Re: Virus Translation Exercise

On Thu, 7 Mar 2002 08:27:59 +0200
"Y.Penzev" <isaacp@...> wrote:

> Shalom, Stephen!
> From: Stephen Mulraney <ataltanie@...>
> >   lenition following the possessive: 'a mhadra' /A wAdrA/ his dog,
> >   'a madra' /a mAdrA/ her dog.
> Waah, thnx Heaven, I've found smbody who knows how to represent Irish with
> IPA (or SAMPA)!! Can you send me the rules of reading Irish (may be done
> off-list, my e-mail is isaacp@ukr.net )? I found NORAID site with a very
> good Irish primer, but the pronounciation is explained through English
> approximate spelling, so it only confuses...
> With friendly attitude,
> Yitzik

I'm sending this to the list in case there are people here who
(1) might be interested in reading about Irish phonology
(2) can correct some of my mistakes

One the negative side, i'm no expert, but on the positive I've done
most of the work, so experts can nitpick (there are some pretty big nits
in there too), I mean clarify.

-------------------------------------------------------------------------


OK, here's my sketch of Irish pronunciation. I am *not* a fluent
speaker, not even 'ardly, but I do find myself in the curious position
of being extremely familiar with this lang since early childhood,
without ever having spoken it for any length of time. As a result, the
spelling and pronunciation seem nearly as familar as English's, as does
the mode of discourse and "sentence shape", even if actual grammar and
vocab elude me.

There are bound to be mistakes here, and where there are not mistakes,
there are confusions of dialects into one big mess (that's what happens
when you've had about 10 different teachers, all with very different
idiolects). I try to note dialect differences where I can, but I
probably miss a lot, and I'm not even going to try mentioning which
dialects have which features ;)

I'll send this to the list in the hope of getting corrections etc, and
since other people might be interested in it anyway, so I'll start at
the start... with leathan & caoil... so this is written for a general
(conlanging) audience, which explains, Yitzik, why I try to describe
russian orthography to you ;)).

I use an apostrophe for the long mark (fa'da)

Irish has (like slavic langs) two series of consonants (phonemically
speaking), called 'leathan' (broad) and 'caoil' (narrow). Caoil
consonants are palatised, while leathan ones are velarised. Rather than
having two graphemes for each leathan-caoil pair, say one for leathan
't' and one for caoil 't', the difference (like in slavic langs, or at
least russian) is indicated by the adjoining vowels. In russian cyrillic
there are two sets of vowel letters, then, e.g. one for /A/ indicating
that the preceeding consonant is (in Irish terminology) leathan, and one
for /A/ indicating that the preceeding consonant is caoil.

Irish only uses one set of vowel symbols, with [i] and [e] indicating
caoilness and [a], [o], [u] for leathanness. Hence there are a lot
of vowel letters in Irish words which are not pronounced as such,
but notate the pronunciation of the consonants. So [fear] ('man') is
pronounced /f'A4/ ('=palatalisation), where the [a] is a real vowel /A/
and the [e] just indicates that the [t] is caoil. For consistency, there
is an orthographical rule "caoil with caoil and leathan with leathan",
which insists that a medial consonant (or cons. cluster) needs to be
surrounded by vowels of the same type. So while the future tense of the
verb [bris] ('break') is [brisfidh], the future of [o'l] ('drink') is
[o'lfaidh], pronounced /brISI/ and /o:lI/ [the [f] is not pronounced in
this ending]. Note that /S/ is the caoil of /s/, and [l] is leathan.
So the [a] in [o'lfaidh] is just there to prevent caoilification (sorry
John!) of the [l].

OK, now the sounds:

VOWELS ------ This is very rough, I really need someone who has studied
Irish vowels to fix this. To get these sounds I'm thinking of how some
simple words are pronounced (by me!) and transcribing, and using an old
grammar book (Dillon & O' Cro'ini'n's "Teach Youself Irish" (1961)) as
a sort of a road map to make sure I'm not missing anything (they don't
give ipa equivalents).

[e] /e/ ~ /E/
[i] /I/
[a] /A/
[o] /o/ ~ /O/
[u] /u/ (this is miles out, I think - /U/ better? - or something between /u/ and /o/?)

Unstressed vowels are sometimes /@/.
Long vowels are ... longer ;). For example, [a'] is somewhat like
English [ar] in RP pronunciation of [farce].

Diphthongs? Well [ia], [ua] are /IA/ and /UA/, real diphthongs.

According to my book, though I never noticed, [ao] is a digraph for
/E:/, or when before an [i] or [e] it's /I:/ (or /i:/?).
Hence [taobh] = /tE:v/ ('side') and it's genitive [taoibh] = /tI:v'/
Well now that  I think of it makes sense. (the book goes for Kerry
dialect, BTW - rather alien to me! I think it's [on average!] Connemara
that I learned at school).


