5-1405msg2.txt

运用贝叶斯网编的一个识别垃圾邮件的程序

Subject: accuracy of historical reconstruction

in write in linguist 5 . 1393 on another topic , martinha @ fub46 . zedat . fu-berlin . de ( martin haspelmath ) say : ) it be true that latin syntax could hardly be reconstruct from ) modern romance language , but neither could latin morphology , and ) even the view of latin phonology that we would get from romance be ) very distort . our reconstruction of protolanguage grammar be ) alway imperfect . . . in look at historical reconstruction do for s . american language ( largely phonological ) , i ' ve often wonder about this . just how much could we trust the reconstruction that we do ? one of my rule of thumb for those language be that if an item be longer than one syllable , it be suspect as be polymorphemic , and if it be longer than two syllable it be almost certainly polymorphemic . the problem be that most attempt at reconstruction ignore this areal phenomenon ( sometime because the datum be simply unavailable ) . from what i know of romance language , i would say polysyllabic morpheme be more common there . if anything , that should make it easier to reconstruct latin , since you have more to work with . ( of course , the morphology of romance language be much better know than that of the language of s . america , which also help ! ) so if latin would be very imperfectly reconstruct , what hope be there for native american language ? has anyone ever attempt , as an exercise in the comparative method , reconstruction of latin from the romance language , then compare the result with the real thing ? or reconstruction of any other attest language from its descendent ?