OE diphthongs/breaking

“Benct Philip Jonsson wrote:

> On 2008-07-21 Tristan McLeay wrote:
>  > Yes indeed. I observed in an earlier email that
>  > in Australian English the phone corresponding to
>  > RP /au/ is very similar to the OE vowel
>  > i.e.a backing diphthong of which both segments
>  >     are low.
> 
> Also some dialects have a /ai/ > /aM/ shift!
> 
> I don't want to spoil the fun or anything, but
> this might be 'evidence' that the OE vowel
> resulting from Germanic *au and written _ea_ was
> actually a diphthong and not a back monophthong.
> In the diphthongist--back-monophthongist debate I
> am a compromissist in that I believe the so-
> called 'long diphthongs' were actual diphtongs
> _ea, eo, io/ie_ /&@/, /EV/ or /e7/ -- actually of”
 

My supposition is that the OE diphthongs were actual diphthongs, not
monophthongs, though widespread monophthongisation occurred in the later OE
period.

 

“(I thought "io" was a variant of "eo", not of "ie", which is the i-mutation
of "ea" and "eo", no?)”

 

<io> can be:

-          West Saxon, later <eo>, = (1) Germanic *I by breaking, as
<liornian> “learn”, <miox> “manure”; = (2) Germanic *I by u- and a/o-umlaut,
as <cliopung> “calling”, <mioluc> “milk”, <liofast> “thou livest”;

-          Anglian and Kentish by u-umlaut = West Saxon <i> (before single
consonants except labials and liquids), as <liomu> “limbs”, <nioþor>
“lower”, <siodu> “custom”, <sionu> “sinew”, cf. West Saxon <limu, niþor,
sidu, sinu>;

-          Anglian and Kentish by o/a-umlaut = West Saxon <i>: <nioman>
“take”, <niomaþ> “they take”, cf. West Saxon <niman, nimaþ>;

-          Northumbrian and Kentish = West Saxon <ie>, i-umlaut of <io>,
Germanic *I, as <hiorde> “shepherd”, <iorre> “angry”, cf. West Saxon
<hierde, ierre>;

-          Kentish = West Saxon <eo> by breaking before r +consonant, as
<hiorte> “heart”, <iorþe> “earth”, <stiorre> “star”, cf. West Saxon <heorte,
eorþe, steorre>; 

-          Kentish by o/a-umlaut = Anglian and West Saxon <i> before velar
consoants, as <stiocian> “prick”, cf. Anglian and West Saxon <stician>;

-          Mercian beside <eo> = West Saxon <ie>, i-umlaut of <io>, as
<iorre, eorre> “angry”, <heorde, heorde> “shepherd”, cf. West Saxon <iorre,
hierde>; 

 

<îo> can be:

-          West Saxon, later <êo>, = (1) Germanic *iu, as <geþîode, geþêode>
“language”, <lîode, lêode> “people”, <þîostre, þêostre> “dark”; = (2)
Germanic *I or *ij + back vowel, as <fîond, fêond> “fiend, enemy”, <frîond,
frêond> “friend”, <þrîo, þrêo> (neuter plural) < *þriju “three”; = (3)
Germanic *î by breaking before *h and *ht, as <þîon, þêon> (Old High German
<dîhan>) “thrive”, <lîoh, lêoh> (Old High German <lîh>) “lend (2nd person
imperative singular)”;  

-          Kentish, beside <îa>, = (1) Mercian and West Saxon <êo>, Germanic
*eu, as <dîop, dîap> “deep”, <dîor> “deer”, <lîof> “dear”, cf. Mercian and
West Saxon <dêop, dêor, lêof>; = (2) West Saxon <îe>, i-imlaut of <îo>,
Germanic *iu, as <dîore, dîare> “dear”, <lîohtan> “give light”, cf. West
Saxon <dîere, lîehtan>;

-          Mercian, beside <êo>, = West Saxon <îe>, i-umlaut of <îo>, as
<dîore, dêore> “deer”, <lîohtan, lêohtan> “give light”, West Saxon <dîere,
lîehtan>;  

 

 

> mean mid height of course! --, /iM/ while the so-
> called 'short diphthongs' were short back or
> central unrounded monophthongs /3/--/6/, /V/--
> /7/, /i\/--/M/.
 

