Tonal and stress accents

[Jens Elmegaard Rasmussen]

On Tue, 14 Mar 2000, jose.perez3 wrote:

>>> From: "Miguel Carrasquer Vidal" <mcv at wxs.nl>

>> [...9  Did (Pre-)PIE have simultaneous stress accent and tonal
>> accent, or did it switch from one to the other in the course of
>> its development?  [...]

> I'd be very thankful for anybody who has paid attention to the issue of tonal
> and stress accent to elaborate on this subject, presenting some sound
> information about how we think that IE differed from most modern European
> languages in its accent.
[...]

I do not think it differed very much: As your mail continues, there is a
high pitch on the stress in Modern Greek too, and the same goes for
Italian. In PIE the two must have gone hand in hand, i.e. there was a
"prominent" syllable in every word, its prominence consisting in high tone
AND stress (greater muscular effort giving a louder sound volume).
   Why must that be so?
   A. HIGH PITCH is secured for PIE by:
      A.1. The Sanskrit term uda:tta 'elevated' used of the prominent
           syllable in Vedic.
      A.2. Greek grammarians' report of a manifestly higher pitch on the
           prominent syllable.
      A.3. In Balto-Slavic, a short diphthong (and a short vowel) has
           falling tone, whereas as long diphthong (and an old long
           vowel) has rising tone: The prominence is on the
           beginning of the last full-vowel mora of a syllable. That
           can be achieved only by assigning high pitch to the most
           prominent part of the syllable.
      A.4. The Slavic, after many changes, the basic accent habit remain
           the same: There is high pitch on the formerly prominent
           syllable in Stokavian Serbocroatian, even after the stress
           itself has been retracted (to the preceding syllable)
	   from the syllable still accented in Cakavian and Russian.
   B. STRESS on the same segment is secured by:
      B.1. The stress accent of Pashto which basically falls on the
	   same syllable as the Vedic uda:tta.
      B.2. The stress accent of Modern Greek which is still on the
           syllable accented in Ancient Greek.
      B.3. The stress of Russian and Cakavian which falls on the syllable
	   carrying high pitch in Stokavian.
      B.4. The Hittite plene writing of originally short vowels in
           accented position, which points to volume, not pitch. The
           effects on the consonantism which is reduced after long
           vowels are the same in Luvian and Lycian as in Hittite.
      B.5. The Lydian vowel reductions in unaccented position observed
           by Eichner.
      B.6. The loss of unaccented vowels in Albanian numerals where
           reductions appear to have taken place so early that the
           relevant accent was still that of PIE, while in the bulk
           of the vocabulary reductions postdate a more automatic
           accent assignment.

There are certainly other indications, but this will suffice to show that
the two features (high pitch and stress) went together in PIE already.

Jens