Data

[iad58]

My disciple who's now a student of Sciences du Langage in Lyon directed me to the workbook for his morphology course—a homemade collection of xeroxed passages from some unnamed anglophone source, likewise homemade, judging by the quality of the print. Two dozen exercises there were, four of them on Bulgarian noun morphology, and this lad's question was: Did they ever speak like this?!

I read through the exercises. Fourscore examples all told, each consisting of a noun in two or three forms (singular and plural in three of the exercises; singular, plural and count form in the fourth). Seven-and-thirty made me raise my eyebrows.

Thirty-seven out of eighty. Very nearly half of them.

One gross mistranslation. A few trivial typos or transcription errors. Half a dozen obsolete words (having been codified late, Bulgarian ages quickly), which had in fact prompted the question.

But also several non-existent morphological phenomena whose likeness is brought forth by data errors. Poor students.

And this is not a fairy-tale language out of Neverland (Adyghe or such …); it is the national language of a very accessible country that many Europeans visit regularly, and speakers shouldn't be hard to find wherever you go. The question that arises naturally is, If they got this one so awfully wrong, can the rest of the data be correct?

(In fact, as far as I can tell, most of the other exercises are all right.)