“I probably don't know as much as you on this area, but having three low
short unrounded vowels (i.e. /&/=ae /6/=ea /A/=a) seems very difficult and
unlikely. I'm not aware of any language that distinguishes more than two
short low vowels of the same rounding. (Forgive me, I don't know how to type
ash on this Windows keyboard.)”

 

I agree. I’m quite traditionalist in that I believe that a back /U/-like
glide was inserted between the original short vowel and the breaking
environment. These new “short” (syllabically light) diphthongs */{U/ and
*/eU/, as well as the old (syllabically heavy) diphthongs */{:U/ (+ the
reflex of “broken” */{:/ > */{:U/), and */i:U/ (+ the reflex of “broken”
*/i:/ > */i:U/) underwent a development that Roger Lass calls “Diphthong
Height Harmony”, i.e. the second element of the diphthong is brought to the
same hight as the dominant first element. (Examples from Lass 1994, 1997):

 

West Germanic examples: *ald, *elx, *baum, *na:x, *biudan, *li:xt;

 

After OE breaking: *{Uld, *eUlx, *b{:Um, *n{:Ux, *bi:UdAn, *li:Uxt;

 

Diphthong Hight Harmony: *{Ald, *eolx, *b{:Am, *n{:Ax, *be:odAn, *le:oxt; 

 

“And aside from when "ea", "eo" represent palatalisation before /A, o, u/,
ISTR that short "ea"/"eo" only actually come from breaking ... if that's
right, it seems to me that the diphthong reading is the simplest and best.
Eventually the vowels which result from breaking (almost always) merge back
in with the original unbroken vowel, which I think makes it even more likely
they're merely short diphthongs.

 

Can short "ea", "eo" ever contrast with short "ae, e"?”

 

Not to my knowledge. They seem to have started out as allophones, but
subsequent developments have lead to partial phonemic re-distribution.   

 

“As for "ie" it is a sound which doesn't make much sense. It merges with y
eventually --- at least in the West Saxon standard --- yet comes from
unrounded vowels + i-mutation. I have seen three --- now four different
readings for it (/i(:)e, i(:)y, I(:), (i:)M/). None had seemed to

account for all the data --- how does the decidedly front unrounded /ie/
merge with /y/ (but not /i/)? how does i-mutation introduce rounding? (If I
squint *just*so* I can see how /I(:)/ might merge with /y(:)/ but not
/i(:)/, but I find it difficult to believe any language would

distinguish all four of /i: i I: I/. Once again the only language I'm aware
of that comes close is my dialect, although there's only three phonemes
there ~[I:(@) Ii I]. Icelandic is ruled out on the grounds that the four
vowels [i: i I: I] only make a two-way contrast /i I/ which permits a
greater range of tactics to distinguish the two phonemes.)”

 

Well, following Lass I would say that since i-umlaut occurred after breaking
the – i-umlauted reflex of */iu/ was */iy/, spelt <ie>. The assimilation of
*/iy/ > /y/ in later West Saxon is not difficult to explain.  

 

“In short I have no idea what to consider "ie". Your notion is somewhat
tempting --- as long as breaking is an ongoing change caused by the phonetic
characteristics of the consonants, then [&A e7] -> [iM] -> [M] is likely
enough, and then [M] and [y] sound pretty similar even if they're pretty
different in how they're articulated. I just find that a bit harder to
swallow for the long diphthongs.”

 

As do I. 

 

> That the OE writing system could use the same
> symbols for both should not be surprising: they
> were similar if not identical and could be
> construed as long--short pairs, and most
> importantly breaking of long vowels **had**
> probably resulted in just these diphthongal
> qualities whiöle breaking of short vowels had
> resulted in these short monophthongs. The *au >
> /&@/ shift just increased the incidence of that
> diphthong. 
 

I have my problems with the schwa-glide. Early OE is such an
“unschwa”-language, there are so many developments that can only be easily
explained by vowel harmony between the stressed and the unstressed
syllables. Even the weaker second elements of the diphthongs harmonise with
the vowels and consonants in the following unstressed syllables. This, to
me, makes schwa-diphthongs unlikely in early Old English. I don’t mean to
rule out, though, that schwa-diphthongs in later Old English may have been a
transitional development towards the late Old English and early Middle
English monophthongs.

