LL-L "Orthography" 2002.03.14 (04) [E/S]

Thu Mar 14 21:55:07 UTC 2002

======================================================================
 L O W L A N D S - L * 14.MAR.2002 (04) * ISSN 189-5582 * LCSN 96-4226
 Web Site: <http://www.geocities.com/sassisch/rhahn/lowlands/>
 Rules: <http://www.geocities.com/sassisch/rhahn/lowlands/rules.html>
 Posting Address: <lowlands-l at listserv.linguistlist.org>
 Server Manual: <http://www.lsoft.com/manuals/1.8c/userindex.html>
 Archive: <http://listserv.linguistlist.org/archives/lowlands-l.html>
=======================================================================
 A=Afrikaans Ap=Appalachian D=Dutch E=English F=Frisian L=Limburgish
 LS=Low Saxon (Low German) S=Scots Sh=Shetlandic Z=Zeelandic (Zeeuws)
=======================================================================

From: "John M. Tait" <jmtait at wirhoose.co.uk>
Subject: LL-L "Orthography" 2002.02.28 (07) [E]

Andy Eagle wrote:

> Many of the 'traditional' practices, similar to english they may
> be, are not nonsensical. They are being used nonsensically i.e. in an
> unpredictable pick 'n' mix kind of way.

I don't agree with this - I'd say traditional practices
are actually nonsensical. While you might defend English
orthography to a great extent, Scots orthography _is_
pick'n'mix, and no alternatives have been proposed that
anybody actually uses. Even my (or yours, John's or Ian's)
spellings are a mixture of Scots and English, the guiding
principle being that English spellings are used if they
don't conflict with Scots pronunciation - ignoring the
fact that they do conflict with Scots spellings.

And:

Andy wrote:

> What are the 'Scots' spellings? I would argue, for example

By this I just meant any non-English spellings used in one's
Scots orthography. It doesn't really matter whether it's a
traditionally Scots spelling or something you've devised
yourself.

A simple example of this is that you could dramatically
simplify your orthography by introducing a rule that <y>
should always be used in Scots for /@i/ and /a:I/ (whether
a length indicator should be added is a separate issue).

However, any such simplifying principles in a Scots
orthography are generally overridden when they conflict
with English spellings, for the sake of recognisability.
This makes the orthography so irregular it can't really
be taught as a system.

I do feel an awful lot could be acheived very quickly in
Scots orthography if the "recognisability" criterion could
be dropped.>

(End of Sandy's postings)

It's perfectly possible to see this, not as a conflict between 'English'
and 'Scots' spellings, but as a basically English-type spelling system
augmented by some features to adapt it for Scots. The whole thing then
becomes a _Scots_ system. There is no conflict between an English and
Scots
spelling system insofar as they both use the same conventions. The
problems
arise (a) where the connotations of English pronunciation are intrusive,
a
problem which could probably be addressed by the way the system is
adapted
in the first place (for example, by using <oo> rather than <ou>) and (b)
where the final stages required to make the orthography consistent -
within
its own presuppositions - are perceived as being unnecessary, and
therefore
rejected (e.g. <dui> rather than <dae>.) Also (c) sometimes the required
consistency measures are perceived as being in conflict, not with
English,
but with other traditions of Scots spelling.

Examples of each kind:

(a) The connotation of <ea> as equating to an [i] pronunciation destroys
its value as a diaphonological (or polyphonemic) representation for BEAT
class words. This could potentially be addressed by spelling the
diaphoneme
as <ae> or <ei>.

(b) The diaphonological spellings of <tui> (adv = English _too_), <dui>,
<shui> are seen as unnecessary because of (c) the lack of a tradition of
final <ui> in Scots spelling.

It would be quite possible to create an orthography which simply
systematised these elements, while using English-type conventions of
e.g.
consonant doubling. (Incidentally, your method of using no doubled
consonants is appropriate for any _radical_ orthography for Scots, in
contrast to other systems, such as the RWS, which don't seem to consider
this.) It would not necessarily have a 1->1 phoneme to grapheme
correspondence - for example, <ch> could represent [tS] when initial and
[x] when medial and final (The way we read, psychologically, means that
we
easily make such distinctions) - but this would not mean that it would
be
an English rather than a Scots system. If such features were applied
systematically, it would be a thoroughly Scots system within its own
presuppositions, English-like or not.

These presuppositions could include:

- The etymological spellings of Latinate and Greek (but not illogical
Anglic - e.g. _tongue_, _steak_) words

- The acceptance of more than one grapheme to one phoneme or polyphoneme
(but not more than one phoneme to one grapheme), so <ei> and <ae> for
/e+/

The results would not be pick-n-mix, because each principle would be
used
in clearly defined ways, and therefore the orthography as a whole would
be
tailored to Scots. It is true that the changes to traditional spelling
required to implement this meet with vehement resistance, but this
doesn't
make the approach nonsensical - not unless you regard anything which
doesn't follow a strict 1->1 phonological approach as nonsensical. It
only
means that normal practice is rendered inconsistent by resistance to the
measures which would make it a Scots, rather than a half-adapted
English,
orthography.

Comments on a couple of the above points from Sandy, which I'll repeat
here:
<<
A simple example of this is that you could dramatically
simplify your orthography by introducing a rule that <y>
should always be used in Scots for /@i/ and /a:I/ (whether
a length indicator should be added is a separate issue).

