BREV TIL: Charles Ernest Bazell FRA: Eli Fischer-Jørgensen (1953-01-24)

24-i-i>3

Dear ar. Bazeii, As it is still January, my resolutions have not quite lost their effect, and I hurry to answer yourt letter from (excuse me, there is of coubøø no date;. - Why should people not quote articles published in a remote place? Perhaps Wells has not known either where the offprints were taken from. - And if your assistants do not send fcany offprints abroad, it will be difficult to have them quoted. I do not think your series is found here in Copenhagen, though I have not tried the University Libra- ry. - By the way, of you have more offprints of "gros ematic definitions'* and your review of Recherches I would be interested. I have borrowed them from somebody else.

i

I do not hink it is possible to identify "unmarked" feature with zero because"unmarked" does not make sense if it is tameri literally, e.g. p/b in German, p may be called unmarii because it. is the representative of the phoneme in neutralisations, out there is zø reason for considering the phonetic quality as zero, aithougn in tnis case it may be argued that acoustically there is voice in b, and nothing in p (physiologically tnere are two different articulations of the cocal chords). - In other cases : nasal - oral, "grave" - "acute" etc. there is no reason at ail. And Jakobson does not seem to have any functional reasons ,nor really any phonetic reasons for his choice between + and *. He seems to mean one quality and the opposite; which one is + and which ~ is irrelevant. He does not even use the terms marxed or unmarked in these cases. - But where the opposition is irrelevant, e.u. nasality in r, .saturation in 1 etc.,there it might be possible to talk of zero, (n is in Jakobsons terminology defined as a nasal, acute and diluted consonant, r only as an intexcepted consonant ( tnus witn many zero-features in this iattex sense).

In light as »unheavy' and *unaakk' tnere is no commutation, since tne expression is tne same. Tnere is a difference between spild-spilt ana awaked-awoken. in the first case there is overlapping u der phonemic conditions, ijaxihaxiasfc g&s&x ( d/t finally, if 1 understand the examtle right ) ±a awaked- awoken, is not phonemicaiiy conditioned. As for my article the meaning was^that tiai basis is indispensibie for establishing tne phonemes of a language, "position"-phonemes included, 3) that these phonemes can then be defined by chair interrelations, tnis myy be wailed formal definitions and clas- sifications , 3) that comparisons between classes of phonemes languages might possibly be made on a purely formal basis sexxM

identification on a substan-

in differet

xx, though I do not quice see how it can be done , ana tnat in is why Inave pre- ferred provisionally to identify units arid ciassaes •• -in different Ian. guages on a faxmaixkx substantial basis, choosing position as the one occurring most often and preferring therefore to define the others by their relations to tnis o. e.

— One of the offprints I am probably sending you is a transia- tion into iiingiish. of an oid and not easily accessible articieof mine, &M£h£xxjL&Kkish&xxsnkxRm&x"PhonologieUebeisichtsberlcht user Arbeiten in germanischet? una romanisene® Hpracné*^ i'he trans!i&&tx$fivM^ been made for Dime's Summer Institute and mimeographed ( as you may Know, even professors of linguistics in America do not generally read German or French) . khxxk Garvin Knows German well, but the translation is simply horrible, so that i nave haa two pages of corrections of the most di- sturbing misunderstandings mimeographed. lours sincerely