Litriú a chosaint: Cuid a hAon

[muckefuck]

If moa1918 wants to hear my brief apologia for Irish orthography, perhaps others do as well. For starters, I find that most of the critics don't actually use it, which I'm afraid means they really don't have the tools to judge it. After all, what is most helpful to the learner or casual user is often confusing to a fluent speaker, and clearly it's the latter that any decent orthography seeks to please. When a discussion of Irish spelling came up in a conlang group I belong to, several people attempted to design new systems. None of these people knew any Irish, so it's no surprise that I found many of their so-called "improvements" actually did away with features I found very useful. And I'm a dilettante myself, so I can only imagine what the opinion of a competent speaker would be.

Part of the reason was that they were working off a transcription that reflected the pronunciation of a single dialect (albeit a central one), and this was not the dialect I'm learning. You might well ask, why keep the "silent letters" at the end of a word like bealaigh (genitive of bealach "way")? Well, for one thing, they aren't silent in Munster where this is pronounced ['bʲal̪ˠəgʲ]. For another, which ones are silent isn't even consistent from place to place: Some Connacht speakers have ['bʲa:l̪ˠə] whereas people further north would say ['bʲal̪ˠi:]. Keeping the spelling with -aigh allows you to accommodate all three variants (and more besides) at once.

This, incidentally, is a major challenge of reforming English spelling. The variation in pronunciation from one end of the English-speaking world to the other is huge and any proposal to "simplify" the orthography soon stumbles over the fact that what's "simpler" for one group may add complications for another. The best solution is probably one that preserves the maximal number of potential distinctions. No single dialect has them all, however, which means that everyone is going to end up having to ignore some. Even the Irish system falls short of this ideal, however, with such simplifications as aniugh to inniu, which works fine for most Connacht speakers, but makes mystifying the Munster pronunciation [ə'nʲuv].[*]

Another flaw in most of those suggestions for a reformed orthography is that they didn't preserve morpheme identity. This is a good example of a case where one man's bug is another's feature. One of the things I find most useful about Korean orthography, for instance, is that the spelling of a root reflects the underlying form. Take a word like 닭 "chicken". The spelling is /talk/ even though the pronunciation in isolation is really ['tak̚]. But add a vowel-initial suffix like /i/ to the end, and the /l/ remerges, e.g. 닭이 ['talgi]. In earlier spelling, both forms would be written just as they were pronounced, respectively 닥 and 달기. So you get a one-to-one mapping of pronunciation to spelling, but now it's more difficult to recognise them as just two variants of the same lexeme or recognise that 기 contains the subject particle rather than belonging to the stem as in 비둘기 /pitwulki/ "pigeon".

I know learners who hate this feature of current Han'gŭl orthography and would happily return to the earlier system, but I find it deeply sensible. Similarly, there are probably those who would have no trouble with the genitive of bealach having the fully phonetic form beala or bealaí, but I find bealaigh much more sensible. For one thing, it allows bealach to take its place among the ranks upon ranks of nouns that form the genitive by making a broad final consonant slender. All you need is a little adjustment rule to convert slender ch to slender gh when following an unstressed vowel and it falls in line with thousands of other nouns like turas (gen. turais), cab (gen. caib), and so forth.

[Coming in part two: More about slender vs. broad, and how the Latin alphabet forces kludgy solutions on even the best minds]


[*] A spelling that accurately reflected the dialect pronunciation would be aniubh or iniubh. Inniu suggests */ə'ŋʲu/.