Caveat for Candice: my OE goes back to 1970, plus a few years thereafter. So
I am no more up to date. And my knowledge, which time has long since eroded,
was never all that vast.
Caveat for everyone else: I was a poor student of phonology (OE and OHG),
finding it somewhat Grimm: more snafu than gyfu in my case.
Anyway here goes.
First orthography: the insular (or Irish Latin, < Old Irish) symbol at issue
is written 'G' when capitalised and like a 'g' but with the two loops open
(ie a tailed 'z') when in lower case. It covers various phonemes. The two
that are relevant here are nowadays contrasted in the Mod E minimal pair
velar 'gear' and palatal 'year'. The phoneme /h/ (from W Germanic /?/) can
be represented by either 'h' or 'g'; hence Mod E suffix '-gh'. Another
phoneme becomes the stop in the '-ng' of English gerunds and is graphed in
ME and in Middle Scots with a slightly different tailed 'z', which survives
in the Mod E orthography and pronunciation of, say, 'Menzies Campbell'. And
so on...
And so on, in fact, to sound. As implied above, the issue isn't voiced
versus unvoiced: /g/ and /k/ are voiced and unvoiced velar plosives in
Modern English, and as such they are quite distinct both from /dg/ and /ch/
and from /y/. It's position in the mouth.
So, with all that in mind, here is the first line of Beowulf:
Hwæt we Gar-Dena in gear-dagum
The 'g' of 'Gar-Dena' (everything up to -na is capitalised in the MS) is
written as capital G, whereas the 'g' of 'gear-dagum' is in lower case, as
are all the 'g's of line 13, including that of 'God', irrespective of
phoneme.
So why alliterate them? Well the visual link is irrelevant. (Campbell notes
that the runes gyfu and cen, though more systematically used for back and
front respectively, are also swapped about.) The modern argument is, I
think, that it's similar sounding allophones of these two quite different
phonemes that actually make it possible: even though one was back positioned
and velar and the other front positioned and palatal, both were spirant or
fricative in terms of how they sounded. Consider, by analogy, the three
'c' sounds of 'keep', 'cool' and 'calm' which move from the front of the
mouth to the back. We normally ignore this distinction but it's there.) Wren
in 1958 (my Klaeber has gone missing for the moment) makes the same point
more or less:
'The alliterating g of [line 1] shows that at any early date the front
['year'] and the back ['gear'] positions of g still retained their original
fricative or spirant quality; for it is this spirantal quality which makes
the alliteration. Pronounced in the later classical Anglo-Saxon of Ælfric's
time, when the MS was copied, there would be no true alliteration.'
You can then go on (that is, One can; I certainly can't) to assert some sort
of dating of the Original Oral Text, based upon this principle.
Grimm's Law, Verner's Law... and now Walker's Law: Just this one post and
then stop.
CW
_______________________________________________
Dozens have gone missing, the decision taken is Elsewhere.
but yes, yes we remain as poetry, pure immateriality.
in the name of the 'current state of things' they murmur to us:
"we went for a stroll, now it's a question of marching!" But this
stroll of ours has brought us a long way off, and now
the horizon is behind us.
(from *Materiali*, Indiani Metropolitani 1977
|