Do we have ablaut in the Bengali verb system?Is it why we have vowel alternation?
If we take ablaut to mean the Indo-European ablaut, we do not see it in the Bengali verb system. Instead, they can be seen in derivational patterns in loanwords from Sanskrit, e.g. বিজ্ঞান
/biggæn/ "science", but বৈজ্ঞানিক
/boiggænik/ "scientific", where i has become the diphthong oi.
But Bengali does have vowel alternations, e.g. the infinitive কেনা
/kena/ "to buy" becomes in the simple present tense for the 1st person singular কিনি /
kini/ "I buy"]. The
/e/ of the infinitive has become
/i/ before the final
/i/ of the verbal suffix.
This is replicated across the conjugation: before
/e/ in the suffix, the vowel of the root is raised if it is a mid vowel:
/e æ o ɔ/ to
/i e u o/ respectively. This is interpreted as a case of vowel harmony or vowel assimilation rather than the Indo-European ablaut.
However, a few verb suffixes are low vowels e.g. the second person familiar (তুমি tumi) in the present continuous is ছো
/tʃʰo/, the first person in the future tense is বো
/bo/, the first person in the simple past tense is লাম
/lam/ and the third person familiar in the simple past tense is লো
These still trigger vowel raising, because of a diachronic change: all these had an
/i/ before it, as can be seen from the সাধু ভাষার shadhu bhasha forms. For কেনা
/kena/ "to buy" :
[সাধু ভাষা shadhu bhasha > চলিত ভাষা cholit bhasha]
(আমি) কিনিলাম (kinilam) > কিনলাম (kinlam)
(সে) কিনিল (kinilo) > কিনলো (kinlo)
Hence the vowel raising in the stem is a trace of the
/i/ that was previously present across all the forms of the simple past. A similar process occurred for the future tense.
There is also a related
/e/ alternation, which occurs only for certain verbs and certain verb forms, and forms the basis of certain conjugation classes. Compare জানা
/dʒana/ "to know" with খাওয়া
/kʰauwa/ "to eat".