However, any such simplifying principles in a Scots
orthography are generally overridden when they conflict
with English spellings, for the sake of recognisability.
This makes the orthography so irregular it can't really
be taught as a system.
>>
This assumes that it's important for orthography to be taught as a
system.
I would argue that we learn to write mostly by reading, and if you read
_time_ and _synd_ then you tend to reproduce them as such. It is more
important to accommodate the reader than the writer, and the important
thing for the reader is to know is that both /-y-/ and /-i-e/ are
pronounced as [@I] or [aI] (depending on the SVLR.) This is why <mind>
is a
bad spelling, because there is no indication that the <i> is not
pronounced
as <I.>. You can argue that this doesn't matter, as native Scots
speakers
will know how to pronounce words shared with SSE (though this argument
could justify any illogicality in spelling.) But this isn't viable in a
situation where Scots is often read, and written, by people who don't
speak
it.

This type of approach is of course messy compared to a 1->1 system, but
it
can be made regular within its own presuppositions. So <y>, and with
following <e> after one or more consonants, both have a defined
pronunciation, assuming that the long and short pronunciations are
allophones except finally, where they are normally distinguished as
<-y(e)>
and <ey>. What should be done away with is spellings such as _mind_,
_wind_, where there is no way to tell how the vowels are pronounced, and
spellings like <ay> v. <aye>, which conflict with other spellings such
as
<kye> and <gey> and should be spelled <aye> and <ey>.

Again, the problems come not from the intrinsic characteristics of the
spelling, so much as from resistance to spelling changes such as <mynd>
rather than <mind> and <ey> rather than <aye>. But if there is such
resistance to this, I can't see why there wouldn't be much more
resistance
to a more radical approach. At any rate, if you can defend English
orthography, and if Scots orthography can be made more regular than
English
- as it can, by abolishing one grapheme to more than one phoneme
correspondences, such as <steak>, <meat>, <bread> (suggested for Scots:
<staik>, <maet>, <breid>) - then Scots spelling would become more, not
less, regular than English. This would involve, of course, the
re-spelling
of words (like <steak>) which are pronounced the same in English as in
SSE,
but which do not suggest the pronunciation, and of traditional spellings
such as <aye> for the word which means English _always_.

Not that there aren't practical problems. The RWS approach, for example,
retains such traditional spellings as _hure_ (no problem with filters if
you're using words in languages the filters don't know!) in spite of the
fact that the u-e spelling is very unusual in Scots (I can't think of
another word off hand, though somebody will probably produce dozens) and
analogical English words (shared or not) have a [ju] rather than an [u]
sound in SSE - e.g. _cure_, _during_ (I can't hear 'during' as a Scots
word
- 'I was a teacher during the war' - 'I wis a dominie _throu_ the war'.)
A
sensible spelling would be 'hour' - but this obviously conflicts with an
English word, and the double <oo> in the popular spelling 'hoor' would
be
anathema to the RWS way of thought. It is obviously because of
considerations like these that <hure> has 'stuck', in spite of its
irregularity. It's a moot point whether this - considering how few words
there probably are with this problem - is a big consideration or not. A
radical orthography can avoid all this mess, of course. (Personally, I
think I'd go for <hoor>).

Some of the problem is knowing what spelling is for. My own $64,000
question is: if I were to teach Scots to English speakers, would I be
able
to give a pronunciation guide which would enable the phonology of Scots
words to be derived from the spelling? I think this could be done by the
sort of changes to traditional spelling which I've suggested above.

For example,

AE and EI

<ae> and <ei> are pronounced [i] in some dialects and [e] in others, but
final <ae> is always pronounced [e].

maet
heid
brae

Y, I-E, -YE and EY

<y> and <i-e> are pronounced [@I] when short and [aI] when long (SVLR
would
need separate explanation), but final <-y> is pronounced [i] or [I] in
polysyllabic words.

time
mynd
bonny

final <ye>, and <y> in monosyllabic words, are pronounced [aI].

kye
try

<ey> (usually final) is pronounced [@I], or [EI] in some dialects.

gey
fey

The above is, I think, by far the most complicated of the explanations
which would be required (more changes could, of course, simplify it
further, e.g. <ie> or could be used for final unstressed [i]/[I],
eg:
bonie/boni - or boanie/boani). The guide to the pronunciations of other
vowel spellings would be much simpler.

As I say, it all depends on your criteria. My first criterion is: can I
write Scots in such a way that a speaker of SSE can read it without
requiring individual explanations of the pronunciation (stress, perhaps,
apart) of individual words which he/she might not know, and can I
explain
this reasonably concisely in a pronunciation guide? This criterion is
_not_
satisfied by any traditional system in use, but could be without too
much
surgery. Secondly, and less importantly from a learner's and reader's
point
of view: are such words as are shared with English, excepting those
borrowed from non-English languages, spelt in such a way that they, too,
conform to the same set of pronunciation rules? This would require more
radical changes - e.g. <tongue> to <tung>, <ever> to <ivver>, etc. The
problem with most systems in use - including the recommendations of the
spelling committee - is that this second criterion is applied to some
words
but not to some others, according to individual preferences (what I call
voting for a three-humped camel). So, instead of <ever>, which at least
suggests the SSE pronunciation, or <nivver>, where the double <vv> makes
the pronunciation explicit, the spelling report chooses to adopt the
English rule against doubling of <v>, and recommends <niver>, which is
more
ambiguous than either of the others.

