Re: Son of dZ

["Raymond A. Brown" <raybrown@clara.co.uk>]

At 3:59 pm -0400 9/4/99, John Cowan wrote:
>Andrew Smith wrote:
>
>> I have finally got around to looking at _An Introduction to the Celtic
>> Languages_ (1995) and reading the section on orthography.  To recap
>> Brithenig has a voiced final fricative which needs to be distinguished
>> from -g, the voiced velar stop when final.  What I've read suggests that
>> this would be -j (so ffelij).
>
>O Hideous. Eo brodest.

Are Welsh 'garej', 'sosej' etc hideous?

I don't see anything hideous per se - tho the overuse of final -j in
Esperanto does give that language a perculiar inelegance  ;)

>> The evidence: Welsh has borrowed j for all
>> positions from English; one orthography for Breton, the Orthographie
>> universitaire, uses specifically it for word final -dZ#,

I think the book must surely be in error here regarding Breton.  Firstly,
final plosives & fricatives are _devoiced_ (as, e.g. in German & Russian)
and voiced only when in liasion.  But, more importantly, from the
information I have the 'Orthographie Universitaire' (officially approved on
16th June 1955 by the French Ministry of education and the Rector of the
Academy of Rennes) uses {j} exactly the same way as the older 1650
spelling, i.e. to denote /Z/ or, in word final position /S/.

My 'Lexique Breton-Français et Français-Breton' uses the Othography
Universitaire.  I'm sure, e.g. Breton  'lañgaj' is pronounced [lãgaS] or,
in liaison, exactly like French 'langage' (from which, of course, it is
borrowed).

Indeed, in Welsh, Breton & Cornish (Why no mention of this?) {j} is used in
_all_ positions, not just the final position, to denote an essentially
"foreign sound", i.e. the English /dZ/ in Welsh & Cornish and the French
/Z/ in Breton.  In Welsh it is found only in borrowed words and, indeed, is
not reckoned as a letter of the Welsh alphabet by purists.  In Cornish &
Breton it and {ch} (= /tS/ in Cornish and /S/ in Breton) have been more
naturalized; certainly in Cornish /dZ/ has developed in native Celtic words
and in both languages the sounds are subject, to some extent, to consonant
mutation.

>>and Manx has j
>> for /dj/ in an orthography created for that language by a native Welsh
>> speaker.

Bishop Phillips was, indeed, a native Welsh speaker but the orthography he
bequeathed to Manx (and which now sets written Manx apart from its Gaelic
sisters of Ireland & Scotland) is a mishmash of some Welsh conventions
(e.g. {y} = /@/, {w} = /w/ or /u/), 16th century _English_ conventions ({j}
was most certainly not used in Welsh in the 16th cent) and some original
ideas of his own (e.g. {th} to denote the dental /t/ as opposed to the
English alveolar /t/).  The result that we have {ch} with two different
values: /x/ as in Welsh for the lenited /k/ (written {c} or {k}); /tj/ i.e.
"slender" or palatized 't' of Gaelic which is similar to English /tS/.

Bishop Phillips used {j} for the "slender d" which is similar to English
/dZ/ and the letter is taken straight from 16th cent. English.

>Brithenig, though, isn't a Celtic language: it's a Romance language,
>and its orthography has a Romance foundation.  The f vs ff

Following the conventions of Old English (still preserved in our 'of' and
'off'), as Welsh did.  But when I suggested that maybe the Old English
convention of {cg} = /dZ/ might be used for the final /dZ/ of Brithenig, it
was summarily dismissed.  Methinks there's a little inconsistency here ;)

But, as Andrew probably recalls (maybe in emails we exchanged before the
list got extended :) , I've never been keen on the use of {f} = /v/ and
{ff} = /f/ in Brithening.  Both Breton & Cornish use {v} and {f}
respectively for the two sounds and, indeed, in medieval Welsh some scribes
followed the post-Norman practice and used {v} = /v/ and {f} = /f/ also.
Personally, had I "done a Brithenig" before Andrew I'd have used an
orthography more like Breton than Welsh.  But I think it's too late now to
change this in Brithenig & most people seem happy enough with it.

>and d vs dd
>are just decorations, and ll is just as much Romance as Celtic,
>though with a different realization.

Yes, Romance usage has both {ll} and {lh} as "modification of l" - tho the
modification is palatalization and not fricativization :)   Following
Spanish use of {ll} and {nn} (the latter now written with one n on top as
the tilde) to denote palatalizations, Basque has {tt} and {dd} to denote
patalized 't' and 'd' so there is some precedence for {dd}.  In any case, I
think that either {dd} or {dh} are likely to have been used by Romance
scribes (unless they borrowed Old English edh ;)

But John's observation that Brithenig is a _Romance_ language is, I feel,
most valid.  It is simply not in Romance tradition for final {j}, which in
earlier times would've been read simply as a variant of {i}, to denote a
consonant.  Also it seems to me very unlikely that a letter would be used
for a sound only in final position.  Elsewhere in Brithenig, as in modern
Italian, /dZ/ is written as (soft) {g}.

>
>I still like ffelig', which can even be accounted for as a false
>etymology, like French poing, doigt (which never had any g there
>since Proto-Romance times); the implication is that the scribes
>believed that ffelig' < *ffelige.  It also works for ync' (inch,
>or less likely ounce) from UNCIU.  Otherwise we need another
>ad hoc solution for final voiceless fricative.

Indeed, we would.  One could, of course, follow Romanian practice where
final -i merely denotes palatalization (final /i/ must be written {ii}) and
have 'ffeligi' and 'ynci' respectively.   But, unlike English, neither the
Romance nor the Celtic langs have eschewed diacritics.  I agree with John
that a final -g and final -c with a diacritic (either a dot or an acute
seem most likely) is probably the best solution.

Ray.