Thl- vs. Fl-

[llama_nom]

--- In gothic-l at yahoogroups.com, Guenther Ramm <ualarauans at ...> 
wrote:
>
> First, thanks a lot for drawing attention to D'Alquen and his idea 
of the Gothic digraphs. For me, the view on those "earlier" 
borrowings (where Greek /e/ and /o/ are "gothicised" into /i/ 
resp. /u/) as reflexes of phonetic peculiarities of the Wulfilan 
dialect (if I understood it correctly) was especially enlightening.


Yeah, it's a very interesting book.  The more traditional view is 
that these anamalous spellings point to Pre-Gothic loans, i.e. from 
before the time of Wulfila's supposed bible translation.  But 
d'Alquen's proposal that they're relics of the original Wulfilan 
spelling helps to explain some otherwise puzzling aspects of the 
writing system, especially why <ai> and <au> were used instead of 
simple letters based on epsilon and omicron.  (Incidentally, 
regarding another part of his study, I wonder if anyone has made a 
similar assessment of Gothic names in Greek sources and loanwords 
into Slavonic.)


> I guess it would also help to explain why the loanwords of the 
Greek second declension with -os in Nom. Sg. so regularly go after 
the Gothic u-stems.


Yes, although the traditional explanations are maybe adequate; even 
with phonemic /O/, this declension might offer the closest match in 
sound with Greek -os nouns.


> Btw, is of any effect for the topic that old opinion that a good 
deal of these "narrow" spellings could be results of a pseudo-
etymological interpretation? - that, e.g. -i- in Makidonja was seen 
as a stem vowel (popularly "dismembered" in Maki-donja), -i- in 
aggilus as a part of the suffix -il- etc.? Cf. the seemingly 
deliberate rendering of Greek bEthleem as bethlahaim, where the 
first epsilon > a (quasi a stem vowel of the first element in the 
compositum), and the second > ai (a diphthong after the Germanic
place-names on -haim). What this bethla- could mean then?


I guess the <o> of Makidonja, if it represents [o:], could be due 
either to the Greek stress or a secondary stress in Gothic, if that 
syllable was pronounced like the second part of a native compound.  
Maybe due to a loss of distinction in unstressed vowels, or a 
raising of [a] > [E], but <e> is very common as a stem vowel in 
Latin transcriptions of Gothic names, cf. Suniefridus = 
Sunjaifriþas.  I wonder what bearing that could have on the question 
of when an how [E] was phonemicised.  D'Alquen saw loss and 
assimilation of postvocalic /h/ as an important stage in this 
process, but another factor could be the 
prepositions/prefixes 'faur', 'faura' and the prefix 'fair-'.  If 
breaking only happened under stress (as seems likely), do these 
represent originally stressed or half stressed forms that have been 
extended by analogy to unstressed positions?  Did analogy extend to 
pronunciation, or is it just orthographic?



>   To the supposed confusion of the letters *faihu and *thiuth. 
Reading Jordanes long ago I noticed the name Safrac (Get. 134, one 
of Ostrogothic "primates"), mentioned also as Saphrax by Ammianus 
Marcellinus and Zosimus. The name itself is most probably of Alano-
Iranian descent (as so many Getica names ending on -ac are). I am 
almost sure that I was not the first to make the following 
supposition, but evidently I had missed the proper literature. In 
short, I supposed that all the above-mentioned authors, successively 
or independently of each other, somehow took this name from a 
written Gothic source (maybe the same source where Gapt-Gaut came 
from), where it stood as *Sathrak(s) < Iranian *Xshathraka- "a 
powerful one" from Avestan xshathra- = Sanskrit kshatra-
 "power", "strength" etc. If anyone knows whether this or some other 
etymology of the name has already been proposed I would greatly 
appreciate the reference. If this one is true, then here is another 
case of the


Interesting.  I haven't seen that idea before, but then I'm not all 
that widely read.  Could the change have been motivated in this case 
by folk-etymology involving the Gothic name element Sab(a)-?



>  involuntary letter replacement caused by graphical likeness (of 
course if we accept that Gothic [f] could ever be written with a phi-
like sign).
>   What concerns our fl-/thl-, so as I understood there are 
etymologies assuming both PIE pl- and PIE tl- for one and the same 
root, aren't there? Rather typical situation for etymologists :) 
Apart from the graphical question I like most the idea that thl- be 
a later and limited (both temporally and territorially) development, 
and
>   > that the assimilative change fl > þl was
>   > conditioned by certain following velar consonants (Krause & 
Slocum
>   > 2006), h, and in one instance q, though not k.,
>   even though
>   > the rule isn't consistently applied: flahtom `plaits', dat.pl.



Nordmeyer certainly explained them all as from PIE *pl.  Unless 
there are strong reasons for positing *tl, the general *pl 
explanation seems simplest.  Now, a question for the statisticians 
among us: which is the bigger coincidence, that fl- spellings are 
confined to Luke and the Epistles (supposedly the areas with the 
least conservative orthography)--this was the distribution that 
Nordmeyer thought most significant--or that the two roots with a non-
velar consonant (out of eight roots attested) should both be among 
the fl- spellings.

VELAR: 4x þl-, 2x fl-
DENTAL: 0x þl-, 2x fl-

The Slocum / Krause hypothesis that requires a special rule for the 
single instance of 'k' seems to me inelegant and arbitrary, 
especially when we can see that 'h' might or might not trigger fl- > 
þl-.  Simpler to suppose that it might or might not happen with any 
velar.  But this begs the question of why.  What qualities would a 
velar consonant have that could cause such a change (let alone what 
quality 'k' would have to make it have a different effect from 'q' 
and 'h')?  It can hardly be assimilation or dissimilation.

Could it be that an assimalative change fl- > þl- happened 
(sporadically? dialectally?) with any sort of consonant after the 
next vowel, except for a dental?  That a dental should block the 
change could be explained as a case of dissmilation.  Note the 
dental after the velar also in 'flahtom'.  That would leave just the 
one anomaly of 'faiflokun', rather than the two problems 
of 'flahtom' and 'faiflokun'.

Finally, does the fact that no one root is attested with both 
spellings suggest that maybe the sound change was not in progress at 
the time of the scribes, but already fixed (allowing the possibility 
of doublets), whether the difference is dialectal or otherwise?  Or 
is this a coincidence (too)?

LN





You are a member of the Gothic-L list.  To unsubscribe, send a blank email to <gothic-l-unsubscribe at egroups.com>. 
Yahoo! Groups Links

<*> To visit your group on the web, go to:
    http://groups.yahoo.com/group/gothic-l/

<*> To unsubscribe from this group, send an email to:
    gothic-l-unsubscribe at yahoogroups.com

<*> Your use of Yahoo! Groups is subject to:
    http://docs.yahoo.com/info/terms/