Wednesday, June 07, 2006

Defending a Case for the Reduction of Large Numbers

A reply to P.J. Williams by Jim Snapp, Jr.

1. Emended Numbers Do Not Erode Any Biblical Theme.

God promised Abraham that his descendants would be unable to be counted (Gen. 15:5). However, the Pentateuch writer did not mean that the Hebrew male population would grow so large by the time of Moses that to count the Hebrew males would be impossible, because no matter how you slice it, the censuses constitute a count of the Hebrew males who were capable of serving in the militia, plus a count of the Levites. And the reduced numbers in the arrays which I presented previously are large; they imply a full Hebrew population of over 200,000.

In the 1200’s B.C. (assuming a late date for the exodus -- but the point stands with an early date for the exodus as well), a group of over 200,000 people would be considered a very large group -- over 2,500 times the number of Abrahamites who had originally entered Egypt. Thus the theme that Abraham’s people were supposed to be fruitful and multiply is not compromised by the reduced amounts for the total population.

PJW: "The large number of Israelites is mentioned in Exodus 38:26 to be 603,550 men, and this number is confirmed by the quantity of tax in the context."

This was explained previously: the thousands-unit in 603,550 is a simple miscombination of the number of officers and actual thousands of men eligible to serve in the Hebrew militia.

PJW: "This number is, moreover, given at great length over the course of Numbers 1-2, and confirmed by multiple totalling."

And the individual sums of the tribes have already been shown to be compatible with the smaller total sum.

PJW: (summarizing) Moses’ objection that it would be impossible to feed "all this people" who number 600,000 men on foot does not fit a scenario in which the population is smaller.

It works as well with a total male population of c. 60,000 as it does with a total male population of c. 600,000.

PJW: Moses' disbelief in God's ability to provide meat is answered by God's provision of quail for one day's journey either side of the camp (let's say conservatively that this means quail for 15 miles in each direction) and 3 feet deep (11:31). This provision suits the large size of camp."

Instead of picturing a quail-rain producing a quail-sea three feet deep, I understand Numbers 11:31-32 to be describing the area where the quail flew (a day’s journey on either side of the camp) and the altitude at which they flew (two cubits above the ground), allowing the Hebrews to easily catch them in nets and swat down, and then collect them into piles.

PJW: "Balak takes Balaam to various points from which he can catch a view of some, but not all, of the Israelite camp (22:41). He emphasises their number (23:10)."

This objection stands against the reconstructions which propose that the Hebrew nation at the time was about the size of a large traveling circus. But these texts are not inconsistent with the amounts I proposed. (Besides, in light of the census-numbers, Balaam’s statement in Numbers 23:10 is hyperbole or prophecy.)

PJW: "The large number of Israelites is again confirmed by the counting of spoil in Numbers 31."

Rather, when one encounters, in the extant text, the statement that 12,000 Hebrews collected this much spoil, it tends to confirm the need for some sort of emendation. It seems circular to refer to one group of large numbers to demonstrate the accuracy of another group of large numbers; both these text-sections were reduced in the earlier proposal.

2. Emended/Reduced Numbers Are Not Inconsistent With Numbers 3:40-43 Or Deut. 9:1.

Numbers 3:40-43 seems to refer to all males at least one month old. A total of 22,273 such individuals seems compatible with a total population of c. 220,000-280,000. While one could raise questions about whether or not this sum was meant only to include underage males, or (a la Keil & Delitzch) males born since the exodus, the reduced sums themselves indirectly answer such approaches -- not so much by showing that the amounts in the extant text are impossible (for only a dense copyist would intentionally make a miscorrection which was palpably impossible) but by showing that the possibility of the extant amounts -- a possibility in the mind of the miscorrecting copyist, at least -- is not the only possibility. Historical feasibility takes the question from there.

Also, the point that the superiority of the occupants of Canaan referred to in Deut. 9:1 may not be a numerical superiority appears neutral. If the Canaanites’ superiority has to do with their building-skills or battle-readiness, then this passage doesn’t impact the numbers one way or another.

3. The Emended/Reduced Quantities Augment, Rather Than Compromise, Reliance Upon Scripture.

PJW: "When evangelicals assert that, despite multiple totalling, thematic links and textual agreement across all the versions, the text of the Bible is corrupt in regard to the number of Israelites, they make it rather attractive not to be evangelical."

