## ------------------------------------------------------------------------ ##

                   Bewnans Ke -- General Impressions

## ------------------------------------------------------------------------ ##

0. The following observations are based on an initial and uneven examination 
of O. J. Padel's preliminary reading of the mss (April 2003). All observations 
are preliminary and require confirmation by careful quantitative analysis.

1. Perhaps the most striking feature of BK is the frequency of whole lines in 
Latin, Middle(?) English, and something which looks remarkably like distorted 
French. Often these line form macaronic couplets alternating with Cornish 
lines. See e.g. BK 292 which consists of 2 lines of Latin, 3 of English and 
3 of Cornish. This stanza also has Eng. <fas> 'face' rhyming with C. <plas> 
'place'. If it is accepted that C /a/ remained a fairly low vowel throughout 
the MC period, then this should indicate that the verse was composed before 
the raising of Eng. /a:/ to /E:/ and beyond as the English vowel shift 
progressed. Given the amount of Eng. in this play it would be interesting to 
extract these lines and present them to one or more ME experts for dating etc. 
Details of the phonology and orthography of the Eng. might then be fed back 
into interpreting the C text. 

2. The text contains a great many words previously only found in the OCV, in 
Lhuyd, or deduced from place names. This might suggest a forger trawling the 
dictionary, but there again is perhaps to be expected, given that these words 
are only in the dictionaries of revived C on the presumption that they 
actually did form part of MC. Some kind of frequency analysis might clarify 
this question. There are also some very Welsh words of which BK 276.2 
<eglynnyon> and BK 438.4 <eglynnyow> is the most blatent example. Another 
less obvious example is _syth_ 'straight, direct' as at BK 287.7 <gans 
gyrryow syth> 'by direct words, in plain words' and BK 427.2 <avel 
bothiak syth> 'just like a hunchback' ???. This is such a commonplace word in 
W that a W speaker might easily have slipped it in without realising that it 
hasn't (hitherto) been used in C. either traditional or AFAIK revived.
 
3. OJP has noted the absence of late MC characteristics, specifically 
<dn, bm> for /nn, mm/ as seen not infrequently in CW. To this example might 
be added the very rare occurrance of Eng. silent final <-e>, and the absence 
of 3 pl prepositional forms like <thethans, thons> etc. Both of these 
developments followed on from the merger of /a, O ,E/ into /a/ in unstressed 
final syllables. The orthography of BK consistently shows <a> for 
Classical MC /E/ in this environment, but the slightly later change of 
/O/ >> /a/ seems to be still in progress. This would date the mss at around 
1550 using the dates given in PSRC, i.e. more or less contempory with Tregear. 
Are we however dating the extent copy or the scribe's exemplar? I.e. was he 
modernising the spelling? Also relevent here is the spelling of _mergh/myrgh_ 
'horses' as <mirth> at BK 378.4 ('at the tails of [wild] horses/ over cliffs 
though I might be dragged') (cf. <marth> for _margh_ at CW 405). PSRC dates 
the change /rx/ >> /rT/ at c1525. 

4. The rhymes however often suggest that at least some of the verses were 
composed before the merger of final unstressed /a, O, E/ into /a/. One 
particularly telling example is at BK 111.1/2 where <ryban> _rybon_ is 
rhymed with <deson> _di'son_. This would have been a perfect rhyme in CMC,
but would have failed in LMC when as the spelling demonstrates, /'rIbOn/ >> 
/'rIban/ but /di'sOn/ stressed on the second syllable retained /O/.

5. MC had, at least initially, two front or central rounded vowels /y/ and 
/9/. The first was identified with the ME reflex of AS /y/ which retained its 
rounding far longer in the SW than elsewhere, and was written <u>. Cornish 
/9/ had no ME equivalent and was spelled variously <e, u, eu, ue>. According 
to PSRC /9/ unrounded to /E/ around 1525, certainly words with this sound are 
generally spelled with <e> in CW. The present text writes <me(e)r> 
consistently for _meur_ 'great', <ur, vr> for _eur_ 'time', and uses <ue> in 
<suel, bues, dueth>. Possibly the unrounding of /9/ was a rolling process that 
proceded word by word throughout the C16. 

