BREV TIL: Eli Fischer-Jørgensen FRA: Charles Ernest Bazell

fj

(ifU

A-osecg’lu Apt o Humeli yaude i'80 usmanbey xstan ul

Dear i = s Pischer--X6rg'9rs9r , I do not expect to hear from you again until next January, when it is to be hoped that you wil renew your excellent resolution. Meanwhile however Y logise for two misunderstandings of mine and correct others of yours, first. -i

apo- ^he latter

The ?dogma' of converging criteria does not in an way exclude the application of the criteria separately. On the contrary, if citeria must be taken together in determining the general outlines of a category, they must be taken apart for the purpose of classifying the units according to the various types of marginality. Your difficulty in untierstanding my remarKs on zero anses partly irom your unwillingness to believe that 1 am mad enough to say what 1 mean when x say'that the'unmarked features® (a contradiction in terms, since the feature is the mark) are just zero, i.e, nothing at all. Partly also perhaps that x have taken over uakobson's system (with all details of which 1 really disagree) for the purposes of argument. -According to this system, n has only one positive feature, namely nasality* uence n simply _is nasality (in a segmental function). The pair haar/ naar represents then a commutation of aspiration with nasality, the phonemes being nothing more than single features. (As for t, which has no positive feat- ures, this is so to speak the allophone which represents zero when this appears in a segmental function, whereas non-segmental zero has no positive manifestation. Of course there is nothing viciously circular about the principle of commutation . as you apply it; nor is there anything glossematic about it either: it is just plain sensible Prague Sch00q. as you say, the content-difference between cat/hat is so obvious that/jone does not investigate whether the expressions are different first: on the contrary one uses it to prove that k and h are not mere facultative ..variants* But then, in quite the same way, the contents of light as 'un-heavy® and as 'un-dark* are obviously different. Hence the syncretism is merely one of expression; not of content. xhe glossematists canr.ot eat their cake and have it. ""hether the units are simultaneous or not would not be of any importance for a glossematic anaylsis, cf. that r) is analysed as ng in Jjanish'*. But this is simply to say that in glossematic analysis the units are not simultaneous. I was hot referring to phonetic simultaneity. How for my misunderstandings. i'ogeby has already pointed out to me that there would be no commutation of cf and t in fhild/shilt (which have the same content). Tn this case the 'overlapping' happens'(by sheer chance) to be between two single phonefiies. in awaked and awoken it is between whole series of phoneihes (overlap of ei-d and ou-n)?). It all seems very trivial. Certainly in you article you did express most clearly that the criteria of iden- tification were partially substantial, and this did not escape me even at the 1st reading* -*-*ut even the frd reading still left me with the impression that the analysis was formal after a certain stage. However i was wrong. 1he offprints without indication of the journal of publication were from the journal of our own department here. I will send you a copy in atonement of my lapse* -^ut it has generally been my experience that linguists p^efeh not to refer to articles published in so obscure a place. Por instance ilulon ells, in his review of Aecherches, was amiaole enough to give a list of my articles * But av&ded mention of those published here* Yet he is quite familiar with them, and has even sent'me some very interesting criticisms* T should indeed be grateful if you ever find occasion to refer to one of our puh- lications. ^ome of my assistants have published work which does not seem to me in any way inferior to the more respectable stuff published in Aurope (let alone the arrant nonsehse which passes muster even m reputuhlepprioddcala^. The aver- age is poor; and this is inevitable so long as every single section of the fac~ ulty publishes a separate journal. -out x find the journal of our English section useful, since 4 have made it a principle never to submit an article except at the invitation of an editor* nere x happen to be the editor myself. 1 ask my- self for an article, and the request is seldom refused!

°incerely