The Eighth Tribe, 1979 (6. évfolyam, 1-12. szám)

1979-11-01 / 11. szám

Page 8

THE EIGHTH TRIBE

November, 1975

I have studied the available literature of the critics and have talked to a number of qualified linguists. With the exception of one all the objections were negative. The only positive argument I came across was that all these apparent coincidences are accidental.

First I will comment on the most impressive negative arguments:

(1) The selection of words in the list show a clear sign of lack of adequate scrutiny in the acceptance of the various entries. This argument is probably correct and undoubtedly applies to some words in table II. If for no other reason because they were collected by a variety of people and many of them may have had different criteria in their procedures. A coordinated approach would probably eliminate many words from that list.

(2) No adequate phoenetical systematica is offered to substantiate the true cognate nature of these words. True, however, that does not mean that no such systematics could be established, especially if we realize that the same sophisticated techniques used for relatively young linguistic relations should not be applied in the deduction of a possible connection which may be as much as 10,000 years older. At least not in the initial stages of the study. In other words, it is possible that the phonetic systematics is there and we may recover it, if we make a consciencious effort.

Actually, the correlation between Japanese and the Finn-ugorian languages has been established by modern techniques. The Graduate School of the Indiana University accepted that for the fulfillment of the doctoral dissertation requirement (Kazár, 1974).

(3) Another negative argument is that a number of the words in the list are de facto loan words and as such they should not be included in the original vocabulary of the respective languages. I accept that in general. That is. when sufficient historical documentary arguments are available to support that. However, in most cases proves are not available and the criteria used for the identification of loan words appears to be arbitrary, and frequently it may be as primitive as: Those words which are common in two languages are loaned from the larger and more developed language to the other. Where the definition of larger and more developed is determined by the present status of the two languages, unless there are historical evidences to the contrary. That is the category of loan words may, in some cases, be used in lieu of the category of words of unknown or questionable origin.

(4) In the comparison of several languages which includes some archaic tongues, the protowords of the modern languages should be used. This is a sound argument, the difficulty with it is a practical one, however. In many cases there are no reliable proto words available, while in some others there are several alternates.

As mentioned before, the only positive argument against the recognition of correspondence between the words of table II is that all these apparent similarities are accidental. I believe that this is the weakest of all arguments. The probability that several hundreds of words would appear to be cognate in several or in all of these unrelated languages is infinetismal. The validity of that statement can be demonstrated by a simple probability exercise.

Let us take nine decks of cards and stack them in the same order. These decks may represent ah original language, where the individual cards are the words and their position in the stacking order designates a meaning. Now we shuffle each deck. If the cards are randomly distributed in all nine decks there should be no other than accidental coincidence between them and the original stacking order.

Let us now calculate the probability of:

A. finding one card in one deck in its original position,

B. finding one card of the same face value but of a different suit in that position, and

C. finding a card of a different face value but of the same suit in the original position.

The probability for A is 1 in 52; for B it is 4 in 52; and for C it is 13 in 52. These are relatively high levels of probability. However, that will decrease rapidly if we add some restrictions: like, what are the probabilities of such coincidence to occur simultaneously in 2, 3,... or 9 decks? The probability of A for two decks is 1 in 522 or 2,704, for B it is 16 in 2,704 and for C 169 in 2,704

In the lexical analogy we may consider in category A words of proven cognate nature, in B words with reasonable phoenetic variations (like p — b) and/or with minor variations in meaning (like rain, drop, water table II) and in C words which sound alike (like hajnal and udzal) and/or with major variations in meaning but still within a definite category (Ike bite and tear). Of course, this analogy is not perfect. Words are more delicate than cards, and similarly, there are fewer sounds and more words in a language than fifty-two. In spite of that, this exercise should clearly demonstrate that the probability of accidental coincidence decreases exponentially