Pagan Anglo-Saxons, Phonology, and Christian lexis 2

In this series of posts I will put forward the following set of ideas. They are not entirely new or original, but I think that there is a strong philological case to support them.

In southern Britain in the fifth century there was a significant Latin-speaking Christian British population. This is evidenced by pre-550 Latin loanwords in Old English and in particular by Christian terminology. The loanwords are most likely to be due to this population adopting Old English as their first language. This may be why there is also an important group of religious terms coined in Old English rather than borrowed from Latin. And it also explains why there are virtually no British loanwords in Old English.

II. Phonological background requirements

In my previous post, I suggested that eight Christian vocabulary items must have been acquired in Old English between the settlement period and the Augustinian mission, and neither before the settlement not after the mission.

In order to establish a case for this it is necessary to have a rough chronological outline of some important sound changes in primitive Old English and early post-classical Latin, and of the phonology of Latin loanwords in Old English.

1. Old English sound changes.

An outline of the major sound changes is necessary, because their presence or absence in words where the conditions for their operation exist enable us to roughly date the assimilation of loanwords, and in the present case Latin religious terms.

It seems to be generally accepted that the migration occupied a long period of time, perhaps two centuries or more. During this period a great many significant sound changes took place which differentiated primitive Old English from its West Germanic relatives. Not unexpectedly changes in vowel quality, alone or in combination, were the most prominent ones, but certain consonantal changes were also important.

Important changes were:

1. A group of low-vowel changes: Common West Germanic /aː/, /a/ and the first segment of /au/ became front vowels; Common West Germanic /ai/ was monophthongized to fill the place of the older /aː/
2. So-called ‘breaking’ which, caused front vowels /i/, /e/, /æ/ preceding certain consonants to become diphthongs (perhaps at first /iu/, /eu/, /æu/, later io, eo, ea).
3. Palatalization of velar consonants /k/, /g/, and /ɣ/ in the neighbourhood (and especially followed by) front vowels
4. Diphthongization of front vowels after palatal consonants (which most authorities believe happened at this stage, though there are arguments for making it subsequent to no. 5)
5. Umlaut, when /i/ or /j/ occurred in the next syllable, of back vowels /u/, /o/, /a/, to become at first /y/, /ø/, /æ/, and also of some diphthongs, with further changes following on
6. Voicing of intervocalic /f/, /θ/, and /s/ to /v/, /ð/, and /z/, and loss of intervocalic /x/
7. Loss of unstressed short syllables in various places, especially after a heavy syllable, and shortening of unstressed long syllables

This package of changes was complete by the time of the earliest written records of Old English, i.e. by the late seventh century. In the absence of written records before the seventh century, it is difficult to assign dates to the changes, but if we are right about the timing of no. 5 (see below), it is likely that they were all completed by the end of the sixth century.

All these changes must have occurred after the start of the migration, except for 1, as all are reflected in insular place names. Even 1 may have still been in progress, or possibly non-Germanic /a/ was borrowed, after 1 was complete, at least in some circumstances, as Primitive Old English /æ/. 2 is reflected in place names such as Beornice.

The important thing to note about this package of sound changes is that most of them were either unique to primitive Old English, or specific to it at the time identified (so, for example, intervocalic voicing and loss of unstressed syllables also occurred in other West Germanic languages but quite independently). However, there is one exception: no. 5, i-umlaut, which, although it clearly occurred well into the settlement period, was shared across the whole Germanic continuum of languages. (Its terminus a quo is secure because both it and the earlier No. 3 affect British place names, and must have occurred during the early settlement period.)

Now this is a singular occurrence. Insular Germanic was well on the way to differentiation from its continental relatives, but right in the middle of that process occurred a major linguistic change which, starting on the continent, penetrated insular Germanic thoroughly. (The possibility that i-umlaut originated in Britain and spread to Germany and Scandinavia seems unlikely.)

The point is that i-umlaut is the kind of linguistic change that, in order to spread, requires a continuum of mutually intelligible dialects that are in geographical contact. It is not like a shift in vocabulary, or even grammar, that could conceivably be passed on through long-distance trade contacts. How could insular Germanic acquire a continental sound change once it was physically separated from the continental dialects? The obvious answer must lie in the fact that the migration and settlement of the Anglo-Saxon was a process extending over quite a long period, and that i-umlaut was introduced by a later group of settlers who had acquired it while they were still on the continent and in contact with mutually intelligible Germanic dialects that had it already.

What approximate dates apply to this process? In rough outline, according to the archaeological records the first continental Germanic finds date from the early 5th century; then there is a marked increase in finds and cemeteries dating from the second half of the 5th century, pointing to an increase in immigration from the middle of that century; and then new intrusive artefact types (of Frankish and Scandinavian origin) appear from the late 5th century implying settlement from other emigration areas. Recent archaeological scholarship (e.g. Härke 2011 [Heinrich Härke, 2011 ‘Anglo-Saxon Immigration and Ethnogenesis’ in Medieval Archaeology 55, 1–28]) suggests that a process of acculturation to the incoming community by the remaining British population was under way during the 6th century, so we can assume a tailing-off of settlement by the end of the first quarter of that century. The introduction of i-umlaut can be fitted into the historical sequence by assuming that it came in with a later stratum of migrants.