The statement is poorly founded, because (a) the textual agreement is granted but its weight is not decisive in light of both the historical difficulty of the immense population-size implied by the extant numbers and the discrepancies between such amounts and other features in the text (such as the Hebrew army’s difficulty defeating the Amalekites), (b) the thematic links are a non-factor (see above), and (c) the multiple totals are already accounted for in the conjecture.

PJW: "The vast majority of scholars from other groups are quite happy for the text to stand as it is."

But the vast majority of non-evangelical scholars do not have a high level of confidence in the historical accuracy of the original text. That is probably a major reason why they feel no impetus for emendation.

PJW: "The most significant thing in attracting evangelicals to abandon classic evangelical views of scripture is when they become convinced that other views show a greater loyalty to the scriptures."

But who’s abandoning a classic evangelical view of Scripture: the person who proposes that the original text recorded some very impressive but plausible large population-numbers, or the person who proposes that the original text recorded historically implausible (i.e., false) population-numbers? What is really in question is not Scripture itself, but the accuracy of the scribal transmission of Scripture (specifically, the scribal transmission of numerical amounts in the Old Testament), and I’m not so sure that there has ever been a focused approach to this specific problem which could be called a "classic evangelical view."

PJW: "The external (mss) evidence for the text is overwhelming. Those who posit that the text is corrupt are positing that large swathes of the biblical text have been systematically corrupted. However, the quantity of text and quality of its thematic coherence on the subject of large numbers easily surpasses the quantity and quality of text needed to establish a whole number of core evangelical beliefs, both doctrinal and ethical."

Of course the extant external evidence will always overwhelm the non-extant external evidence. However the internal evidence is another story. The greater historical plausibility implied by the conjectures, combined with the relative simplicity of the mechanisms by which the proposed original amounts were enlarged, combine to outweigh the case for the originalness of many extant numbers. Plus, one could say that the external evidence for the text is equally overwhelming in cases which are clearly incorrect -- Ahaziah’s age of 42 in II Chronicles 22:2, versus 22 in II Kings 8:26. Inasmuch as the external evidence is not given the final say in cases like that, it should not be given the final say in these cases either -- especially since the extant text itself has some suggestive inconsistencies, which were referred to earlier (such as the (apparent) number of men in the Transjordan tribes, and the reference to 40,000 in the song of Deborah).

PJW: "If you're going to say that the large number of Israelites is textually corrupt, it would surely be easier to argue that John 3:16 is an interpolation."

No it wouldn’t. This particular emendation implies that special phenomena were involved in the enlargement of the numbers, and we have at least one demonstrable example (in Josephus) of one such phenomenon at work. And a fine interlock can be shown to occur between the reconstructed amounts and other (uncorrupted) amounts found in the extant text. Such confirmatory evidence would be lacking in regard to arbitrary assertions of interpolation in John.

PJW: "Corruptions common to MT and LXX as well as other versions would have to have taken place considerably prior to the time of the NT. Those who argue that the text is corrupt are likely to have to conclude that the truth about the exodus had been lost before the time of the NT."

Granted. But whatever difficulty this poses is of essentially the same kind as other cases where disparate amounts are recorded in parallel passages in the MT and LXX (for example, II Kings 24:8/II Chron. 36:9 and the lists of the returners in Ezra and Nehemiah).

4. As a Branch of Historical Research, Textual Criticism Can Legitimately Answer Historical Questions.

It is unsustainable to (a) admit that the extant text is historically problematic, (b) deny the need for conjectural emendation by stating that the issue is best left to a different field of study, and then (c) affirm one’s confidence in the reliability of the extant text. If the extant text is historically problematic, then the problem needs to be solved by historical investigation, and textual criticism is such an investigation -- a historical investigation limited to part of the picture, namely the history of the text’s transmission. Biblical archaeological research is another such investigation -- an investigation limited to another part of the picture, the discovery and analysis of the physical and cultural backgrounds of entities and events mentioned in the Bible. When archaeological research shows the improbability of a population of over 2,000,000 Hebrews at the time of the exodus, and textual criticism shows the probability of a series of textual corruptions (as well as the internal consistency of an emended text), it seems to me that the logical thing to do is not to hand the problem to archaeologists, because they will hand it right back. The logical thing to do is to gauge the historical probability of each option -- (a) Moses led over 2,000,000 Hebrews out of Egypt or (b) copyists corrupted the numbers -- and use the more likely scenario as a working hypothesis. (There are more options than just these two; I mention only them because they seem to be the only ones already on the table, so to speak.)


