Pre-PIE Word Formation
From: Patrick Ryan
Message: 55112
Date: 2008-03-13
We have been discussing Pokorny's PIE *bho:i-(dh-), 'be afraid'.
In order to elucidate a few principles of pre-PIE word formation, I
introduced Pokorny's *bheidh-, 'persuade, have faith in', to stimulate a
discussion of pre-PIE vocalism but I have decided to start a new thread to
investigate this in detail.
First, let me say, I believe that any 'laryngeal', i.e. PIE *H, representing
pre-PIE [?, h, ¿, H] lengthen adjacent vowels, which in the pre-PIE stage
were [e], [a], and [o], causing them to retain their specific qualities
while all unlengthened vowels collapsed into *A(blautvokal), which was
realized as *é, *o, or *Ø in PIE; it is possible that there was an accentual
realization of *A between *o and *Ø that might be represented as *6 (<6> for
schwa; not to be confused with *&, the zero-grade expression of _any_
lengthened vowel).
It is possible through internal and external comparison to isolate three
pre-PIE roots:
1) *bhe, 'foot';
2) *bha, 'chin'; and
3) *bho, 'leg'.
In PIE, these are only recognizable as suffixed root extensions:
1) -*bh(A), 'track, formant of animal names';
2) -*bh(A), not identified at present;
3) -*bh(A), 'formant of place names'.
For this discussion, I will only be looking at 1) and 3).
We can see from PAA, where we have a root *ba?/y/w, 'walk, go', representing
Nostratic *bwu?/y/w after vowel leveling (pre-PIE *bho?; PIE *bho(:)H), that
these three formants were utilized to produce the most basic verbal
activities associated with the basic noun (pre-Nostratic *p?wu, 'leg').
Though I cannot point to a PAA cognate, I, for reasons that will become
clear, also reconstruct pre-Nostratic *p?wi, 'foot'. The most basic verbal
activity associated with 'foot' is 'stand'; and I hypothesize a Nostratic
*bwi? with this meaning (pre-PIE *bhe?; PIE *bhe(:)H).
Looking at PIE *bhe(:)H- first, I believe we can see it in Pokorny's 2.
*bhe/e:-, particle of asseveration, 'true'; the semantic connection is
'standing (in front of), constituting a fact by visual observation).
A formant that is evidenced by languages all over the world manifests itself
in PIE as _derivational_ -*y.
Added to PIE *bhe(:)H, it produces *bhe(:)Hy/i, 'true (by sensual
observation)', a stem listed for several PIE-derived languages under 2.
*bhe/e: as *bhe:/ei.
This stem is the source of Pokorny's 1. *bheidh-, 'persuade, have faith in',
which I analyze semantically is 'place as true, cause someone to place as
true' (this is a revision of my previous suggestion that the final element
means 'hold' rather than 'place'). The final element is PIE 2. *dhe(:)H-,
'place'.
The original form of 1. *bheidh- is *bhé(:)Hidh(&)-, and there are possibly
some faint traces of a former *bhé:idh- but most PIE-derived languages have
simplified the root to *bhéidh-.
I assume on the strength of comparative evidence and internal analysis that
PIE once had a root *bho(:)H- (from pre-PIE *bho?; Nostratic *bwu?;
pre-Nostratic *p?wu?), which meant 'go (away) by walking'.
Adding derivational -*y to it, produces *bho(:)Hy/i, 'pertaining to go
away'.
In this form, it shows up in PIE as Pokorny's *bho:i-, 'be fearful', which
can not be seen more specifically as 'be flighty'.
Adding *dhe(:)H-, 'place', produces 'place flightiness in someone,
frighten', and the original form will have been *bho(:)Hidh(&)-. This
survives in PIE as the adjective *bhoidho-s, 'frightful', and the vowel has
been shortened to [o] as is the case with [e:] in *bheidh- from *bhe(:)Hidh-
above.
PIE 2. *bheidh-, 'weave', is a derivation from pre-PIE *bha(:)idh-, but let
us leave that discussion (if anyone is interested) for another time.
By one of the corollaries of Murphy's Law, any well-organized structure can,
without rhyme or reason, be disorganized as the universe strives to return
to original chaos.
For this reason, I cannot guarantee that the actual original form of either
word underlies a cognate in any given PIE-derived language.
Shortening any of the long vowels could produce *A in PIE with its
accompanying lack of vowel quality specificity.
Patrick