Another problem with existing systems is that most people who are
involved
in them accept almost as if it were an axiom the idea that words that
sound
the same but have different meanings should be spelt differently - this
automatically creating spellings, such as _redd_ , which break normal
rules.

It's possible that a traditionalisation of a radical orthography and
radicalisation of a traditional orthography could meet in the middle. It
is
difficult to radicalise a traditional orthography simply because of its
familiarity, whereas if you begin from the radical end you are moving
from
the already unfamiliar, and thus are more aware of the underlying
phonological issues and less hampered by habit. However, you perhaps
then
become familiar with the unfamiliar to the extent where you reject
familiarising measures which would be essential to fulfil the criterion
of
readability for those literate in English - such as etymological
spellings
of Latinate and Greek words - which Andy and I would consider not so
much
important as inescapable. It would depend what your aims and objectives
were.

John M. Tait.

----------

From: "John M. Tait" <jmtait at wirhoose.co.uk>
Subject: LL-L "Orthography" 2002.03.10 (03) [E/S]

Sandy wrote,

>A'v been daein plenty but A see you'v no chainged!

Heh - I dout no! Parteecularly, I hinna lairnt ti keep ma big mooth
steekit...
>

>Could you complete the argument? Why is there little point
>in pursuing something because literary and academic figures
>reject it?

Well, I suppose I'm only speaking from my own perspective, where the
practical purpose of any orthography is that it be accepted widely
enough
to assist literacy in the language. This is a tall order in any case as
far
as Scots is concerned, as it would require some sort of official backing
and/or popular acceptance within a wide enough section of the
population.
Both these cases are a tall order in the case of Scots, but it seems to
me
that the more radical the orthography the less chance it has of
acceptance.
At the popular level, a radical orthography (any orthography, but
especially a radical one) would be likely to gain acceptance only if
deliberately taught, which brings us back to the necessity for official
backing. In Scotland, there is no official backing for any attempt to
teach
Scots at any level other than that provided within the national
curriculum,
which is essentially that of encouraging the use of pupils' native form
of
English - an aspiration which is honoured mostly in the breach, and even
where there is some attempt to implement it, is interpreted mostly in
terms
of encouraging children (usually on special occasions such as Burns Day
or
when one of the one or two specialists in Scotland is present) to
produce
their own 'native dialect of English' off the cuff. This approach - of
eliciting whatever it is that children happen to speak - is a good
enough
idea as far as it goes, but is in my opinion a different thing from
teaching Scots as a language. However, it fits in well with the current
attitudes in academia - stemming ultimately, I suspect, from Labov and
popularised in this country mainly, I think, by Trudgill - with the
result
that any implementation of 'Scots teaching' in school, including
crucially
the preparation of materials, is likely to be dominated by academic
linguists trained in this school of thought. One of the aims of this
approach in the national curriculum is explicitely stated to be that
exposure to non-standard language teaches children the value of a
standard
language - i.e, standard English, implying that Scots - or, the 'native
form of English' - is to be taught explicitely as an example of
non-standard dialect to be contrasted with standard English. This, too,
fits in well with the academic position, where Scots is 'taught' by
people
who regard it as intrinsically non-standard, its value lying in its
non-standardness, and who themselves use only English for practical
purposes. The only possible outcome of such an approach is to enhance
the
status of standard English - an outcome which the powers that be might
be
persuaded to support. As, therefore, the acceptance of an orthography
depends on official backing, and not only officialdom but those who
ostensibly represent Scots to officialdom are opposed to orthography in
general and to radical orthography in particular, the Faroese or Catalan
scenario which involves an orthography being accepted in official
circles
by people who actually want the language to be used is not forthcoming.
As
far as acceptance only by those who use Scots is concerned, again, much
smaller areas of agreement are practically impossible to achieve.

This brings us back to what I said before ( and, probably, before, and
before...!) The only way to get an orthography accepted would be if it
were
to be agreed upon by a substantial number of those who represent Scots,
so
that most materials written in Scots would use it. That this is not
likely
is demonstrated by the Spelling Comatee - not only the fact that its
conclusions were not accepted, but more importantly that the most
influential figures either did not bother to turn up, or disassociated
themselves with the final result. As another illustration, a recent Arts
Council grant was given to a group to produce material in Scots for
schools, the three prominent names in the group being a Scots author
whom I
remember being vehemently against standardised spelling on the old
Scotslang group (I can't quote what he said about it because of the
filters); a writer and editor whose opinions on Scots as intrinsically
unrespectable I think I quoted recently; and the author of the Grammar
Broonie, which does not know the difference, for example, between 'ae'
and
'ane', and is written in a form of Scots which breaks many of the rules
which it itself recommends. Again, we have Scots being taught
essentially
as dialect. In such a situation, it seems to me that the only thing to
do -
not because it is the best idea, but because it is the only possibility
-
is to use a more or less traditional orthography, but develop it along
diaphonological lines where possible.

This raises another question - as to how far it is possible to express
the
essential characteristics of Scots within a more conventional
orthography.
To some extent this depends what is meant by 'essential'. However, I'm
going to banish this train of thought to a separate posting.

