Re: [tied] Re: Singulative

From: Patrick Ryan
Message: 37675
Date: 2005-05-06

----- Original Message -----
From: tgpedersen
Sent: Friday, May 06, 2005 4:58 AM
Subject: [tied] Re: Singulative


I thought the consensus was pre-PIE -k > PIE -x? So *gWena-k >
gunax, gWenakó > gunaikó ? That still doesn't explain the -i-. BTW
isn't the situation with the root *poh3(i)- "drink" similar, with an
unexplained -i- around a laryngeal?
Patrick writes:
One of the phenomena that amazed me when I first started these studies was the correlation I thought I had detected between  /G/ and /x/  (representing velar spirants, voiced and unvoiced) in other languages with PIE *gW and *kW.
I fought against accepting this correlation not because it was impossible but because I could not understand the mechanisms involved. I still cannot. But, having looked at many, many pairs from Egyptian, Sumerian (and others), and PIE, I have become convinced that I have no choice but to accept it.
So, if pre-PIE still had velar spirants (in addition to thse sounds that the laryngeals represented), then the direction of development is pre-PIE /G/, /x/ ->  PIE /gW/, /kW/. Either at the pre-PIE/PIE transition point or earlier, the source language had velar spirants which became /gW/ and /kW/ in IE.
How PIE/IE kept these phonologically distinct from the glides that developed when pre-PIE changed *e, *a, and *o to glide (Y/W) + *e/o, I do not know. I speculate that in the velar series, pre-PIE /go/ and /ko/ may not have become /gWe/o/ and /kWe/o/ but sacrificed vowel quality preservation in order to maintain the integrity of /gW/ and /kW/ from /G/ and /x/ but, of course, I do not know. I have only observed that there seems little confusion if any.
As for the other part of your question, -ko (diminutive) and -no (singulative) and others had some kind of independent existence outside of there employment as suffixes, and the modified word frequently occurs with /j/, an adjective formant, that usually appears as *I; so, for example, we have *-no/*-ko etal. but also frequently *-ino/*-iko etal. which really should be segmented *-i-no/*-i-ko.
In the word above, the addition of *-ko  (diminutive, 'child') has required the stem to form an adjective through addition of *-i- before accepting the suffix-element.
Of course, this applies to nouns. With verbs, *j/i indicates imperfect and irrealis.