1406, as they were commanded to do in Lev 25:1–10, explains the following facts:
First, for the Jubilee years: The Hebrew text of Ezek 40:1, by saying that it was both Rosh HaShanah (New Year’s Day) and the tenth of the month, establishes that Ezekiel saw his vision at the beginning of a Jubilee year. Only in a Jubilee year did the year start on the tenth of the month (Lev 25:9). The date was the Day of Atonement, Tishri 10 of 574 bc.34 Since the Jubilee year was identical to the seventh Sabbatical year,35 the first year of this cycle must have been forty-eight years earlier, starting in 622 bc. 1406 bc, the year that Israel entered Canaan that can be derived from the chronological note of 1 Kgs 6:1, was 784 years, or sixteen Jubilee cycles earlier than this date, thus showing that 1406 would have been the first year of a Jubilee (and Sabbatical) cycle. This is in agreement with an entry into Canaan in that year, since Israel was to start counting the cycles when they entered the land of Canaan (Lev 25:1-10).36