BREV TIL: Eli Fischer-Jørgensen FRA: unsure (1956-04-26)

Istanbul 26/4/o6

Dear PIscher-Jorgensen, Ao of course I have not cited ^jelmslev's letter. I consider that permission should always be asked to cite an author's private com arnications, with only one or two exceptions, une exception is of course, when one wishes to attribute prior- ity to him. (Thhs I allowed myself to mention that a pdnt against Hjelmslev had already be^n made by you and Di&derichsen at a meeting, in my paper on the sememe; though even here I don’t think I said I’d had it from a letter of yours.) ■'hit your revelation (which of course I shan't cite either!) that though the example is mistaken it has been used more than once, has a consequence you strangely do not anticipate. This wrong example is a good example of "incompatibility", the relation that Hjelmslev told Biertsema was "improfitable to seek". But Hjelmslev 'found it without seeking; and has be;n citing since long an example of the very thin he assures us is of no importance! It is not true, by the way, that syntagmatic relations must always be involved in the examples of paradigmatic relations. if. the term "paradigmatic" is used in a manner logically parallel to "syntagmatic”. For then it will not be used to mean "the paradigmatic relations of un;ts in the same paradigm”: the notion of paradlg- matics should not appear twice in the definition. There is paradigmatic aH:to fa Ion - determination whenever a may altoffi’be commuted with B, but not vice-versa (e.g. German t and. d. ). There is no*need to mention any syntagmatic relation; and there would not be even in the case of Latin genders given by Hjelmslev, if gender oc- curred only with substantives. But the extreme case is that of the fourth relation that Hjelmslev wished to /dmit, namely paradigmatic incom atibilityC of which the relation of verb and noun would be an example)/itrø//:the paradigmatic relation here by definition excludes the units being membersv// 'the same syntagmatic råations* ■^ven Hjelmslev’s example illustrates this in a less gross way; since if noun-gen- ders are not invariably computable, the# are not members having exactly the same syntagmatic relations to other units. *ence what Hjelmslev does in his example, is to bundle the genders into one paradigmatic class/ on the grounds that they have similar syntagmatic relations to other units, and then classify the/ cases according to their sucess, partial cess, and even (if one loads him with the necessary fourth relation) also zero-sue- | cess/ in fulfilling the conditions within the limits of a certain syntagmatic type. In other words he uses the paradigmatic criterion twice, onee for putting the units together, and once for separating the cases. There Is nothing parallel to this in syntagmatics. I considered Whitfiels’s review very unfair. No doubt he was angry at a rather silly criticism Siertsema made of his translation. It is a very unequal book, and it is easy to nick out things which are just nonsense. The worst of all is the ridiculous preface. But the author Is certainly more ca,. able of Independent thought than the good Whitfield. ~ou will doubtless give a sympathetic though critical re- view. Unfortunately you. were so kind to Cohen that it will be difficult for you now to make clear that S. is better! I have no doubt the quarrel with Lotz had something to do with Martinet's leaving Columbia. But I am surprised to hear that a Frenchman has to believe in Jakobson's binarity In order to find America tol.dte.bleJ* or was it J. who brought M. to the States? Anyway the break with J. means that M. is now able to expres -? his disap- proval of J.'s binarity without beating about the bush. I am personally quite in agre ment with M. here, and so I believe are you. nnd what a dull theory it is! Yours sincerely

sue-