Thanks, Hal. Here are some of my thoughts on the root (dhAtu)
of kiT- and how it relates to the 'durative', present tense.
kiL-/kiT-/kiN- are related set (e.g., aakaNum 'it should happen',
pOkaNum 'I must go' etc., has this kiN- dhAtu in it. Like kiTakkaTTum < kiTakkiTTum ...)
Your 7 essays on spoken Tamil published in J. of Tamil Studies
are available here.
The Tamil Aspectual System : A Transformational Grammar, 1972-1977:
Hope people here, Selva, Iraamaki, and linguists read these important
papers and tell us. I just read the first essay.
Chapter I: Tense, Mode and Aspect in Tamil (JTS, Dec. 1972)
pg. 56
"Thus we have aspect markers /viTu/ 'completive',
/pOTu/ 'completive with nuance of malicious intent';
/vay/ 'keep for future use', /ko/ or /kiTu/ 'reflexive, self-benefactive';
(/kiTu/ seems to be a back-formation from the past of /ko/,
which is /kiTT-/ in non-brahmin dialects; in some dialects,
/kiTu/ is the underlying form, and /ko/does not occur at all); [...]
/ko/ would have to be considered a defective verb in colloquial Tamil,
occurring only in compounds and alone only in the neuter future
form /koLLum/ 'it contains' or as a verbalizer for certain lexical verbs
no longer freely occurring, such as /ottu-ko/ 'agree'. Historically
/kiTTiru/ 'duraitive' seems to be derived from /ko/ po /iru/ but
there are some reasons why it is preferable to consider /kiTTiru/
as a single item, with no morpheme boundary internally. There
are also some reasons why this is not a good idea, and these will
be discussed later."
In spoken Tamil words like teriJcukkO, paarttukkO, puriJcukkO,
vaccukkO, etc., the final common part is -O- (ஓ). ஓ can be
called either as முன்னிலை அசைச்சொல் or வியங்கோள் வினைமுற்று
விகுதி (involving some degree of command ஏவல்). I do not think
this -O- vikuti is from the verb koL-/koNTu-, nor it is from kiT- (kiTTiru etc.,)
For example, when the consonant cluster -nt- (in spoken Tamil
-Jc-, (for example, aRintu (LT) > aRiJcu (ST)) word is followed by the
auxiliary verb, koL, there is no sandhi increasing to -kk- ie.,
in LT we only have terintukoL, purintukoL, aRintukoL where as
we have terintukkO, purintukkO, aRintukkO, etc., Note the difference
of -k- when koL joins, and doubling -kk- when the vikuti -O- is used.
This -O- vikuti is neither from the verb, koL- nor from the verb, kiTa-.
These three lexical items ((i) -O- vikuti (ii) kiT- (iii) koL- are distinct
and need to be separately treated. Otherwise, problems in understanding
spoken Tamil (ST) will occur. For example, on pg. 77
" (74) /seruppe pOTTu-kiTTu kOyilukkuLLe pOkakkUTAtu/
[...]
In this sentence, element (4) /kiTTu/, the participial form of /ko/,
gives a notion of 'simutaneity' in addition to its more usual notion,
as in 30.b of 'self-benefactive or 'self-oriented'. This notion of
'simultaneity' of wearing shoes concurrently to going into the temple,
is of course more clearly aspectual, since it has to do with relationships
of actions to each other, duration of time, etc., This notion is not
present in sentences like (30.b) which makes it difficult to define /ko/
exactly, since its meaning varies."
(30.b) is /itu eTuttukkalAm/. Perhaps, you mean /itai eTuttukkalAm/
and it does not have /kO/ at all. The -O- vikuti in -kO occurs only word-finally.
-O-, koL-, kiT- need to be handled as distinct words and not as same ones.
Next, will give M. RaghavaiyangAr's paper of kiTu, kiTAy examples.
My proposal on how Tamil starts to create the present tense,
from kiT-/kiL-/kiN- words and why kiL- became kil- as a means
to disambiguate. Of course, kil- give rise to kinRu, kiRu, kiRRu-.
At the same time, the two-word combo, "aanin2Ru" was tried
because of the simplicity of kil- (< kiL-), "aanin2Ru" lost out and
not used in spoken Tamil (not even in literary Tamil) even tho'
Nannuul suutram lists the three as present tense markers:
aanin2Ru, kiRu, kin2Ru.
Regards,
N. Ganesan