Re: digraphs (was: Rhotics)

In the last episode, (On Saturday 07 July 2007 12:49:36), Jeff Rollin wrote:
> In the last episode, (On Saturday 07 July 2007 12:43:10), R A Brown wrote:
> > Benct Philip Jonsson wrote:
> > > On 6.7.2007 Jeff Rollin wrote:
> > >  > In general, I detest digraphs. The only excuses for them,
> > >  > in my view, are either that they are used in the same or a
> > >  > similar way as those in the language of the conlanger's
> > >  > intended audience (for example, I assume that the audience
> > >  > for most people on this list would consist of English
> > >  > speakers, in the main, so using "ch" for a sound like that
> > >  > in "chutney" is ok),
> > >
> > > I might grudgingly use digraphs for that reason e.g. if I
> > > used a conlang in a story which was also adressed at a non-
> > > linguistically inclined audience,
> >
> > Quite so, I might _grudgingly_ use them in such circumstances. But I am
> > not over-fond of digraphs. Indeed, there are possibly some on this list
> > who may recall that a few years back, when on that other list there was
> > a proposal to reform the auxlang Novial, I made the modest suggestion
> > that its two digraphs, _ch_ and _sh_, should scrapped.
> >
> > [snip]
> >
> > >  > OR the reading is intuitive - thus using "h" for
> > >  > aspirates, etc.
> >
> > 'intuitive'? You mean because an aspirate has an /h/ off glide, I guess.
> > In other words it can be analyzed as two sounds, hence two letters.
> >
> > But - and I thought this was obvious - that is *precisely* why I
> > _suggested_ using two letters for taliesin's sound which he currently
> > denotes as _H_, i.e. it is, as taliesin himself has confirmed, two
> > sounds, therefore it seemed reasonable to me that he might consider
> > using two letters.
> >
> > I was somewhat taken aback that such a suggestion (and that's all it was
> > - the decision what to use after all is taliesin's, not mine) should
> > prompt a bit of soap-box ranting.
> >
> > > I will in general only use digraphs for aspirates,
> > > affricates and secondary articulations (e.g. consonant + y
> > > for palatalized consonants in Kijeb).
> >
> > Yes, indeed - and taliesin's sound is surely just as composite as an
> > affricate.
> >
> > In short, I am not in principle in favor of using digraphs for _single_
> > sounds; nor did I suggest this. Nor, on the other hand, am I dogmatic
> > about it.
> >
> > For example, I certainly would not favor spelling /v/ in a reformed
> > English spelling as _bh_ on the analogy of _ph_ = /f/. In fact I would
> > strongly argue in favor of _v_ and _f_ for the two sounds respectively.
> > But if a conlang has a system of lenition similar to Gaelic, then I
> > think a very good case can be made out for using digraphs to represent
> > the single sounds /v/ and /f/. (Note: they are *not* aspirates, they are
> > simple fricatives).
> >
> > It surely depends upon the internal structure of the language and, if it
> > has one, the conhistory of the language and (just as importantly,
> > probably) the predilection of the language's author    :)
> >
> > I have stated why I personally do not think r-caron is a good idea for
> > taliesin's _H_, and what my own preference of the choices are, and I
> > also _suggested_ a two-letter combo might be considered for a two-sound
> > combo. But I am certainly not going to get on my soap-box about any of
> > this. In the end, surely, it's taliesin decision that matters.
>
> Sorry if I offended! I had another look and yes, you're right, The Phoneme
> Currently Written As 'H' is a composite. You're also right that "if a
> conlang has a system of lenition similar to Gaelic", it makes sense to use
> digraphs to represent /v/ and /f/. OTOH, IMAO that's looking at it more
> from a "I want it to look like Gaelic" standpoint than a strictly
> linguistic one - there's no reason, for example, why you couldn't have a
> language that /sounded/ like Irish (or Gaelic), but /looked/ (as far as
> possible) like Welsh, English, Finnish, or even Russian if you're prepared
> to make you readers learn the Cyrillic alphabet!
>
> And yes, in the final analysis, it IS taliesin's language.
>
> So again, sorry if my soapboxing ruffled feathers, that certainly wasn't
> the idea.
>
> Jeff
I suppose it also depends on how big you want your readership to be - a
fantasy novel like LOTR is surely famous enough to attract an audience of
people who will persist in pronouncing Minas Tirith, for example, as /main@s
tirith/, even in the face of Tolkien's deliberately un-linguistic appendices
and (in this day and age) Intarweb resources like Fauskanger's. This is
probably the reason why Tolkien chose "v" and "f" instead of the equivalent
Welsh "f" and "ff", though why he then chose to use "ch" for /x/ is less
clear: It appears in English in "loch", but in almost everything else, is the
aforementioned /tS/ or /k/ - you might get away with the latter but any
Sindar listening to Hobbits or Men pronouncing "i chammad" / "the clothing"
as /ai tSamad/ is likely to be non-plussed (in either sense reported by
wiktionary, and possibly both simultaneously).

Jeff

--
"Please understand that there are small
European principalities devoted to debating
Tcl vs. Perl as a tourist attraction."

                            -- Cameron Laird

Reply