---- Original Message ----
From: stlatos
To:
cybalist@yahoogroups.com
Sent: Sunday, October 28, 2007 4:46 AM
Subject: Re: [tied] swallow vs. nighingale, SWALLOW
> --- In cybalist@yahoogroups.com, "Grzegorz Jagodzinski"
> <grzegorj2000@...> wrote:
>>
>> ---- Original Message ----
>> From: stlatos
>> To: cybalist@yahoogroups.com
>> Sent: Wednesday, October 24, 2007 8:15 AM
>> Subject: Re: [tied] swallow vs. nighingale
>
>>> Not at all; if there are other Latin words with l>r in a specific
>>> environment they could show a regular rule.
>>
>> Could a rule be irregular :-) ?? (Eng. "rule" < Latin "regula")
>>
>> But seriously: could you list words with l > r and without this
> change, and
>> formulate the rule first than negate my words?
>
> I think the change is l>r. between V_V if followed by n., but more
> exact conditions for the environment may exist. I reconstruct many
> more sounds for PIE than most, which incidentally allows more regular
> rules.
>
>>> The change of *gWdelu+ >>
>>> hiru:do: 'leech' suggests it is so.
>>
>> One example, or even two examples, means no example (especially that
>> there are more differences between Greek bdella and Latin hiru:do:,
> including the
>> vowel in the root).
>
> It was probably contaminated by *gYher+ 'grasp', and maybe hirundo:
> itself. Sometimes e>i by a pal. in Latin; the rule in this case may
> be e>i/#CY_C+sonor. [or only specific combinations?] especially if o>e
> in the same position later.
>
> Instead, please extract from a Latin dictionary all
>> words with intervocalic -l-, as well as all words with intervocalic
>> -r- which have -l- in Greek. Compare both lists, and formulate
>> rules...
>
> Well, one rule already exists and is well-known: l-l > l-r with
> many examples.
It would be very well if it was so. I mean that there are NO strict
(Neogrammarian) rules of dissimilation. But, examples are much better than
outtalking :-).
For l ... l:
1. l ... r in forms with -a:lis and -clum,
2. r ... l in caeruleus < *caeluleus, fragellum < flagellum (both attested,
so not a rule!),
3. n ... l in cuntellum < cultellum,
4. 0 ... l in cavilla < *calvilla
For r ... r:
1. l ... r in pelegri:nus < peregri:nus,
2. r ... l in fragla:re < fragra:re, cri:blum < cri:brum,
3. 0 ... r in taberna < *traberna,
4. r ... 0 in agrestis < *agrestris, praestigiae < *praestrigiae.
One more word on leech in another post.
In Greek, there are also "weak rules" of dissimilation.
For l ... l:
1. l ... r in gló:ssargos < gló:ssalgos, kephalargía < kephalalgía,
2. 0 ... l in ékpalgos < *ékplalgos, phaûlos < *phlaûlos,
3. i ... l in paipállo: < *palpállo:.
For r ... r:
1. Not changed: bárbaros, ararísko:,
2. l ... r: klêros < krâros,
3. 0 ... r: phatría < phratría, thipóbrotos < thripóbrotos, détron <
dértron,
4. r ... 0: drýsphaktos < drýsphraktos, thermastís < thermastrís, krokódilos
< *krokódrilos
5. i ... r: maîtys (Cretan) < márty(r)s.
Finally, for l ... r:
1. r ... l in Áglauros > Ágraulos
In addition, l > r or r > l in some words, such as kríbanos < klíbanos,
amérgo: < amélgo:, aderphoí (in Attic inscriptions) < adelphoí, blýo: <
brýo:, alkéo: < arkéo:, Cretan lákos < rhákos. Sometimes l > n, like in
béntistos < béltistos, Doric Kentai < Kéltai etc.
In Modern Greek -r-forms are yet more frequent - but the change l > r was
never regular.
Sometimes inversion took place (korkódeilos < krokódeilos), sometimes
distant (dríphos in inscr. < díphros), and sometimes -r- appeared ex nihilo
(straté:r < staté:r). We even may give conditions which were conducive for
inversion: after or before dentals, in vicinity of e, i - but they were only
conditions of a weak rule. A strict phonetic rule never originated.
None of the above changes is regular, and most are present in more than one
word
Grzegorz J.
___________________________________________________________
All New Yahoo! Mail Tired of Vi@...@! come-ons? Let our SpamGuard protect you.
http://uk.docs.yahoo.com/nowyoucan.html