The deriders of orthography are of course inconsistent in that, while
being
opposed to much less radical orthographies than Sandy's because of their
supposed unreadability (based on their difference from English spelling)
they lionise Tom Leonard, whose orthography is deliberately non-English.
The difference, of course, is that Leonard's spelling is designed to
suggest illiteracy, which falls within the area of academic sympathy and
supports the underlying ideology that Scots is to be used as an example
of
a non standard (and, by implication, unrespectable) mode of language,
with
the implied demonstration of the advantages of standard English.

There are, of course, other points in developing a radical orthography -
such as that it casts light on aspects of the language, and of
orthographic
problems in particular. It all depends what you mean by 'point'. As an
auxiliary orthography  alongside written texts - as on Scotstext (I can
never remember which letter to capitalise - could lead to a long
discussion
on reading psychology) it has an obvious use, as Ron points out. It's
notable, however, that many dictionaries which used to use re-spelling
systems (such as the OED) now use IPA, and I wonder whether anyone
interested enough in linguistics wouldn't just as well use IPA or SAMPA
-
those not interested probably wouldn't bother with the alternative
orthography either. The biggest advantage of an alternative orthography,
however, is that it is much easier to read a connected text in it than
in
IPA or SAMPA.

Another point is that it forces you to decide between certain
representations which can be fudged in traditional spelling - again,
mostly
the <ea> spelling. As an example, here's the first part of Tam o
Shanter,
with things that I would alter.

Qan capman biliz lëvz ði stríts,
An drúÞi níbirz níbirz míts,
Az merkit déz iz wërin lét,
An fók biginz ti tak ði gét,
Wí Þiñksnæ on ði lañ Skots meilz
Ði mosiz, watirs, slaps an steilz,
Ðat laiz atwín us an úr hém,
Qår sits úr sulki, sulin dém,
Geðirin hir brúz leik geðirin stoorm,
Nursin hir raÞ ti kíp it warm.

You can see the alterations I’ve made - two íí -> ë changes (The English
<ea> spelling is a hint, and cp. Shetlandic [le:v] and [we:r]; I’m not
representing vowel length, assuming that it’s SVLR conditioned) and one
ó
[o] for o [O] - though in fact, I say either [fOk] (Shetlandic) or
[fVuk]
(North East). (Incidentally, _bonny_ is [bonI] even in dialects which
distinguish /O/ and /o/, so this would be <bóni>.) Also <a> instead of

in e.g. <atwín>, as you already have in <an> and <az>. And <sulen> to
<sulin>, which I presume was an oversight? I could have made others,
such
as proposing a new single representation - such as <ey> - instead of
<ai>
and <ei>.

(I'm tryin ti think on a Scots word wi the soond at ye spell <oi>. In
Shetlandic the'r _twa_ sic soonds, /2i/ in e.g. _point_, an /Oi/ in e.g.
_gloy_, roch strae; but i the NE, as faur's I can think, aa sic words
like
_point_ is sayed like 'pynt'.)

It's a matter of opinion whether this focussing on phonology is
ultimately
an advantage, or whether it would just lead to more disagreement among
people who might otherwise agree to fudge. After all, my discussion with
Andy about <ea> is similar, in that I would use <ae> and/or <ei> for
_all_
BEAT words and hopefully dispense with the <ea> spelling - with its
underlying connotation of an SSE pronunciation - altogether, but if even
this amount of radicalism seems OTT to Andy, you can imagine what the
SNDA
and their ilk would say about it. One of the things I discovered on the
spelling committee was that even _tiny_ concessions to a phonological or
diaphonological representation which produce unusual-looking forms (e.g.
<tui> and <dui> rather than <tae>, <dae>) are regarded as _obviously_
unacceptable and unnecessary even by those who are _in favour_ of an
agreed
orthography. How much more so by those who aren't, and who simply don't
see
why anything should be spelt any way in particular. I personally am
inclined to focus upon phonological representation, but to try to
incorporate this within an orthography which gives a traditional
appearance. Again, this raises the question of how far this is possible,
which I've banished to another posting.

Another thing is that, when you start inventing things from scratch,
everyone has their own ideas about it. I might start to invent an
alternative orthography and come up with something unlike yours, or at
any
rate, as different from yours as my 'normal' spelling is different from
your 'normal' spelling, even if I were to start from your basic
proposals.
Again, there is the problem of getting agreement even between a few
people.
At least, with 'normal' spelling, there is some agreement - for example,
over the use of <th> for both /D/ and /T/. The use of <dh> or _edh_,
however phonologically sound, is introducing another element for people
to
go their separate ways over. I still maintain that more phonological
information is required in the Scots vowel system, but I don't think the
lack of a distinction between /D/ and /T/ causes any practical problems
(initially, anyway, though it might with words such as _kythe_, where
there's an SVLR issue.)
>
>> _Per se_, the type of orthographic approach Sandy is using reminds me of
>> the one Sir James Murray used in his Biblical translations, the purpose
>> of
>> which was to put over the pronunciation of his own border dialect.
>> Samples
>> are available in the (now also out of print) _History of the Scots
>> Bible_,
>> by Graham Tulloch.
>
>It's not the same. Murray was a professional lexicographer for
>the English language who believed that no Scots writing was of
>value unless it provided a precise snapshot of the spoken language
>at a specific place and time. My approach is that there doesn't have
>to be an exact representation - it just has to be teachable and as
>consistent as it can be made (not entirely consistent - dialect
>variation makes that impossible).

