Let's see how the rules work out:
I'll start with a.p. c verbs.
Athematic:
vêmI' (vê~dê [?]), vê'dêti, vê'dêla
Late PIE: *wóid-h2ai, *wid-mé (perfect, mobile),
[analogical:] *woid-eh1-téi, *woid-eh1-láh2 (oxytone).
Hirt: does not apply (but inf. woid-éh1-tei, l-ptc.
woid-éh1-l-ah2).
Winter: *wai:d-mí, *wai:d-é:-tei, *wai:d-é:-l-a: (root
becomes acute)
Meillet: *waid-mí, *wai:d-é:-tei, *wai:d-é:-l-a: (present
root becomes circumflex again)
-Dybo: *waid-mí, *wái:d-e:-tei, *wái:d-e:-l-a:
da(d)mI', da'ti, dala'
Late PIE: *didóh3-mi, *di-dh&3-més (mobile), *doh3-téi,
*d(o)h3-láh2. The Slavic present root goes back to *dod-,
which would seem to be a conflation of the weak root
*did(&3)- with the strong root *(di)doh3-.
Hirt therefore applies only to the inf. (*dóh3-tei) and
perhaps the l-ptc (*dóh3-lah2).
Winter: *do:d-mí, *dó:ti, *dó:la: (and *do:lá: with mobile
accentuation, after the present).
Meillet: *do~dmí, *dó:ti, *do~lá: (mobile)
Dybo does not apply.
jesmI', by'ti, byla'
Late PIE: *h1és-mi, *&1s-més (mobile), *bhuh2-téi,
*bhuh2-láh2.
Hirt does not apply (Francis-Normier: *ésmi, *bhu&2-téi,
*bhu&2-láh2).
Winter does not apply.
[ésmi => esmí, by rearrangement of verbal mobility]
Meillet: *esmí, bu:téi, bu~lá:
-Dybo: *esmí, *bú:tei, *bu~lá:
êmI', ê'sti, ê'dla
Late PIE: *h1éd-mi, *h1(e)d-més, mobile. *h1ed-téi,
*h1ed-láh2.
Hirt does not apply.
Winter: *e:dmí, *e:dztéi, *e:dlá:
[analogical after bad-láh, *bad-lá, *bad-lá: and *nes-láh,
*nes-lá, *nes-lá: the l-ptc. was replaced by *e:dláh,
*e:dlá, *e:dlá:]
Meillet / -Dybo: *e~dmí, *é:stei, (*é:dlah, *é:dla) *é:dla:
C-thematic:
ve`doN, vesti', (vedlU', vedló,) vedlá
Late PIE: *wédh-o: (barytone thematic => mobile), *wed-téi
(oxytone), *wedh-l-ós, *wedh-l-óm, *wedh-l-áh2 (oxytone
o-stem => mobile).
Hirt does not apply.
Winter does not apply.
[analogical after o-verbs, the l-ptc. becomes AP(b):
*wed-láh, *wed-lá, *wed-lá:]
Dybo does not apply.
Meillet applies vacuously.
bo`doN, bosti', (bodlU', bodlo') bodlá
Late PIE: *bhódh-h2a(i), *bhédh(h2)-me (acrostatic => AP I).
Hirt does not apply.
Winter does not apply.
+Dybo: *bad-ó:(N), *badz-téi, *bad-lá: (AP a => AP b)
[Merger with mobile C-thematics: *bàdo:(N), *badztéi,
*bad-lá:].
Meillet applies vacuously.
Note that Meillet, +Dybo and -Dybo can be applied
simultaneously for rule ordering purposes, as their range is
mutually exclusive (Meillet affects a.p. c, +Dybo affects
a.p. a, -Dybo affects a.p. b (i.e. old oxytones)).
gry~zoN, gry'zti, gry'zla
Late PIE: *gWrúh1g^h-o: (= Grk. brú:kho:) or *gWréuh1g^h-o:
(= Lith. gráuz^iu), barytone thematic => mobile;
*gWruh1g^h-téi, *gWruh1g^h-láh2 (oxytone o-stem => mobile).
Hirt: *gWrúh1g^ho:, *gWrúh1g^htei, *gWrúh1g^hlah2 (AP(a)).
No effect on *gWreu&1g^ho: and perhaps analogical l-ptc.
*gWreu&1g^hláh2.
Winter does not apply.
[The mobile present was maintained, but the vocalism was
adapted to that of the infinitive. If an analogical mobile
l-ptc. existed, it was reshaped to AP(b), after nesló, and
then -Dybo applies: *gr(a)u:zló => gr(a)ú:zlo, merging
accentually with any l-ptc. that should have resulted
regularly from Hirt's law: *grú:zlo]
Meillet / (-Dybo?): gr(a)u~zo:(N), gru:'zti, gr(a)ú:zlo
kla~doN, kla'sti, kla'dla
Late PIE: *kl(a)h2-dhí (imperative).
Hirt's law should apply, but was apparently blocked by the
oxytone suffix -dhí. Based on the imperative (*kla:dí), a
mobile thematic verb klá:doN/kla:detí, kla:d-téi,
klá:d-lah/kla:d-lá: was formed. The l-ptc., as with all
C-thematics, was later reshaped into AP(b) -lÚ, -ló, -lá.
Meillet / -Dybo apply (kla~do:(N), kla:stéi > klá:stei,
kla:dlá(:) > klá:dla(:)).
pa~doN, pa'sti, pa'dla
Late PIE: unclear. An acrostatic o-grade verb (a.p. a,
bosti-type) would have remained barytonic due to Winter's
law (*pad- > *pa:d-, and +Dybo does not apply). Maybe an
old perfect (like vêdêti): *pe-pod-, *pe-pd- is mobile.
L-ptc. reshaped to *pa:d-lá-, then with -Dybo pásti, pádlU,
pádlo, pádla.
pa~soN, pa'sti, pa'sla
LIV proposes a desiderative *páh2-s/*p&2-s-', which is
mobile, and where the weak stem would have avoided Hirt's
law.
stri~goN, stri'gti, stri'gla
Whether *streig- or *strig- does not matter.
No Hirt, acute by Winter, analogical AP(b) l-ptc., -Dybo
retracts the accent in inf. and l-ptc.
sê~koN, sê'kti, sê'kla
LIV acrostatic *sé:kH-/*sékH- would not have given a mobile
paradigm, unless secondarily. I have no better ideas.
vI`joN, vi'ti, vila'
LIV *wéih1-e- (=> mobile), escapes Hirt's law. L-ptc. (c.q.
inf.) may originally have been *wih1-ló- (c.q. *wih1-téi)
which would have triggered Hirt's law, but was replaced by
analogical *wei&1ló-. Accent retraction (vi''ti) by -Dybo.
lI`joN, li'ti, lila'
LIV *le:iH-, Rasmussen *leh1i-. Assuming thematic verb
*léih1-e-, see above.
pI`joN, pi'ti, pila'
Root *pih3- (< aorist), thematized: *píh3-o:, *píh3-e- (=>
mobile), *pih3-téi, *pih3-láh2.
Hirt does not apply (*pi-, *pi&3-).
Winter does not apply.
Meillet / -Dybo: *pI`joN, pi''ti, pi~lá.
z^i~voN, z^i'ti, z^ila'
Root *gWih3(w)-. See above (present tense *gWih3w- >
*gWi&3w-, so no Hirt).
We'll do the rest some other time.
=======================
Miguel Carrasquer Vidal
mcv@...