This agrees well with estimates of the date of i-umlaut proposed by language specialists, viz. the first quarter of the 6th century or at the very latest around 550. Other pointers to the same dating can be adduced, but the above argument seems to be the most certain.

Probably the whole bundle of sound changes was completed during the sixth century, so that by the time of the Augustinian mission, pre-literary Old English had already much of the phonological character that we see reflected in the very earliest texts.

2. Late Latin sound changes.

Latin loanwords shared by the related West Germanic languages show few, if any, of the sound changes that affected later Latin and emerged in the Romance languages. They are borrowings from near-classical Latin that occurred early enough to be spread through the West Germanic speech community. Reasonably secure examples include Old English casere, ceap, cycene, mydd, mylen, mynet, pund, pytt, sicor, stræt, weall, win. These all have West Germanic cognates; they predate the settlement; they participate, where applicable, in the primitive Old English sound changes listed at A; and they are irrelevant to the present discussion.

Later Latin loanwords in Germanic languages increasingly show the results of these proto-Romance sound changes. Early Gallo-Romance changes may be presumed to have relevance also, since Gallo-Romance is the form of Latin which most closely bordered on primitive Old English.

Below is a rough framework of the relevant late Latin sound changes occurring in early borrowings into primitive Old English, followed by subsequent Gallo-Romance (GR) changes that may be relevant:

1. lenition of intervocalic /b/ to /v/ (first century)
2. reduction of intervocalic /v/ to /w/ and then zero before /u/ and /o/ (third century)
3. palatalization of /k/ before front vowels and /j/ (fourth century)
4. palatalization of intervocalic /g/ before front vowels to /j/ (fourth century)
5. lenition of intervocalic /g/ not before front vowels to /ɣ/ (fourth century)
6. lenition of intervocalic /t/ to /d/ (around the end of the fourth century)
7. lenition of intervocalic /p/ to /b/ (around the end of the fourth century)
8. lenition of intervocalic /k/ not before front vowels to /g/ (late fourth century)
9. (GR) secondary palatalization of velars before /a/ (late fifth or early sixth centuries)
10. (GR) fronting and raising of /aː/ to /æː/ (early sixth century)
11. (GR) lenition of new intervocalic /b/ to /v/ (sixth to seventh centuries)
12. (GR) lenition of new intervocalic /d/ to /ð/ (sixth to seventh centuries)
13. (GR) simplification of long consonants: e.g. /bb/ to /b/, after new intervocalic /b/ (from earlier /p/ (around the end of the fourth century) had changed to /v/ (sixth to seventh centuries)

Why do the subsequent Gallo-Romance changes need to be brought into the picture? Because Gallo-Romance was the nearest ‘contiguous’ form of early Romance to primitive Old English. A priori one would expect loanwords from vernacular Romance, as opposed to literary ecclesiastical Latin, occurring during the post-late Latin period, to have the phonology of Gallo-Romance. And if there were a British late Latin dialect, one would expect that it might develop somewhat along the lines of this form of Romance. I will suggest below that in certain respects the British late Latin dialect did not resemble Gallo-Romance; but these differences do not affect the list given above.

3. The phonology of West Germanic and primitive Old English reception of  Latin loanwords.

A third consideration is the small set of phonotactic rules that affected the reception of loanwords. A major principle relates to the place of stress within polysyllables. As regards phonology, certain sounds existing in late Latin did not occur as phonemes in West Germanic, and (as normally happens with oral word borrowing) nearby phonemes were selected instead. These constraints evidently applied to primitive Old English for a certain period.

1. the main stress was shifted to the initial syllable of polysyllabic words from the penultimate (or occasionally antepenultimate) syllable where it fell in Latin words
2. substitution of voiceless /p/ and /k/ for /b/ and /g/ (Germanic dialects at this period lacked the voiced stops /b/ and /g/ in medial (and final) position). There is good place name evidence for this.
3. late Latin /o/ was borrowed as /u/ before a nasal (primitive Old English only permitted the high back vowel /u/ before a nasal consonant)
4. Late Latin /o/ (and possibly /e/) was also borrowed as /u/ before a ‘high’ front or back vowel /i/ or /u/ in a following syllable
5. Late Latin /e/ long or short appears to have been borrowed as /i/ in syllables that were tonic (stressed) or post-tonic (after the stressed syllable) in Latin but became post-tonic in Germanic.
6. Late Latin /ia/ was borrowed as OE io or eo.
7. Late Latin /aː/ entered prOE as /aː/ later than the shift of WGmc /aː/ to /æː/ and subsequent shift of WGmc /ai/ to /aː/