There is a similarity, however, in that both are based upon a
phonetic/phonological approach, and that both emphasise pronunciation
rather than holding with historical spellings. At any rate, they're more
similar to each other than either is to traditional spelling. My script
for
showing Shetlandic pronunciation on my website is also similar in some
ways
- but I was intending to change that to SAMPA! (Ower thrang evenou
threipin
aboot spellin...mair excuses!)
>
>
>> A few comments (apologies if my cursory glance has missed some of the
>> adaptations from the original proposal):
>>
>> 1. The original proposal seems to have some confusion of phonetics,
>> phonology and morphology. The omission of the /O/ phoneme, for example,
>> is
>> another of those features which tend to standardise in favour of Central
>> dialects. If /O/ is a phoneme in some dialects, I don't see how it can
>
>I did mention that - I was really hoping for solutions from
>those better-versed than myself, not problems!

I hope it's no me at ye think is better versed nor yersel!

My solution would simply be to include the /O/ phoneme! Let's say, <o>
for
this, and o-acute or <oa> for the /o(:)/ phoneme. I'd have to look a bit
closer at your proposals to figure out how far this is involved with the
idea of length - ie, it would be possible to characterise the difference
as
length rather than quality, as long as SVLR length wasn't represented.
In
the original proposal, where non-phonemic SVLR length was represented,
length wouldn't have been enough to distinguish /O/ and /o/.
>
>
>> be derived from. On the other hand, is it necessary to distinguish
>> between
>> final plural [s] and [z], when they both express the plural morpheme?
>> Similarly, the short and long pronunciations of the phoneme in _time_,
>> _byre_ is usually determined by the SVLR, and where it isn't is subject
>> to
>> dialect variation. Both of these considerations suggest that it is best
>> to
>> use one grapheme for it, because native speakers will know how it is
>
>You say "similarly" but in fact these two points contradict
>each other. If you don't distinguish the plural /s/ and /z/,
>then the SVLR isn't a reliable guide any more. It's not a
>completely reliable guide anyway, of course!

Hmm. Maybe I'm labouring under a misconception here, but I sort of
assumed
that, since all vowels (ie, vowels which can be long) are long before a
morpheme boundary, then they must all be long at the end of a word, and
the
pronunciation of the plural morpheme will therefore be determined by
what
comes before rather than the other way round. I could be completely
wrong,
of course, as I haven't looked into it. Can you elaborate?
>
>> pronounced in their own dialect, whereas a distinction is redundant
>> where
>> it is determined, and intrusive where there is dialect variation. Surely
>> if
>> we'r talking about diaphonology, we first need to implement a
>> phonological
>> rather than phonetic approach. At any rate, it should be clear which
>> elements are phonetic, which phonological, which diaphonological, and
>> which
>> morphological, and which of these principles is used where.
>
>I don't agree that all those aspects of linguistics need to be
>given weight in an orthography. The important thing is having
>a representation of each of the phonemes in the language and
>applying them consistently enough for people to be able to see
>the sense in them.

I didn't say that they all had to be given weight - obviously, an
orthography can be based on different principles, or a combination. What
I
meant was that, in inventing an orthography, you have to be clear which
principles you're using where. If you're emphasising phonology, for
example, then you have to begin with a clear idea of the inventory of
phonemes. (Probably here I'm wittering on about generalities not
directly
related to your own proposals!)
>
>
>> As regards plurals, I would suggest that there's a back-influence from
>> English orthography. Because English uses <s> for /z/ (the phoneme)
>> sometimes, as well as <s> for [z] where the difference isn't
>> phonological,
>> we may have a tendency to change _all_ <s>s to <z>s even where they both
>> represent the same morpheme, rather than just where the /z/ is a
>> phoneme.
>
>I think you're being far too pessimistic about people's,
>especially children's, linguistic abilities. Besides, you
>said "may" - you haven't put it into practice, so unless
>you can find some examples, there seems no reason to cite
>this as a problem.

The mental process I'm guessing at is more or less as follows:

1. English often uses <s> for /z/, for example, <rose> /ro:z/

2. Conscious of the problem this might cause with the pronunciation of
Scots words, in a radical orthography we tend (quite rightly) to propose
<z> for /z/, so, for example, <uiz> /2:z/, <jilouz> /dZ@"lu:z/ rather
than
<uise>, <jalouse>.

3. Having corrected this inconsistency in English spelling where it is
likely to cause problems with the pronunciation of unknown words, we
then
go on to change <s> to <z> in situations, like the plural morpheme,
where
the pronunciation is - as far as I'm aware - determined by what comes
before, and thus not a practical problem. In fact, if we hadn't had our
attention drawn to plural [z], we probably wouldn't notice that it was
different from plural [s].

What I'm saying is, then, that the spelling of plurals as <z> and <s>,
rather than just using one of these - and assuming that the
pronunciation
is predictable - is unjustified representation of unimportant voicing
spilling over from justified representation of important voicing.

As for pessimism, I've got a big savings account of it, and I'm getting
interest on it all the time! However, it's not children's linguistic
abilities I'm pessimistic about - it's adults' attitudes, much the same
as
in your illustration about Colson (sp?) numbers.
>
>> You could argue, I suppose, that there's no reason why the plural
>> morpheme
>> shouldn't be written phonetically, but I would suggest that this would
>> create problems, ultimately involving rules for writing it (a bit like
>> we
>> have to do with the preterite morpheme <-it>, <t>, <ed> because the
>> influence of English means that we can't trust people to pronounce a
>> single
>> spelling the Scots way.)
>
>The preterite morpheme isn't the same sort of thing as the
>plural - it varies between dialects, whereas the plural
>morpheme is consistent. Again, perhaps you underestimate
>people. Why can't you "trust" them? Who are "people"? Scots
>speakers? Foreign-language learners? English-speaking actors
>working with Scots scripts?

