5-1288msg1.txt

运用贝叶斯网编的一个识别垃圾邮件的程序

Subject: comparative method

the recent discussion of limitation on the comparative method contain several assertion that such limitation have be and be be use to justify resistance to proposal of remote relationship , include amerind and nostratic . to my knowledge this be absolutely false . in every instance that i be aware of in which perceive temporal limitation on the comparative method have be mention , it be by way of explain why it be there be no relationship beyond a certain degree of remoteness on which there be consensus , or by way of make prediction as to what historical linguistics will ultimately be able to achieve . i do not know of a single instance in which someone have argue : such and such a propose relationship be associate with a time-depth of x year . this exceed the temporal limit of the comparative method . therefore the proposal must be wrong . if anyone can provide evidence of such an argument be make i would be most interest . in the particular case of amerind , the objection have be two-fold : ( a ) the datum be riddle with error ; ( b ) the argument be not of a sort that most historical linguist consider probative . since greenberg and ruhlen do not apply the comparative method disbelief in their claim could not rationally be base on perceive limitation of the comparative method , nor have it be . ironically , even if greenberg and ruhlen do attempt to apply the comparative method , amerind would not provide a very good instance of the above hypothetical argument . since even conservative estimate of the time-depth possible with the comparative method range from 6-10ky , cand since greenberg and ruhlen adhere to the low chronology for the people of the america that place it at roughly 12ky bp , their date for amerind do not radically exceed the perceive limit of the comparative method . in sum , whatever the validity of propose temporal limit on the comparative method , and i agree that such limit be far from exact , the view that this have anything to do with reaction to greenberg 's work on amerind and similar work be a red-her . to evaluate such proposal , look at the datum and look at the methodology , not at the allege ( and generally unknowable ) motivation of the critic . you 'd think that the irrelevance of ad hominem argument except in matter of credibility of witness would not need to be repeat constantly . bill poser bill poser , first nation study , university of northern british columbium , 3333 university way , prince george , british columbium , v2n 4z9 , canada 604-960 - 6692