From: dgkilday57
Message: 70253
Date: 2012-10-24
>I will read it as soon as I can.
> 2012/10/23, dgkilday57 <dgkilday57@...>:
> >
> > --- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy
> > <bhrihstlobhrouzghdhroy@> wrote:
> >>
> >> 2012/10/18, dgkilday57 <dgkilday57@>:
> >> >
> >> > --- In cybalist@yahoogroups.com, Bhrihskwobhloukstroy
> >> > <bhrihstlobhrouzghdhroy@> wrote:
> >> >>
> >> (...)
> >> >> > As for reconstruction, Old Indic bhinná- 'broken &c.'
> >> >> > expectedly
> >> >> > means 'a fragment, bit, portion' as a m. substantive (Sir Monier
> >> >> > Monier-Williams, A Sanskrit-English Dictionary Etymologically and
> >> >> > Philologically Arranged with special reference to Cognate
> >> >> > Indo-European Languages, Oxford 1899 : 757); its prototype
> >> >> > *bhid-nó-s
> >> >> > would regularly yield Celtic *biddo-s (cf. MacBain 1911: 36
> >> >> > *bid-do-)
> >> >> >> Gaulish *Biddos (*<Biddus>, maybe directly attested by Bingen
> >> >> > <biddu[>).
> >>
> >> > DGK:
> >> > Kluge's Law should have given Celtic *bitto-s. The gemination in
> >> > Biddu[s]
> >> > is likely hypocoristic, from a compound name whose prototheme was
> >> > 'bite'.
> >>
> >> Bhr.:
> >>
> >> It isn't Kluge's, it's Stokes' Law; tt < *tn + stress, dd < *dn +
> >> stress. You may not believe in it, but that's its formulation
>
> > DGK:
> > Superseded. See Miguel's comments in message #56156.
>
> *Bhr.:
>
> An extremely short comment without any additional argumentation (pro or contra).
>
> Best formulation:
>
> Ernst ZUPITZA, «Ãber doppelkonsonanz [sÄ«c] im Irischen», Zeitschrift
> für vergleichende Sprachforschung auf dem Gebiete der indogermanischen
> Sprachen. Begründet von A. Kuhn. Herausgegeben von E. Kuhn und J.
> Schmidt. Band XXXVI. Neue Folge Band XVI (Gütersloh, Druck und Verlag
> von C. Bertelsmann, 1900 [IV, 668 S.]), S. 202-245.
> Kluge's Law is best understood if operating after 1stYes. Now, Gallo-Latin <beccus> 'beak' opens a can of worms. If we temporarily disregard the issue of some Romance reflexes pointing to close /e./, we might notice the similarity to Old English <becca> 'ligo' i.e. 'hoe', explained by N. van Wijk (IF 24:232-3, 1909) on the basis of *bHeg- 'to break', the weak noun continuing either *bHegnon- or preferably *bHegon-, *bHegn- with Kluge operating on some of the oblique cases, the geminate then being generalized. If a similar thing happened in Celtic, *bekkon- might have signified 'pecking bird, woodpecker, figpecker, etc.', reinterpreted in Gaulish as 'beaky bird', whence the extraction of *bekkos 'beak'. But several things have to go right for this explanation to work.
> Lautverschiebung, so its very input is different from Celtic (where
> the Law - for those who believe in it - operates before
> dephonologization of */p/); such difference is reminiscent of
> GraÃman's Law in Greek and Indo-Aryan respectively, so OK for
> considering it one and the same tendence, but since the Scholars who
> investigated it were two or more it's fair to mention them.
> Anyway, this is just a matter of labels. What's crucial is the origin
> of geminates in Celtic (and Ligurian, according to the other message
> of Yours). That would deserve a thread for its own, but I don't want
> to start it because I already know what would be its major impasse:
> someone accepts reconstructions of PIE accented words only when their
> continuations are attested by diagnostic languages (especially Old
> Indic and Greek), other ones admit that back-projections from Celtic
> can exhibit different accentuation (for morphological reasons). It's
> like lengthened grade...
> A disappointing detail is that the whole controversy about the LawThurneysen still has followers of his proposed *-tn- > -nd- in Latin, which I had to reject in the post mentioned.
> originated in a personal quarrel between Stokes and the young
> Thurneysen (when reviewing Stokes' Urkeltischer Sprachschatz); since
> then, Thurneysen's followers (Pokorny, Karl Horst Schmidt, Patrizia de
> Bernardo Stempel) have all felt themselves compelled to reject the Law
> >> *Bhr.:Yes, but not impossible. I consider it highly likely that Umbrian <felsva> 'vegetables' vel sim. is borrowed from Sabine, although the great majority of Italic words in Umbrian is native Umbrian. And the very widespread 'pot' seems to have diffused from Trier (in my view of Ligurian origin; Sean prefers Gmc. with a */b/-initial IE root).
> >> Peoples and words do move around, of course. The problem (not just
> >> mine) is: do You think that, ceteris paribus, every word of substrate
> >> origin has definitely the same probabilty to have gone through
> >> diatopic movements than to be simply an in-situ heritage?
> >
> > DGK
> > I do not understand the question. Can you give examples?
> >
> *Bhr.:
> Let's consider only words that we agree in assigning to a given
> substrate, e.g. Gaulish words in Gallo-Romance. Let's take into
> consideration only words that are attested in just one dialect
> (without any Latin or Standard French evidence), say Lausanne's
> Franco-Provençal; I know that this isn't the case of bidet, but I'm
> trying to show the attitudes in a prototypical situation.
> To put it on a personal level (just for the argument's sake), I would
> explain virtually all words of Gaulish origin in Lausanne's dialect as
> Helvetian relics (and those also attested in Genevan as ancient
> isoglosses between Allobrogian and Helvetian); I wouldn't like the
> possibility that some word could be a Treverian one, that was brought
> South of the Jura at the time Trier was residence of the Western Roman
> Caesar and has been preserved only in a conservative milieu like
> Burgundy, disappearing elsewhere and emerging only in Swiss Romand. I
> think You too would agree that the latter possibility is less
> probable.
> Now, let's turn to words with a much wider diffusion. Of course,If I understand you correctly, you are referring to a word which is ATTESTED over a wide area, and asking whether I think it is equally likely that the word spread in antiquity as later. I would say no. I do not believe the a-priori probability (to the extent that means anything) of the two outcomes is equal. Working with substrates is so opaque that some scholars (e.g. Larry Trask) reject them entirely. If substratal words commonly survived over a wide area, this would not be the case. I think it is much more likely in practice that a substratal word attested over a wide area is like 'pot', enjoying its widespread popularity late. But I do not discount the existence of substratal words spreading widely in antiquity and staying widespread. I simply think the probability of the latter is lower, without trying to quantify it, since I have made no statistical study.
> everybody knows that a substrate word can have survived only in a
> limited region and subsequently spread to the rest of its present-day
> area. Nobody would deny, on the other hand, that - unless reasons
> testify to the contrary - the word's area can have been the same or
> even larger in ancient times and the word can have directly survived
> everywhere in an independent and parallel fashion.
> The question is, then: do You think that both possibilities (I
> stress: without additional phonological reasons) have precisely the
> same probability (50% and 50% respectively)?