John R. Perry’s ‘A Tajik Persian Reference Grammar’ (2005) is a case in point.
The book does not clearly distinguish between phonological and phonetic levels of representation. According to the book, the ‘letter name’ of Ӯ is ‘ū’ and its ‘sound’ is ‘/u̇ / [u̇ ]’ (p. 37). Incidentally, in print, the u̇ here is an overdot u (!)
There’s not much clear description in the book either. It bombards you with a large number of unglossed examples and expects you to work out what grammatical points they represent.
‘A Persian Corpus’ would have been an apt title.
A linguist will be frustrated by such terms as ‘quasi-tense’, ‘quasi-article’, ‘quasi-coordinate’, ‘Quasi-future tense’, ‘Participial Quasi-Future Tense, ‘Quasi-passive’, ‘selective relatives’, ‘selective pronouns’, and ‘pseudo-direct speech’ and be baffled by the complete absence of grammatically glossed sentences.
Added to verbal agentives or suitable nouns, it (-ак, ک-) forms concrete instrumentives (p. 419).
Whatever that means.
Quasi-terminology.
So ic-al (!) is a ‘complex adjectival formative’. Very helpful.
An Iranist’s sentiments at times take precedence over plain description in this book. Every example is accompanied by its Arabo-Persian script equivalent though basically no one reads or writes Tajik in Arabo-Persian script.
As a result, it contains lots of overly long paragraphs padded up with unnecessarily repeated examples in different scripts as well as paragraphs that do not concern Tajik at all, e.g.:
In Perso-Arabic, especially after -ī ی-ِِِ, this variant may be written separately: راضی گی ‘satisfaction, consent’, قاضی گِی ‘qozi-ship, status of judge (p. 421).
Getting rid of such impertinently appended paragraphs and Arabo-Persian transliterations from the book would have halved its thickness, which would have been welcome.
