Subject: data , judgment thereon , and the teach of linguisticsin linguist 5-1247 , benjus wald say > i once hear a syntactician suggest that in introductory syntax > class certain theory . . . have to be teach as if they be > true , without attend to know fallacy in the theory . > that way the student get a foundation for learning how to make > syntactic arguments . the fallacy would then be expose in more > advance course . ( would the way of make syntactic argument > change too ? i guess whoever do n't take the more advance syn - > tax class can go to hell , or assume that theory of syntax do . > > to make this semi-anecdote more relevant maybe the syntactician > be insist that introductory class may not challenge the > grammaticality judgment use as datum for an illustrative analysis > ( sound familiar ? ) . otherwise the lesson could not be teach . > that 's a form of discipline that may be hard to unlearn and hard > for the teacher to resist take advantage of in the more advance > course - - any syntax teachers out there care to comment ? ( if you > do n't i ' ll say " i think not ! " ) i suppose i ' m take up benjus 's gage , though i want it understand that i ' m very much in sympathy with what he have to say . i remember a little over 10 year ago i be take a course in lexical phonology , and get into an argument with the prof . because he insist that one of the co - rollary of the theory be that ' people ' be inherently plural , which explain the non-existence of the overt plural ' people ' . i insist that the word ' people ' certainly do exist and that therefore any theo - retical claim that it do n't be ipso facto falsify . ( cf . rev . vius 9 : ' a great multitude which no one could number , from every nation , from all tribe and * people * and language ' . now , presumably ' people ' in this passage means something like ' ethnic group ' , and be not therefore precisely equivalent to ' people ' mean ' many person ' , or ' the set of all human being ' . i would also note that one common mean of ' universe ' be ' the set of all exist thing ' , which would also imply that the word ' universe ' ought not to exist either , but it do . for one thing , there be a variant usage of ' universe ' which make it roughly synonymous with ' galaxy ' or ' group of galaxy ' , of which of course there be many . it be also freely use in talk about ' parallel universe ' , etc . ) > from a paedagogical point of view i think it be perfectly legitimate for a linguistics teacher to say , ' here 's a bunch of string , with attendant grammaticality judgment . what generalization can we draw from these datum ? ' we can save ourselve from the threat of the protest of a student who happen to be fluent in the language under discussion against a gram - maticality judgment by invoke the notion of ' idiolect ' and claim ' these judgment be an accurate reflection of one particular individu - al 's own linguistic behaviour . what can we deduce about the internalize grammar of that individual ? ' this , of course , assume that what we as linguist be concern about be language as a psychological / cognitive phenomenon , and i believe that that be legitimate . but we be also , in my opinion legitimately , concern about language as a social phenomenon . in which case language be not ( simply ) the property of an individual mind but of a community , and ' gram - maticality judgment ' may be different at the communal level from the individual level . i think as teacher we need to be honest with our stu - dent about this , which means allow ourselve the luxury of a certain type or level of judgment on one occasion and a different one on another , but as much as possible each time make it clear to our student what level we ' re work at . address the broader issue that have come up follow joseph stemberger 's post in linguist 5-1163 , i think we also need to make clear the difference between ' grammaticality ' and ' acceptability ' judg - ment . i have an introductory lecture , prepare for a course i have as yet have no opportunity to teach , in which i attempt to explain that a give string ( e . g . , ' 3 be angry ' or ' my toothbrush be pregnant ' ) may be perfectly grammatical but semantically or pragmatically anomalous , and may be reject for that reason , while another string ( e . g . , ' tooth - brush my be orange ' or ' there be a rabbit in garden ' ) may be fully in - terpretable and therefore not semantically anomalous but nevertheless violate certain constraint of grammar . i would be incline to reject any claim that , as georgium green occasionally put it ( facetiously ) in class and elsewhere , ' the rule of grammar must include the rule of arithmetic and all other real-world knowledge ' . thus , if my son be to say , ' the moon be make of green cheese ' i may challenge him on the * content * of his statement , but not on its * con - struction * . dr . steven schaufele 712 west washington urbana , il 61801 217-344 - 8240 fcosw @ prairienet . org * * * * o syntagma linguarum libereminus humanarum ! * * * * * * nihil vestri privarus nisus obicibus potesti ! * * *