Figure 2. Output from the first run of Dawkins' simulation.4,5 prevent a gene from functioning point, then only the letter positions There are significant advantages to than of successful configurations. mismatched with the target are the approach published in this journal. • The evolutionary assumption, that reshuffled for every 'generation'. The It makes transparent that the example a novel, complex function missing intermediate most closely matching the is indeed a 'rigged fruit-machine' affair, in an original population has goal is flawlessly retained every time which both respondents failed to evolved to produce it in a des- as the starting point for a new recognize. In addition to the obvious cendant population has been generation. flaws already discussed in my article, it allegedly demostrated by Dawkins Now, Dawkins fails to inform us as becomes easy for others to add to this through the following 'proof': a path to how many progeny permutations are list. For example, if Dawkins' analogy over time of discrete and viable generated to choose from. An astro- had any validity, several other outcomes mutations is assumed to exist and a nomically large number would lead to are predicted: mechanism is assumed to exist instant success everytime. A smaller • Lengthening the target sentence which cannot fail to converge to any number would provide less op- results in a vanishingly small and every outcome by permuting the portunities to move in the correct probability of finding any mutation original genome. As pointed out in direction. I used the worst possible which does not lead closer to the the journal article, accepting that scenario in the paper to show goal starting from the first gene- kind of argument and the assump- convergence is unavoidable: only one ration! Consider a random sequence tions would make it possible to surviving mutant is produced per the length of an encyclopedia, to 'prove' anything since the outcome generation, offering a choice between mimic a real genome. A single can never fail to be met. it and an identical copy of the parent. mutant progeny, to fail to make The objections offered only The analysis, based on well-known progress within that generation, strengthen the argument that Dawkins' binomial probability distribution would have to fail to match any of example is irrelevant to the real world. formulas, revealed that any target the millions of letter positions sentence cannot fail to be matched, available. This is virtually Royal Truman irrespective of length and starting letter impossible statistically. Worse still, Ludwigshafen configuration using Dawkin's con- Dawkins' example would allow straints. perhaps another 1000 mutants to GERMANY Clearly, the only effect of extra also try within the same generation mutants per generation is that the period! sequence converges more rapidly to • As a long sentence converges to the the goal! It becomes less likely to target, eventually it would be far reproduce only misaligned letters. Also, more likely that one of the letters Cainan in :36: occasionally several letters would be already correctly lined up would insight from concurrently 'corrected' within that misalign than that the one or two generational attempt. The proof lies in remaining letters should get lucky Josephus the much lower number of iterations within that generation. This reflects that Dawkin's reported simulations the real-world constraint that there Because Luke 3:36 has the extra required than mine, for which 10,000 is a vastly greater number of name Cainan in the genealogy simulations were run. nucleotide arrangements which can compared with Genesis 10 and 11,

CEN Technical Journal 13(2) 1999 75 and 1 Chronicles 1, sceptics have would have included it! If the LXX copies, and in the Vulgate Latin, used this difference to attack biblical contained the reading, Josephus and all the Oriental versions, inerrancy. However, as Sarfati either omitted it by mistake (which even in the Syriac, the oldest of pointed out,1 the difference was not is not likely) or held the reading in them; but ought not to stand an error in the God-breathed (2 low esteem. We know that when neither in the text, nor in any Timothy 3:15-17) original auto­ Jerome (c. AD 347-419/420) trans­ version: for certain it is, there graphs of Scripture, but one of the lated the Vulgate (Latin translation never was such a Cainan, the son extremely few copyist's errors in the of the ) in the 5th century AD, he of Arphaxad, for was his manuscripts available today. To put did not use the LXX in spite of son; and with him the next words this into perspective, we have the Augustine's (AD 354-430) pleadings should be connected.' original text to 99% accuracy in the because Jerome said it was too The evidence from Josephus Old Testament and >98% in the New inaccurate. He used the Hebrew text and Gill shows conclusively that the Testament.2 Most of the variation in which did not include the variation. extra name Cainan is not part of the remaining <2% is merely The great Baptist theologian Dr God's original Word, but due to a stylistic, and not a single doctrine of John Gill provided further strong later copyist's error. Thus it cannot Christianity relies on a debatable support that Cainan was a spurious be used as an argument against text. Sarfati, after Morris,3 thought addition. He summarized the textual biblical inerrancy (nor can it support that the error came about when an evidence as follows in his major ideas of gaps in the Genesis early copyist of Luke inserted the Bible commentary.6 genealogies). extra name, and from there it was 'Ver. 36. Which was the son of incorporated into later Cainan, &c. This Cainan is not Larry Pierce (LXX) manuscripts. mentioned by Moses in Ge 11:12 Winterbourne, Ontario This is strongly supported by nor has he ever appeared in any CANADA information from the Jewish Hebrew copy of the Old Testa­ historian Flavius Josephus (AD 37/ ment, nor in the Samaritan References 38-c. 100) about the genealogies. version, nor in the Targum; nor The following is a table comparing is he mentioned by Josephus, nor 1. Sarfati, J.D., Cainan of Luke 3:36, CEN Tech. the genealogy from the Hebrew text, in 1 Ch 1:24 where the genealogy J. 12(l):39-40, 1998. the LXX, and Josephus4 giving years is repeated; nor is it in Beza's 2. White, J.R., The King James Only after the Flood at birth of the son most ancient Greek copy of Luke: Controversy, Bethany House Publishers, Minneapolis, MN, pp. 38-40, 1995. mentioned. it indeed stands in the present If Josephus did not use the LXX, copies of the Septuagint, but was 3. Morris, H.M., The Genesis Record, Baker Book House, Grand Rapids, MI, pp. 280-283, he must have used some document not originally there; and there­ 1976. fore could not be taken by Luke based on the LXX for it repeats too 4. Josephus, Jewish Antiquities Books I-IV, many of the mistakes of the LXX to from thence, but seems to be Harvard Press, Cambridge, MA, p. 73, 1930; be a chance occurrence. It appears owing to some early negligent Loeb Classical Library No. 242. that at the time of Josephus, the extra transcriber of Luke's Gospel, and 5. Note that was not 's firstborn. generation of Cainan was not in the since put into the Septuagint to Genesis 12:4 says Abraham was 75 when he LXX text or the document that give it authority: I say "early ", left Haran, and this was soon after Terah died at 205 (Genesis 11:32), and the difference Josephus used, otherwise Josephus because it is in many Greek 205-75 means Terah was actually 130 years old when Abraham was born, not 70 (Archbishop Ussher (1581-1656) seems to have been the first modern chronologist to have noticed this point). The latter figure refers to Terah's age when the oldest of the three sons mentioned was born — probably Haran. 6. Note on Luke 3:36, In: John Gill, D.D., An exposition of the Old and ; the whole illustrated with notes, taken from the most ancient Jewish writings (nine volumes), London: printed for Mathews and Leigh, 18 Strand, by W. Clowes, Northumberland- Court, 1809. Edited, revised and updated by Larry Pierce, 1994-1995 for Online Bible CD- ROM.

76 CEN Technical Journal 13(2) 1999