  1. -- "Ahaziah’s age of 42 in II Chronicles 22:2, versus 22 in II Kings 8:26."

    That is a horse of a different colour, as this verse IS a textual variant unit. LXX mss have '20'.

    But the problem Jim points out is a real one: in the MT as we now have it, some numbers MUST be wrong. Going to the versions can resolve a few of these discrepancies, but not all. In fact it could bring up more problems than it solves, were we to pick any one versional mss and run with it.

    The question Jim seeks to answer is:
    Given that there are some #'s in the MT that we KNOW are wrong (based on internal evidence), and probably were so as far back as the time the NT was written (in light of I Cor 10:8), what's to prevent us from emending other numbers that we only THINK might be wrong (based on some internal but mostly external evidence)?

    Evangelicals don't have the luxury of ignoring this question, any more than we can just leave any questions of the real history of 1st century Christianity to the agnostics and new agers.

  2. Professor Williams,

    Thank you (and colleagues) for continuing to post on this thread. The subject of numbers in the old testemant has been quite bothersome to me. And the more archeology I read, the more troublesome they become. Indeed, the more troublesome much of the Exodus conquest model becomes. (But that is a seperate issue.) I very much appreciate you grappling with the numbers and presenting a logical possible emendation.

  3. "But who’s abandoning a classic evangelical view of Scripture: the person who proposes that the original text recorded some very impressive but plausible large population-numbers, or the person who proposes that the original text recorded historically implausible (i.e., false) population-numbers?"

    First off, thanks Jim for providing us with a helpful example of a text-critical investigation that accepts the verbal accuracy of the autographs as admissible evidence in favor of a set of readings. Given that the Truthfulness of God's Word is a known quantity, there is never any reason that this axiom should be neglected in either higher or lower criticism of the Bible, especially in a dialogue in which all participants have already committed to this dogma in faith.

    However, leaving aside the details of your solution to the problem of the large numbers in Numbers (and I admit they are a difficulty with which we must reckon), I share a certain natural sensation with PJ that an insistence that the numbers must be reduced evidences a certain effrontery toward the text.

    First off, there is an array of assumptions behind the claim that the large numbers are implausible. These working assumptions of archeological scholarship about estimating ancient populations are based on evidence (both material and textual) that itself must be evaluated just as the text of Numbers has here been subjected to evaluation. And, though I don't speak from personal knowledge of current trends in archeological scholarship, is it not the case that at the bottom of these evaluations is an evolutionary scheme of understanding the origins of human civilization? And if we as evangelicals appraise the data we have that is pertinent to such early stages of history, are we not bound to incorporate it into our own system in ways that will appear bizarre and even implausible to those who take a secular approach? It seems almost inevitable to me that we will picture the world in 1400 B.C. very differently than nonevangelicals will. The numbers involved in the exodus and conquest are incredible; but the Bible includes alot of incredible stuff. We may similarly reduce the age of Methuselah or the height of Goliath as implausible numbers (and with more external support). But how will we silence the talking donkey? And what will we be left with when we're done?

  4. trierr,
    Thanks for yours. If I might recommend two things to read, which do not solve the numbers problem, but may help with a small part of the problem of the who issue of the exodus and conquest, I'd commend the following two items 1 and 2 (neither of which is wholly to my liking).

  5. Second link doesn't work. The book is James Hoffmeier, Israel in Egypt (OUP, 1999).

  6. Good point Eric.
    If I may say so I noticed a reference in Jim's article to the Late Date for the Exodus. There are enough problems in Egyptian Chronology already without using it to overthrow the Biblical date for the Exodus--which is right there in I Kings 6:1 (with 'late' and 'early' dates varying by only 40 years textually).
    The problems which led to the late date can be traced to the Manethan framework being off by up to several centuries during the timeframe covered by the OT narrative. Solving this series of chronological discrepancies (which complicates the dating of every world event from the fall of Jerusalem back to the Flood) has occupied evangelical scholars since Isaac Newton, but has received scant attention and almost no support from those committed to an evolutionary understanding of ancient history.

  7. "... has occupied evangelical scholars since Isaac Newton". I know that the word 'evangelical' is getting ever broader, but I think that given Newton's denial of the Trinity it would be wiser not to class him as 'evangelical'.