I don't think it matters who the 'people' are. Anyone who sees an <it>
ending on a verb is likely to pronounce it as [@t], unless they are
confident enough in their own command of Scots to know otherwise. This
means, in practical terms, that people who don't have a grasp of Scots
pronunciation write all sorts of words with <it> endings because they
think
it makes them look more Scottish (I presume), e.g. *haiverit and
*stammygasterit, and others copy them because they assume that this is
the
Scots spelling and maybe even the Scots pronunciation. So you can't
trust
people to either write or read such endings in the way you can trust
them
to correctly interpret <-ed> endings in English as either [d] or [@d] -
ie,
because they already know how to pronounce the words. It's not
inconceivable that some people might put <z>on the ends of the wrong
words,
for exactly the same reasons, if they got the idea that this was a
'Scots'
spelling. You might even get people using <dh>, or _edh_, in the wrong
place, likewise. (Another example is the willy-nilly use of the <an>
ending
- traditionally used of present participles but not verbal nouns -
because
it gives a more Scots appearance, although the people who do use this
ending rarely know where and where not to put it, as the distinction has
disappeared from most Scots dialects.)

In short (irony intended!) I agree that the situation with preterite and
plural endings is different, in that you can trust people neither to
pronounce preterite endings properly if they're all written the same,
nor
to write them properly if they're written with different spellings. You
_can_ however trust people both to pronounce and write the plural
morpheme
<s>, but if you introduce another spelling you open an extra possibility
of
(a) it being written wrongly owing to some misguided attempt to
emphasise
some imagined feature of Scots as opposed to English, and (b) that
mistake
being perpetuated in pronunciation. Why fix it if it ain't broke?
>
>> 2. The representation of Shewa and Aitken's Vowel is a problem. I find
>> the
>> use of <u> as initial Shewa very odd - my pronunciation in this case
>> doesn't sound like [V]. This is bound to be a problem in any orthography
>> where the graphemes have exact values. There isn't the scope for the
>> sort
>> of dialect variation which isn't regular. How do you decide when to use
>> <u>
>> and when <i> for an unstressed vowel?
>
>You can't phonemically - it's a feature of both Scots and
>English that unstressed syllables tend to be reduced. English
>takes an etymological solution, but this is something that has
>to be learned on a word-by-word basis. As I explained originally,
>this is reflected in the large number of "i"s you find used in
>unstressed syllables in this orthography I'm suggesting.
>
>However, using "i"s copiously for reduced syllables has an
>advantage - it makes it possible for a person to read an
>unfamiliar word and if there's only one non-"i" syllable
>they can normally assume the non-"i" to be the stressed
>syllable. So although it may not be possible to show all
>variant pronunciations of such syllables, the advantages
>can be greater than the disadvantages.

Why not use <i> for initial ones as well, then? Or <a>, which already
appears in some words in your Tam o Shanter example?
>
>This is something you need to bear in mind - it's easy enough
>to tear things apart, but the disadvantages you see in what
>you're ripping into may be nothing compared with the
>disadvantages of what you're used to!

I'm perfectly well aware of the disadvantages of what I'm used to - and
of
the advantages of a more regular orthography. However, even if the
argument
between historical and phonological orthographies was clear cut (and it
isn't) there would still be the problem of people's reactions.

Moreover, it's not the radical orthography itself that I'm criticising -
rather, I'm questioning its viability in the real situation, not because
of
any fault of its own, but because of existing attitudes. If it were
possible to choose between a terrible orthography which would gain
widespread acceptance and an excellent one that wouldn't, I'd go for the
terrible one.

I keep mentioning Faroese. There had been several other attempts to
create
a Faroese orthography, most notably by Jakob Jakobsen, a prominent
linguist
and also compiler of An Etymological Dictionary of the Norn Language in
Shetland. His Faroese orthography - based on phonological principles -
was
rejected in favour of a historical one, which has proved to be very
successful. Perhaps a phonological one would also have been successful,
but
the important point is that one was accepted - perhaps it didn't matter
which.
>
>> And what about variation in stressed vowels - for example, I say
>> ["antr at n],
>> but in the NE it's - to my ears - ["Vntr at n]. The well-known tendency of
>> some Central dialects to pronounce classic Scots <i> as [V], and of NE
>> ones
>> to pronounce classic Scots <a> as [V], raises questions about exact
>> representation, in an approach which doesn't take account of the
>> probability that the historical spelling represents an original
>> pronunciation which has then become merged with other phonemes in
>> certain
>> dialects, but not in others.
>
>Well, variation exists, and things like that will have to be
>ironed out. I mean, are we afraid to step outside in case of
>what might happen? And again, what I'm suggesting is still
>better than what we have now - it's still going to be a lot
>more teachable and learnable than the current schizographic(TM :)
>state of affairs.