Dan  

 

 

 

-----Original Message-----
From: Tristan McLeay
Sent: Tuesday, July 22, 2008 2:59 AM
To: CONLANG@listserv.brown.edu
Subject: OE diphthongs/breaking 

 

Benct Philip Jonsson wrote:

 

> On 2008-07-21 Tristan McLeay wrote:
>  > Yes indeed. I observed in an earlier email that
>  > in Australian English the phone corresponding to
>  > RP /au/ is very similar to the OE vowel
>  > i.e.a backing diphthong of which both segments
>  >     are low.
> 
> Also some dialects have a /ai/ > /aM/ shift!
> 
> I don't want to spoil the fun or anything, but
> this might be 'evidence' that the OE vowel
> resulting from Germanic *au and written _ea_ was
> actually a diphthong and not a back monophthong.
> In the diphthongist--back-monophthongist debate I
> am a compromissist in that I believe the so-
> called 'long diphthongs' were actual diphtongs
> _ea, eo, io/ie_ /&@/, /EV/ or /e7/ -- actually of
 

(I thought "io" was a variant of "eo", not of "ie", which is the i-mutation
of "ea" and "eo", no?)

 

> mean mid height of course! --, /iM/ while the so-
> called 'short diphthongs' were short back or
> central unrounded monophthongs /3/--/6/, /V/--
> /7/, /i\/--/M/.
 

I probably don't know as much as you on this area, but having three low
short unrounded vowels (i.e. /&/=ae /6/=ea /A/=a) seems very difficult and
unlikely. I'm not aware of any language that distinguishes more than two
short low vowels of the same rounding. (Forgive me, I don't know how to type
ash on this Windows keyboard.)

 

And aside from when "ea", "eo" represent palatalisation before /A, o, u/,
ISTR that short "ea"/"eo" only actually actually come from breaking ... if
that's right, it seems to me that the diphthong reading is the simplest and
best. Eventually the vowels which result from breaking (almost always) merge
back in with the original unbroken vowel, which I think makes it even more
likely they're merely short diphthongs.

 

Can short "ea", "eo" ever contrast with short "ae, e"?

 

As for "ie" it is a sound which doesn't make much sense. It merges with y
eventually --- at least in the West Saxon standard --- yet comes from
unrounded vowels + i-mutation. I have seen three --- now four different
readings for it (/i(:)e, i(:)y, I(:), (i:)M/). None had seemed to

account for all the data --- how does the decidedly front unrounded /ie/
merge with /y/ (but not /i/)? how does i-mutation introduce rounding? (If I
squint *just*so* I can see how /I(:)/ might merge with /y(:)/ but not
/i(:)/, but I find it difficult to believe any language would

distinguish all four of /i: i I: I/. Once again the only language I'm aware
of that comes close is my dialect, although there's only three phonemes
there ~[I:(@) Ii I]. Icelandic is ruled out on the grounds that the four
vowels [i: i I: I] only make a two-way contrast /i I/ which permits a
greater range of tactics to distinguish the two phonemes.)

 

In short I have no idea what to consider "ie". Your notion is somewhat
tempting --- as long as breaking is an ongoing change caused by the phonetic
characteristics of the consonants, then [&A e7] -> [iM] -> [M] is likely
enough, and then [M] and [y] sound pretty similar even if they're pretty
different in how they're articulated. I just find that a bit harder to
swallow for the long diphthongs.

 

> That the OE writing system could use the same
> symbols for both should not be surprising: they
> were similar if not identical and could be
> construed as long--short pairs, and most
> importantly breaking of long vowels **had**
> probably resulted in just these diphthongal
> qualities whiöle breaking of short vowels had
> resulted in these short monophthongs. The *au >
> /&@/ shift just increased the incidence of that
> diphthong. 
 

FWIW although I don't think any dictionary will agree with me, pairs like
"vowel" and "Val" or "Powell" and "pal" are homophones or nearly so in these
parts.

 

>Contrary to belief the OE writing
> system was by no means 'perfect' or 'one to one':
 

Hence one reason I like it so much =)

--

Tristan.