BREV TIL: Louis Hjelmslev FRA: Charles Ernest Bazell (1956-02-20)

c

Rumoli Gaddeal 80, Koseoglu Apt. 3, Osmanbey Istanbul 20/2/56

Dear Hjelmslev, In case you replied already to the letter and did so by ordinary post, this was no received on account of the floods in Thrace which have held up all land- mail, for the last two or three weeks. Only air nail is now received, we shall exnect the land-mail ultimately to come through, but if there has been any cor© responrence it would be safer to send a copy by alr-mail(if one exists* person- ally I don't keep copies of my letters but imagine that you do.) Meanwhile another point which your let er raised. Before, I had imagined that permutation was regarded on the same level as com utation, perhaps even as a special case of corrmitation fcom utability of AB and Ba where A and B are not individually inter-commutabled. But from your letter it S'ems that permutation stands to syntagmatlcs as comnntation stands to paradigmatics• I only half understand this. Of course, the permutable units may occur togetht in a . so there is a both-and relation between them, not an either-or relatior as with commutation. But when one permutes A and B in AB, to obtain Ba, one is surely doing something analogous to com uting A and B; with this dlf erence, that in the former case one shifts the relations (of sequence) between the terms, while in the latter case one shifts the terms themselves. Th< distinction would se^m one between relations and relata rather than between syntagmatlcs and paradigmatic But the main point remains as before-- why there is an asym etry between the syntagmatic and the paradigmatic relations. I can understand why you should ex- elude the relation of incompatibility from the syntagmatic relations: once the thre relations have bee given, the fourth (A selects non-B -- incompatibility of A and B) is left as a residue, and nothing further is added by including it. But then, why not exclude the corresponding paradigmatic relation!'not both A and B in the same paradigm]!), rather than a different relation? But perhaps an answer on this point lies buried in the Thracian floods? Yours sincerely

—é~— — A**.