Actually, it's schizographic (c) J.M. Tait :)>

It doesn't matter how teachable and learnable anything is if nobody
teaches
or learns it. Colson numbers again. Not to mention the Dvorak keyboard,
the
Klavier method of reading piano music, and doubtless umpteen other
sensible
alternatives to established applecarts. I used to think that the
relative
non-establishment of Scots meant that you could begin more or less from
scratch, but I learned that 'establishment' is relative.

(Incidentally, about the Dvorak keyboard, I've heard the theory that
it's
actually _worse_ for RSI sufferers, because its layout encourages a more
rigid hand position with less finger movement. You never can tell...)

Another illustration. As you know, I use a system of pronunciation
script
for Shetlandic (soon to be implemented on my Shetlandic website, if I
don't
get any more bugs in my macros) where I use a diacritic to indicate the
non-phonological change which affects certain vowels before voiced
consonants, e.g. <bat>, [bat], in my script \bat\, and <bad>, [b{d], in
my
script \bäd\. This shows the variation in pronunciation while preserving
the identity of the phoneme. In my abortive excursion on the outskirts
of
academia, however, I discovered that this was regarded as too difficult
for
students to learn, in spite of the fact that I had intended it to be
easier
than the Hobson's choice of phonetic or phonological representation
which
are the only alternatives. It doesn't matter how well suited it is to
the
representation of the subject matter if it breaks the established
conventions.
>
>> This brings out another problem. This type of orthography enables Scots
>
>Have you considered a solution?

Well, the use of a single grapheme for the plural morpheme, and another
for
both the [@I] and [aI] diphthongs, would be along the right lines, I
would
say.
>
>> A similar question arises with the spelling of the [W] sound. In a
>> from-scratch orthography, it would be silly to use <wh> for this, when
>> some
>> other diaphonological spelling (q, qh, quh, fh) could be used to
>> accommodate both the usual [W] and the Northern [f] pronunciation, and
>> sever the mental connection of <wh> with the SSE pronunciation for
>> Northern readers.
>
>You talk about this as if it was yet another problem with the spelling
>I'm suggesting. In fact I've allocated <q> for /W/.

I'm afraid I'm mixing up all sorts of problems here, and failing to
distinguish between (a) particular problems with your actual orthography
-
most of which could potentially be addressed - such as the omission of
the
/O/ phoneme, the use of two letters and for what I hear as
Shewa,
and the above (I think it was actually Ron who suggested <wh>?!); (b)
the
intrinsic characteristics of a phonologically-based orthography -
whether
we regard them as problems or not - such as the fact that they suggest a
particular pronunciation more readily than a historical one; (c) the
practical and political problems associated with orthography in the
Scots
context.

>> 3. The real problems arise with the attempt to process the original
>> orthography into a more traditional form. The best illustration is
>> 'dangerous' - a Romance (Old French) word borrowed with its French
>> spelling. The automatic 'traditionalising' process works well with
>> Anglo-Scots words, but, whereas the phoneticological (!?) spellings of
>> words like 'dainjiris' look all right in the context of other 'weird'
>> spellings, they stand out like a sore thumb when in the company of other
>> words which look 'normal'. This is why, in my opinion, any orthography
>> which sets out to have a traditional appearance must use etymological
>> and
>> morphological spellings as well as phonological and diaphonological
>> ones.
>
>Now wait a minute - you're making an argument for a maximally
>complex orthographic system purely on the basis that some
>spellings look "weird" or "stand out like a sore thumb" to
>you. This is completely subjective - what looks weird to you
>will look normal to someone who's used to it. How about
>restricting considerations mainly to phonemics on the basis
>that "weirdness" doesn't matter?

It's all relative, and a matter of context - that's why I put 'weird' in
inverted commas. It's other people's reactions I'm anticipating - not my
own. Nothing seems very weird to me - even writing backwards - I can
read
Hebrew, remember! Not only that, but I've spent Sunday afternoons
inventing
hypothetical alphabets for hypothetical languages. But a radical
orthography will certainly seem 'weird' to most people, and the unusual
spellings of Latinate and other words within a context of words which
are
spelt in a more familiar way will seem even more weird than they would
among other weird spellings. Of course this is subjective - but the
whole
idea of getting an orthography accepted is subjective.

Restricting considerations to phonemics is, unfortunately, exactly what
we
can't do. We have to consider the human, and political, aspects of the
problem.
>
>> As regards the acceptability of any such approach, the first question
>> which
>> must be addressed is the unacceptability of _any_ orthography, other
>> than
>> the _ad hoc_, to those who largely represent Scots in Edinburgh.
>
>So I'll address the question thusly - aren't the Edinburgh
>academics just a scapegoat for the fact that you're not able
>to provide any solutions either?

I wiss! I coud come up wi a hale heist o solutions, o ae kynd or
anither,
but for a solution ti be a solution it maun redd somethin. An solutions
at
naebodie taks tent o disna redd naething. The fact is at frae-scratch
orthographies is gey an aesy ti cleck, aince ye understand the
phonology,
an ken whit principles ye'r ettlin ti uise.  Solutions is mair harder.

