Sunday, 13 August 2017

Pagan Anglo-Saxons, Phonology, and Christian Lexis 9

The thesis advanced in this series of posts is that there existed vernacular Latin-speaking communities in sub-Roman eastern Britain, some of which were Christian. At least some of these were the communities encountered by the Germanic settlers, who first arrived as foederati guarding the major civitates, but later set up their own small communities.

Archaeological evidence and Gildas point to a fair degree of peaceful coexistence between the communities at various periods before and after the ‘Saxon revolt’. I suggest that it was in this context that the earliest Latin loanwords entered Old English, and that the reason why there are few Celtic loanwords in Old English is that the ordinary language of lowland Britain was late Latin; the settlers did not encounter British-speaking communities until they had penetrated much further west. Perhaps even the reason why the British were known by the name Wealas as well as by their own name Brettas was that the former term referred at the time of settlement to Latin-speaking foreigners, as it did in other early Germanic languages; though there are arguments against this view.

Some of the Latin vocabulary that entered Old English at this time was Christian. It seems less likely that the settlers would have embraced Christianity; though given the archaeological evidence that cultural customs were borrowed in both directions, it is not entirely impossible. It should be borne in mind that very soon after these communities had developed into kingdoms, modelling their kingship on neighbouring monarchies, they did embrace Christianity. It is, however, more likely that some resident communities, pagan and Christian, shifted their language to Old English. It is quite possible that the residents did this readily, considering identification with the incoming culture as generally beneficial. Hostility to the incoming culture on the part of the inhabitants is an assumption for which we have only slim evidence. Identification with North Sea Germanic material culture seems, on the basis of archaeological evidence, to have been widespread. We could even understand some of the tensions existing within sub-Roman Britain as related to disagreement over whether to adapt to the powerful North Sea culture or to remain connected to the disintegrating Roman imperial legacy.

On this hypothesis it is suggested that in the course of adopting Old English a resident Latin-speaking Christian community carried over many Latin words, including Christian ones, and they also made ‘folk’ coinages for Christian concepts using Old English terms, some of which would have been unlikely to be coined by theologically sophisticated Roman missionaries. As has been pointed out, there are parallels for the acquisition of loanwords from language A in language B when an A-speaking community shifts to using language B.

A term which may be associated with the acculturation that I am putting forward is Læden / Leden. This word does not (as far as I am aware) have West Germanic cognates, and in any case, although it is not a specifically Christian word, it has it in common with them that it was unlikely to have been added to the vocabulary of ‘barbarian’ continental Germanic speakers; they already had their own terms for foreigners and their language (e.g. the etymon of Wielisc). It shows the proto-Romance sound change, lenition of medial /t/ to /d/. It shows i-umlaut. Hence it probably belongs to the group of pre-550 insular borrowings described in earlier posts. If there were a Latin-speaking community in southern Britain in the fifth century their term for their language would have been Latinum (or *Ladinu). The hypothesized shift by this community to speaking Old English might have been the reason why this form of the word was the ordinary word for Latin in Old English rather than one borrowed from ecclesiastical usage with t and perhaps a and i.

If there actually was a British vernacular late Latin, from which loanwords of the above second group were transferred, what phonological characteristics might it have had?

As regards vowels, the presumption is that the system was the same as that of the Latin from which Bryttonic borrowed; i.e. distinctions of length were maintained. The short high vowels /i/, /u/ were not merged with the long mid vowels /eː/, /oː/ (compare Welsh cadwyn ‘chain’ < Latin cateːna, Welsh ffydd ‘faith’ < Latin fides; Welsh ffurf < Latin foːrma, Old Welsh loc ‘place’ < Latin locus).

Short /i/ and /u/ may have been lowered in closed syllables, but retained in open ones. For example the primitive Old English base of chest seems to have had /e/.

OED (unrevised): ‘Old English cest, *ciest, cist, cyst (< *cesta) strong feminine, apparently an early adoption of Latin cista, < Greek kiste box, chest. Compare Old Frisian kiste, (Middle Dutch kiste, Dutch kist), Old High German chista (Middle High German and German kiste) < *kista strong feminine.’

The continental West Germanic forms all go back to a Latin form with /i/, whereas the Old English one goes back to one with /e/ (subject to Palatal Diphthongization). The OHG initial <ch> seems to imply an early loan. So the continental forms may reflect a very early borrowing before late Latin lowering of /i/ > /e/. It is difficult to see why this early continental loan did not reach primitive Old English (unless cist reflects it, rather than showing /i/ < /ie/). The Old English form has to be older than the first quarter of the sixth century, since palatalization is older than I-umlaut. This seems to make it an early loan, insular, and showing /i/ > /e/ in its Latin etymon.

(This would contrast with apparent raising of short mid vowels in closed syllables: cf. early British gravestone inscriptions reading ic iacit.)

The mid vowels were not diphthongized (either the reflexes of Latin /eː/, /oː/ as /ei/, /ou/ or the reflexes of Latin /ε/, /ɔ/ as /iε/, /uɔ/).

Latin long /aː/ remained separate from short /a/, as in Gallo-Romance. The reflex of Latin ae seems also to have been maintained as a separate long mid open vowel.

It seems to have shown late Latin voicing of intervocalic voiceless consonants, though only /d/ appears in Old English loans because primitive Old English replaced intervocalic /b/ and /g/, which it lacked, with /p/ and /k/. Probably Latin intervocalic /b/ was already /v/ as this is a very early Latin change. Intervocalic /g/ (where not palatalized) is likely to have been /ɣ/.

The palatalization of Latin /k/ before front vowels and /j/ was not very far advanced, perhaps only to /kj/. Intervocally it had not been voiced. Latin /g/ in the same environment = /j/, but after /ŋ/ presumably = /gj/. There is some evidence that initial /gj/ was borrowed as /j/, hence palatalization may already have been in train. Old English gimm ‘gem’ may be an insular borrowing from Latin. It shows early raising of /e/ to /i/ before a nasal and initial /j/ is confirmed by Laȝamon (Caligula)’s ȝimme. Possibly Latin initial g before front vowels had become /j/, rather than /gj/, borrowed as the same sound in Old English.

No comments:

Post a Comment