5-1435msg1.txt

运用贝叶斯网编的一个识别垃圾邮件的程序

Subject: re : 5 . 1414 native speaker intuition

on monday , 5 dec 94 , jule levin respond to the follow : ) ) tony bex raise some very interest question in the post ) ) about native speaker intuition . however , there be at least ) ) two other possible explanation for ` could of ' spelling which ) ) plausibly may be offer by naive and not very literate ) ) native speaker of british and american english : ) ) 1 . the ` of ' segment may be a pure homophone which be both ) ) a verb and a preposition without any link other than pronunciation ) ) between them . i think most people would hesitate , for example , ) ) to claim that ) ) a ) he may of say that and ) ) b ) the queen of heart ) ) use the string pronounce ` of ' in the same way . ) ) 2 . ` of ' in ` could of ' may be some kind of preposition ) ) rather than a verb , along the line of ` up ' in ` wake up the baby ' ) ) and so on . ) ) ) ) either of the above may be the native speaker 's understand ) ) of ` of ' - it would be interest to know if anyone have ever ) ) query this by ask native speaker of english in some fashion ) ) or other . levin comment : ) i do not believe that ' of ' be or can be consider a verb here , just because ) it replace the orthoepic ' have ' auxiliary . ) it be an abduction or erroneous hypercorrection of the construct form ) ' woulda ' , ' coulda ' , etc . ) i have be collect sample of the latter in " standard " prose , i . e . , ) newspaper , magazine article , for 20 year , intend to write about the ) construct in " future english " , since that may be the telo of this form . ) it be steadily rise in acceptibility , have appear even in a wm buckley ) column ( even if the latter would insist he be be ironic ) . the problem ) be that as these construct become more and more widespread , ordinary ) letter-writer to the newspaper , try to upgrade their prose style from ) a * no longer analyzable * coulda , shoulda , etc . , but know that the latter ) be somehow not strictly kosher , expand or reanalyze them to a false ' of ' ) construction : " he should of stand in bed . . . " ) therefore i submit that ' of ' here be nothing at all structurally , it be like ) the ' b ' in debt . the real linguistic question be the status of the ) construct vis-a - vi their original compound aux structure . instead of ) engage in the typical amateurish ask of native speaker what they think ) about it , devise a long sheet of sample sentence with ' have ' construction ) mix in with ' of ' construction , and see if native speaker can ) non-randomly pick them apart . well , yes . but this rather miss my point . who be a native speaker ? many british speaker rarely / never use the ' coulda ' construction . they could hardly , therefore , be build on the kind of expansion suggest above . but surely , they are native speaker . similarly , levin 's proposal would only include those speaker who be literate . tony bex