6. An unusual characteristic of BK is the frequent use of the digraph <ue> to 
spell /y/ in words such as _sur, pur, fur, tuz_ (i.e. <suer, puer ...> etc) of 
which there are many examples, especially when these words form the rhymes. 
Perhaps /y/ had become unrounded in the author's English, so that <u> alone 
no longer suggested this sound. The digraph <ue> might then be seen as an 
attempt to signal the fact that this vowel was both rounded <u> and fronted 
<e>. I would not take this to indicate any confusion in speech between /9/ 
( >> /E/) and /y/. The two vowels are contrasted in the rhymes towards the 
end of stanza 237 :

par thennvgh suer    par dhennowgh (?) sur
ny gerth war leer    ny gerdh war leur 
rych os ha fuer      rych os ha fur 
hath gallas mer      ha'th galloez meur 

7. Although MC has been reconstructed with contrasting open and close 'o' 
phonemes /O, o/, when long and stressed as the penultimate syllable of a 
polysyllable they are not distinguished in the texts. Presumably they were 
identified with ME open and close 'o' which are themselves both commonly 
written <o> (later <oa> and <oo>). When short, the close sound was often 
identified with ME short /u/ and written <u> in MC (there is little 
confusion with MC /y/ as this had generally become /I/ when short, and MC 
had at best a very marginal /u/ phoneme) e.g. _arloedh_ /'arloD/ is almost 
invariably written <arluth>. Thus paradoxically, these two phonemes were 
more clearly distinguished when short than when normally long (so called 
'half long vowels'). When extra-long in stressed monosyllables the close 
/o/ might be written <oy>, perhaps recalling its derivation from OC <ui>. 
This was almost regular in PA, and not uncommon in the Ordinalia, although in 
both cases words with /O/ might sometimes be written with <oy> in error.
In addition, in syllables bearing metrical stress (e.g. the final 
syllable of a line) the normally unstressed final /o/ of a polysyllable might 
also be written with <oy>. E.g. <(g)alloys> PA 135.6, 224.6 beside the usual 
<gallus>. This use of <oy> to distinguish /o/ from /O/ is absent from later 
texts. BM uses <oy> indiscriminately to signify any long 'o' sound, writing 
for instance _moz_ 'go' and _doz_ 'come' as <moys, doys> etc., while 
Tregear and CW avoid <oy> altogether in this context. Despite evidence from 
LC texts and from place name studies (perhaps not as clear as some might 
wish for) that two contrasting 'o' phonemes persisted throughout the MC 
period, in could be argued (and of course _has_ been argued at length and 
most vehemently by NJAW) that the texts in fact show that the /o/ vs /O/ 
distinction was neutralised at some time during the course of MC, perhaps 
even before the date of the extent texts, with the <oy> spellings in PA etc. 
being the result of orthographic conservatism. The text of BK however 
rather spoils this viewpoint. As indicated above, the treatment of final 
unstressed /E, O/, the spelling <mirth> for _myrgh_ etc. all indicate a 
date of around 1550. However when long in stressed monosyllables, and 
finally under metrical stress, the close /o/ sound appears to be written 
<oy> more frequently and more consistently than in any of the previously 
known MC texts, PA included. Note for example how _arloedh_ is several 
times rhymed with the English word 'forsooth' (itself suggestive of /o/ 
rather than /O/) and that on many of these occasions the spelling is 
<arloyth> (and sometimes <forsoyth> too! both in stanza 275) a spelling 
hitherto unkown for what is a fairly common word in the corpus. The 
consistent use of <oy> in BK to distinguish /o/ from /O/ is therefore 
powerful evidence that there really were two 'o' sounds throughout MC. 

8. In Welsh the word for 'God' is _Duw_ (/dyw/ in MW). This is an irregular 
development, the expected _*dwyw_ /du:jw/ being restricted to old compounds. 
Interestingly, a handful of words which in W have related forms /-u:jw, 
-u:jv, -o:jw, -o:jv/ etc, have consistently caused problems for Cornish 
lexicographers (who spell them in a variety of ways), while at the same time 
frequently turning up in MC texts as rhymes for 'God'. The simplest 
explanation is that MC had an hitherto unrecognised diphthong _uw_ /yw/ which 
developed regularly from OC /u:jw ~ u:jB ~ u:jM/ after the /o:j/ and /u:j/ 
diphthongs had merged. Words with _uw_ can be identified by the following 
criteria :

*** Their etymology shows that they would have had /u:jw/ etc in OC;
*** In OCV they are spelled <ui(f), uy(u)>;
*** In PA and the Ordinalia they have a characteristic orthographic 
    profile : 50% <u, v>; 40% <ew, ev, (yw, yv)>; 10% <uw, uv>; 
*** They are generally rhymed with _Duw_ and with each other. 

As with some other sounds in MC, /yw/ had no equivalent in ME, and so was 
approximated either with <u> (ignoring the off-glide) or with <ew, yw> 
(ignoring the rounding of the initial element). Nevertheless the 
'correct' spelling is found occasionally eg. RD 1601 <del yv duv pur>. The 
spelling of 'God' tends however to be stereotyped (either <du ~ dv> or 
<dev ~ dew>) in each text. BK exceptionally uses both forms. 

