Saturday, 13 December 2008
Nustr Padr, ke sia i llo gel:
sia senghid tew nôn:
gwein tew rheon:
sia ffaeth tew wolont,
syrs lla der sig i llo gel.
Dun nustr pan diwrnal a nu h-eidd;
e pharddun llo nustr phechad a nu,
si nu pharddunan llo nustr phechadur.
E ngheidd rhen di nu in ill temp di drial,
mai llifr nu di'll mal.
Per ill rheon, ill cofaeth e lla leir es ill tew,
per segl e segl. Amen.
Revelation time. It's not Sard or Corsican, neither Aragonese nor Asturian; not Rhaeto-Romansh and certainly not Walloon.
In fact, it's a clever attempt to imagine a form of Romance language that might have developed in sub-Roman Britain had spoken Vulgar Latin swamped and killed off late British, as it did the closely-related Celtic language Gaulish.
For this contra-factual history to have happened, certain things would have had to have occurred. The interaction of Latin and British followed two phases, each the mirror image of the other. That we can discern these two distinct phases is due in part to the fact that the ways in which languages borrow words and structures from other languages with which they are in contact follow certain patterns. We need first to distinguish two types of borrowing: structural and lexical. Structural borrowing refers to the use by language A of a grammatical contruction properly belonging to language B. An easy example in English is the Hiberno-English use of 'after' to make a perfect tense: 'I'm after breaking the window', i.e. 'I've (just) broken the window', representing a literal structural borrowing of Irish Tá mé tar éis an fhuinneog a bhriseadh. But lexical borrowing, on the other hand, merely refers to the use by language A of vocabulary items drawn from language B.
Structural borrowing is generally a sign that the language being borrowed from is weakening, because it originally arises from imperfect learning of the target language. A British example is the Welsh pluperfect tense: Celtic did not originally have one, and the Welsh pluperfect is ultimately a structural borrowing of the Latin one. Compare:
Latin: amaverat, 'he had loved', which looks like the [perfect stem of the verb] + [the imperfect of the verb 'to be'] (erat, 'he was').
Welsh: doethoedd, 'he had come', which looks like the [perfect stem of the verb, i.e. doeth] + [the imperfect of the verb 'to be'] (oedd, 'he was')
This would have arisen when native speakers of British Latin had to adapt to speaking late British/Neo-Brittonic, with idiosyncratic results as they translated idioms literally and transferred grammatical structures wholesale. A kind of British would have arisen spoken by people whose familial language was or had until recently been Latin, heavily influenced by Latin grammatical structures: the Welsh pluperfect is a kind of fossil of one of these.
On the other hand, huge lexical borrowing of ordinary words occurs in rather different circumstances. Welsh has a large number of Latin borrowings (about 700 odd in total), but a significant proportion of them are very ordinary words for which there must have been a common British term. They include words like 'fish', 'leg',
'arm', 'cabbage', 'bridge', 'beard', 'green', 'cloud', 'wise',
'floppy', 'smock', 'fruit', 'people', 'cauldron', and so on. Now large-scale replacement of ordinary lexical items like this occurs into language A from language B when language A is weakening and language B strengthening. This class of borrowings is not, you'll notice, made of things which the Romans brought with them which the British didn't have before; they're not like English 'bungalow' or 'pyjamas'. (There is a large class of such borrowings in Welsh, but they're not our concern here.) They are ordinary, everyday words, and thus signs not of enrichment but of an accelerating process of linguistic impoverishment, reflecting social uncertainty and a period of low political status for British speakers. It's not as though no ancient Briton ever thought to put a slightly floppy cabbage into a cauldron with some fish until the Romans came along and gave them words for these objects. You can see this in modern Welsh, where native speakers will sometimes use an English calque even where there is a very common and widely-used native word.
So, what all this indicates historically is that from the 1st century to the late 4th, Latin expanded aggressively at the expense of British. Latin speakers had higher status than British speakers, and in a situation of unstable bilingualism British speakers borrowed huge amounts of vocabulary from Latin, and probably spoke Latin with some British grammatical features. This is when these ordinary Latin words came to replace British ones. Eventually, a large proportion of Romano-Britons in the lowland zone were speaking Latin as their native language. Had this situation continued, British would have died out, and instead of Welsh, Cornish, and Breton we'd have some strange form of Romance language. (This is exactly what occured in Gaul, and we call their strange Romance language 'French'.) The Nustr Padr quoted above aims to imagine what this Celto-Romance language would look like today. The many Latin words borrowed into British underwent exactly the same sweeping phonological changes as did native British words, and the language of the prayer represents those phonological changes extended to cover late British Latin as a whole, with the inclusion of 'Celtic' grammatical features such as consonant mutation. (So the word for 'heaven', cel, from Latin caelum (cf. French ciel), is soft mutated after the article to give gel.)
So why did this not happen in the real world? In the early 5th century, with the withdrawal of the legions, the linguistic shoe was rather suddenly on the other foot, and British expanded aggressively at the expense of Latin, recouping its former territory. In a second situation of unstable bilingualism, everything went into reverse. Latin speakers now spoke British with some Latin grammatical features (e.g. the pluperfect tense) and presumably a lot of British vocabulary. Eventually British Latin died out altogether as a spoken language, probably around 600. If Caesar had managed to conquer Britain in the mid 1st century BC, as he conquered Gaul, then British Latin might well have had enough oomph behind it to have swamped British altogether, as Gallo-Latin swamped Gaulish.
Anyway, the prayer above is the creation of an ingenious man from New Zealand called Andrew Smith, and you can take a walk in the alternative linguistic past at his website here.