Consonants

~ means that both sounds seem right to me, but they may well be
allophones or dialect variations. When I write, in the leathan column,
a simple phone like /t/, bear in mind that this could be velarised to
distance it from the caoil form. A few items might be peculiarites of
my own pronunciation (e.g. the bilabials, suspicious because I have a
fondness for bilabials), so I note them, since I don't know if anyone
else uses them (they're hardly wrong though - they don't interfere with
anything else)

Grapheme     Leathan         Caoil


[p]          /p/             /p_j/
[ph]=[f]   .........................       I tend to pronounce [ph] as [p\]. Seems
                                           more natural. (wrong?) Might arises by
                                           analogy from my habit with [bh].
[b]          /b/             /b_j/
[bh]         /v/ ~ /w/       /v_j/ ~ /w/   Dialectical. Possibly the same distribution
                                           as for [mh].  Probably as a result of not
                                           knowing whether I speak with a /v/ or /w/
                                           here, I tend to pronounce [bh] as [B].
                                           Maybe I heard someone else do it...?
[f]          /f/             /f_j/
[fh]         silent          silent
[m]          /m/             /m_j/
[mh]         /v/ ~ /w/       /v_j/ ~ /w/   (I think /v/ in Munster: /w/ elsewhere.
                                           Dialectical anyway.)



[t]          /t/ ~ /t_d/     /t_j/ ~ /c/   (& can be /tS/ or /t_jS/ rather, medially.
                                           If you're slurring or speaking quickly)
[th]=[h]
[d]          /d/ ~ /d_d/     /d_j/ ~ /\J/  (& can be /dZ/ or /d_jZ/ rather [unless you
                                           have two tongues], when medial. Again, if
                                           speaking quickly)
[dh]         /G/             /j/           I'm sure this varies a lot with dialect,
                                           but I can't quite seperate the distribution
                                           in my mind from the caoil-leathan
                                           distinction.  My confusion might extend to
                                           [gh] too. Help!
[n]          /n/ ~ /n_d/     /n_j/ ~ /palatal nasal/  - can't find the xsampa for it!
[nn]         /nn/            /N/           finally [or medially & finally in some
                                           'lects?], else /nn/



[c]          /c/             /c_j/
[ch]         /x/             /C/
[g]          /g/             /g_j/
[gh]         /G/             /G_j/

[h]          /h/             /hj/ ~ /c_o/  (_o=lowered: somewhere between the
                                           fricative /c/ and the approximant /j/)
[l]          /l/             /l_j/ ~ /L/
[s]          /s/             /S/
[sh]=[h]
[ts]=[t] when it occurs initally as a result of the definite article. But in some
         dialects, I suspect /ts/ is how it's pronounced.
[r]          /4/ ~ /r/       /4_j/ ~ /r_j/  Initally, may be just /4/ ~ /r/ If caoil
                                            [r] is *not* inital, it can approach the
                                            sound of /z/ or /Z/. Not sure how to
                                            describe it in IPA, but it arises I think
                                            fairly naturally if you try a trilled
                                            palatized /r/.




Eclipsis doesn't introduce any new sounds except that I would pronounce
initial [ng] as /N/.

As a test, I'll try writing out my pronunciation of a line or two of
Flann O'Brien's "An Beal Bocht" (the poor mouth). I won't be consulting
my table above, but saying and transcribing. Just for fun I'll choose
a section where there's some English, written in phonetic Irish from a
monolingual Irish speaker's point of view ;) (BTW, part of the joke is
that every boy in the class is called "Jams O'Donnell" ;)) )

Two following vowels aren't a diphthong unless I bracket them.
(How do you really indicate diphthongs in IPA/X-SAMPA?)

This is not competelty consistant - It's just what came out of my mouth
without much premeditation.


"*All spik English?*"
 ...................

"*All spik, sor*", arsa an Seanduine.
 ...............   ErsA An SAndIn_j@

Ansin   thug an duine  uasal mise   faoi deara  i mo sheasamh ar chu'l an
AnSIn_j hug  An DIn_j@ uAsOl m_jIS@ fwe: d_jAr@ I mo h_jAsAv  @4 xu:l  An

tSeanduine  agus labhair se' go borb liom.
t_jAndIn@   Agus lAw@r_j Se: g@ borb l_jum.
                    (r_j ~ 4Z nearly)

"*Phwat is yer nam?*" a du'irt    se'. Bhi' mise oilte    go maith ar an obair
 ..................   @ du:r_jt_j Se:  vI:  mIS@ ol_jt_j@ g@ mAc_o @4 An obIr_j
                                                                          (or near 4Z)

a bhi' idir  la'mha an la' sin   againn agus d'fhreagair gan aon chu'thail e'.
A vI:  IdZI4 lAv@   An lA: SIn_j OgwIN  Agus d@r_jAgIr_j gAn a@n xu:hAl_j e:

"*James O'Donnell, sor.*"
 .....................



Comments? Corrections? - many needed!

(in particular can anyone clean up the mess I've made of the vowels, and
the [dh] and [gh] sounds?)

stephen