Day is night in Dravidian?

[Jean-Luc CHEVILLARD]

At 11:12 17/04/99 EDT, Palaniappa at AOL.COM wrote :

>DEDR 3656 has Ta. nAL day, early dawn, forenoon..etc.
> What DEDR does not list is the word naTunAL,
> a compound of naTu + nAL  meaning midnight instead of midday.

>If we also consider that DEDR 3621 has Ta. naL meaning 'night', we can see
>that at first the word "nAL" (derived from the root *naL- meaning 'to be
>dark') must have meant 'night'. Then, how did it come to mean 'day'? Another
>meaning of "nAL" not listed by DEDR but preserved in Tamil and Malayalam is
>'daily lunar asterism' more explicitly referred to as "nALmIn2", a compound

>Are there other languages where the word for day is based on the word for
>night?

Later, Periannan Chandrasekaran added:
>Actually the word "nAL" itself stands for daily lunar asterism without
>the assistance of teh word "mIn2":
>"nALoTu peyariya vizu marattu" (neTunalvATai 82)

One has to add to that very suggestive (but probably not conclusive)
discussion the following facts:

A1. the word nAL is frequent inside the 33000 lines
 that constitute the Classical Tamil corpus:
 it occurs almost 400 times; in its most frequent use,
 it does not refer to day-time  (as opposed to night-time)
 but refers to a unit of time that is contained inside a longer period
 (like a month, a life-time, etc.); that unit is somehow
 a 24 hours period; [see for instance: oru nAL "one day",
 pal nAL "many days", cil nAL "not-many days",
 eN nAL tiGkaL "a moon that is eight days old" (litt. "eight day moon"),
 num nAL "your days", etc.

A2. It can sometimes refer to the morning time, some examples being
 Narr. 358-9: ciRu veN kAkkai nAL irai peRUum (see also Narr.21-11)
 PuRam 123-1: nAL kaL uNTu nAL makizh makizhin2
    "if a man's drunk from morning on" (transl. A.K. Ramanujan 1985)
     "[Anyone], if he drinks toddy in the morning,// and gets happily drunk
     by the time he holds court,"  (transl. G. Hart 1979)

A3. it is used inside tolkaappiyam to refer to some "special days/dates"
  like the nakSatra paraNi (=bharaNI, T.Lex) according to iLampUraNar
  but this does not mean it was its main use

A4. There is at least one place where it can refer (with a specifier)
  to night time (according to UVS): it is Kuruntokai 332-1
  vanta vATaic cil peyaR kaTai nAL
   "late at night, when cold winds blow, and drizzling rain falls"
    (transl. Ludden & Shanmugam Pillai, 1976)
   [= ... nALin2 kaTaiyAmattil (U.V.S.)]

B1. On the contrary, naL is not a very frequent item inside the CT corpus
 and it is difficult to be sure of its exact meaning
 (it probable has to do with "middle"; should be connected with naTu &
naTuvaN)
 + We meet 7 times with the expression naL iruL, like in:
   kuRun 107-3: naL iruL yAmattu "in the dense dark of midnight"
   perum 155: naL iruL viTiyal puL ezhap pOki
   "birds awake from sleep at dawn when darkness flees" (Chelliah 1962:115)
 + What about PuRam 246-15: naL irum poykai
  ("large pond [of lotuses]", Hart 1975:103)
 + we also meet with "naL en2al", like in
  naL en2 kaGkul (8 occ.) "in the dead of night" (A.K.Ramanujan 1967:92)
  naL en2 yAmattu (11 occ.) "in the night" (ibid. p.77)
B2. there is one item, naLi, that occurs 50 times and
  could be connected with naL since it has some common shades
 of meaning with it like "denseness"
 (but this post is becoming to long
 so that I'll leave it for others to comment and to argue)

So my conclusion is an absence of conclusion.
We cannot be sure that nAL is derived from naL
What Palaniappan says (night --> day) could be true (it is thought-provoking!)
 but the main CT data does not seem to be very conclusive for that.

Greetings

-- Jean-Luc CHEVILLARD (Paris)