The words thought to have included _uw_ are :
_Duw_ 'God, god' (also _duwow_, duwes_ etc);
_guw, wuw_ 'spear' and it's homophone _guw_ 'woe' (both W _gwayw_);
_gluw_ 'sharp, penetrating' (W _gloyw_ 'bright, shining');
_ruw_ 'king, majesty' and it's homophone _ruw_ 'oar' (both W _rhwyf_)
         [n.b. Lhuyd's "ref" 'oar' seems to be a separate word which 
          probably means 'shovel', W _rhaw(f)_, MC _*reuv_, << /*ra:ma:/, 
          although confusingly this gives the Irish word for 'oar' _ra:mh_]
_suw_ 'suet, tallow' (B _soav_ 'tallow'; W _swyf_ 'froth, scum, yeast');
_pluw_ 'parish' (W _plwyf_);
_g(w)uwenn_ 'sinew';
_duwon_ 'trouble';
_Gwennuwar_ 'Guinevere' cf. W _Gwynhwyfar_.

Clearly the good rhymes for _Duw_ were limited and semantically challenging, 
and while no one seems to have managed to juxterpose God with either suet of 
an oar, all of the other monosyllables are pressed into service in the texts 
(_ruw_ for the first time in BK). 
Note especially RD 2579 ff :

Ow stons fue | crous a pren 
Kyns en myghtern | den ha DEV  /dyw/ 
Yn le basnet | war ow fen 
Curyn a spern | lym ha GLEV   /glyw/ 
Ol ov ysyly | yn ten 
Hag a wel | the lyes PLU      /plyw/ 
Yn golon | dre'n tenewen 
The restye | syngys ow [leg. o] GU  /gyw/ 

In the previously known MC texts _duw_ is rhymed rather rarely (given the 
subject matter). Perfect rhymes with other /-yw/ words account for 13 examples 
(22%), followed by 22 cases of 'acceptable' rhymes with /-y/ (37%) although 
these include a 'clump' of 6 rhymes with _Yesu_ between PC 1631 and 1744. 
The 'obvious' rhyme with /Iw/ accounts for only 19 occurances (32%) mostly 
from BM, and the remaining 9% of cases (5 examples) are a very mixed bunch of 
'bad' rhymes with other -w diphthongs. 

If the text of BK is genuine it would be expected to follow a similar pattern, 
whereas a forger, like a writer of revived Cornish verse, might be expected to 
take the dictionary and mss spellings of _Duw_ at face value and use rhymes 
like _hedhyw_, _gwiw_ (with "Dyw") or _tu_, _tru_, _lu_ (with "Du"). Unless 
he had independently deduced the existance of the /yw/ diphthong (which is 
not of course impossible) he is unlikely to go out of his way looking for 
spears, woe and royalty!

A preliminary search through the initial reading of BK has located 16 rhymes 
involving _Duw_ and two more involving other /yw/ monosyllables. They are 
as follows :

_Duw_ with : 
    _guw_ 'woe'   8  [17, 43, 65, 66, 330, 381, 386, 
                                       and 303 if <vew> is read as _wuw_]; 
    _guw_ 'spear' 5  [212, 279, 349, 228 reading <dew> for <dowt> and 
                                332 reading <caerevew?> as _... re dhuw_];
    _ruw_ 'majesty' 1  [178];
    _klyw_ 'hears'  1  [36];
    <haggew> ???    1  [32]

_guw_ 'woe' with _tru_ 'alas' 1 [145 reading <trv trevath ew> as <trevath trv> 
_guw_ 'spear' with _ruw_ 'majesty'  1  [227]

That is 15 perfect rhymes, one 'acceptable' rhyme, one 'bad' rhyme, and one 
unknown. Note also that _ruw_ which has not previously occurred in a MC text, 
is found three times in BK with one of the predicted spellings <rv> for a /yw/ 
word, and it both cases where it is used as a rhyme it is paired with other 
/yw/ words, _Dyw_ (BK 178) and _guw_ 'spear' (BK 227). Since the idea that 
MC had a /yw/ diphthong has not been published or widely circulated, the 
fact that BK appears to conform rather well to this theory would seem to 
validate both text and theory.

## ------------------------------------------------------------------------ ##

Conclusions 

The spelling of BK suggests a date of c1550, although rhymes and other 
evidence point to at least some of the material having been composed rather 
earlier. The frequent spelling of /y/ in stressed monosyllables as <ue> is 
not seen elsewhere, neither is the spelling of _Duw_ 'God' in more than one 
way in the same ms. The regular and consistent distinction between the 
vowels /o/ and /O/ in stressed monosyllables, by spelling the former <oy> has 
previously been seen only in C15 mss, and then often sporadically. This 
provides additional evidence that the distinction persisted throughout the 
MC period and on into LC. A little known theory that certain words in MC had 
a diphthong _uw_ /yw/ is confirmed by their spellings and rhymes in BK. This 
also provides evidence that the text is probably genuine. 

## ------------------------------------------------------------------------ ##

# Breuzyansow ha ewnansow dhe : 
# Comments and corrections to : 
# bk@carrot.clara.net 
# 9 Mis Me 2003 

# End of File