Div ye mynd Colin sent us, a whilie back, a leet o the SNDA objections
ti
the spellin comatee (no, at that time I dinna think, ti the conclusions
o't.) Ane o thaim wis at thay didna gree wi spellin words different frae
English. In fact, as Colin pyntit oot, the'r a lot o words i the SNDA
dictionars - like _baud_ - spelt different frae English at wisna spelt
different bi the Comatee. I'm gey shuir at the SNDA didna pey nae heed
ava
ti Colin's comment. Tho facts - contrar ti the saw - is chiels at dings
gey easy, ideology is a chiel at nae amoont o facts can ding. It's no
scapegoats, but goats o the contermashious thrissle-chowin kynd, at's
the
problem!
>
>Gin A seem a bit greetin-face, John - juist think o me bein
>tae you what you ar tae thae Edinburgh folk!  :)

Nou haud on! I'm the big heid-ane greetin-face aroond here, an ye needna
think ye'r cowpin me aff ma bar stuil....

John M. Tait.

(PS - I hope this comes over in a readable format, without those
annoying
line breaks I seem to have been getting recently.)

----------

From: "John M. Tait" <jmtait at wirhoose.co.uk>
Subject: LL-L "Orthography" 2002.03.12 (04) [E]

Sandy wrote:

>I've been checking examples of these sort of words against
>the SND and I think I've made a mistake in trying to
>differentiate between variant schwa-like sounds by use
>of <u> instead of <i>.

>Do you think it would work to write "ibút" for "aboot", "iwå"
>for "awa" and suchlike, ie use "i" consistently for what's
>marked invariably as a schwa in the dictionary? Although this
>seems more like [V] to me, I don't think it's a difficult
>thing to take on board as in dialects like mine I think people
>are used to a certain amount of ambivalence between [V] and
>[i]/[3].

I would have thought so. My pronunciation of such words is more like
[3],
which is the /I/ phoneme, so I would have thought that would be ok.

However, you still have the problem that people who do pronounce it as
[V]
might make the opposite objection! In other words, if you're making a
more
exact representation of someone's pronunciation, you're making a less
exact
representation of someone else's! One question is - if such words are
written traditionally with <a>, does this mean that the sound _was_
originally [a], and that it later 'eroded' to [V]/[3], or was it always
just a way of spelling an indeterminate sound?
>
>The first vowel of "antrin" is only listed as [a] or [Q] in
>the SND (the unstressed second syllable accounts for most
>pronunciation variants but only through combinations of
>reduction and metathesis, which I don't think need to be
>addressed by an orthography). This suggests one way of
>dealing with spelling variants - we already have dictionaries
>that narrow down choices in their phonetic transcriptions,
>could we take these as authoritative and leave anything else
>to personal-dialect writers and recorders?

Well, the tendency to pronounce written <a> as [V] is a NE
characteristic -
'untrin', 'grunnie', 'new lumpies for aal', 'Muggie', etc. Again, I
think
the _original_ sound would have been [a] - as it is in Shetland - so
<a>
would be the way to spell it.

The question of the relationship between [I], [3] and [V] would bear
investigation. My Shetlandic pronunciation is probably irrelevant here
(except insofar as I have a definite distinction between the /I/
([I]/[3])
and /V/ ([o]/[2]) phonemes) because I pronounce /I/ as [I] or [3]
depending
on the voicing of the following consonant (e.g. _pit_, [p3t], _bid_
[bId] )
but I don't think this is the case in Mainland dialects. Is there any
rule
that governs the pronunciation [I]/[3] in your own dialect?

I believe that there is a particular tendency in Scots as a whole
(though
not in Shetland) for /I/ -> [V] after /w/ - e.g. 'wull', etc - and a
more
Central tendency for /I/->[V] generally - e.g. you write <shulin> but I
say
[sh3lin], which to me is an /I/ phoneme. In dialects where /I/ becomes
[V],
and where there is also an /V/ phoneme, then the change is obviously
phonological in that dialect - though not if /V/ were to be realised
differently in the same phonetic environment - but that doesn't answer
the
question of how it should be written in Scots as a whole. Again, my view
(cop-out?) is that, if it is traditionally written , then the best
idea
is to keep that grapheme, on the assumption that variation from it will
likely either be allophonic, or will represent phonemic mergers only in
particular dialects. To put it another way, you could turn to the most
conservative dialects to find out what the original/underlying phonemes
are
most likely to be - in this case, Shetlandic is useful even though it
can't
be treated as a dialect of Scots for orthographic purposes. This would
tend
to favour English-like spellings (<shilin>, <wil>) which in a radical
orthography might conflict both with - perhaps now more widespread -
Mainland pronunciations, and with the desire of Scots writers to
emphasise
non-English pronunciations.

In principle, this is a bit like the glottal stop question. Apparently,
one
of the derisory remarks made by, I think, politicians about the idea of
Scots road signs is 'How do you spell a glottal stop?' We know that the
glottal stop in Scots is a realisation of /t/ (occasionally /p/) in
certain
circumstances, varying with dialect as to what those circumstances are.
We
know this from the traditional spelling (in this case, the English
spelling) and this impression would be reinforced by phonological
theory.
Thus a field linguist who knew no English and nothing about the history
of
Scots would (I assume) come to the same conclusions as we do from the
traditional spelling. If the Scots were a tribe in the Amazon and a
Wycliff
Bible Translator had invented an orthography for Scots using <t> for the
phoneme whether it was pronounced [t] or [?], based on a phonological
analysis, it is possible that a native Scots speaker might not even
notice
the difference in pronunciation, just as most Shetlanders are not aware
of
their pronunciations of e.g. /a/ as /{/ until they compare it with SSE
pronunciation. But, because we are aware of the different SSE
pronunciation (although, in fact, many SSE speakers now use a glottal
stop as well) we do notice the glottal stop, and some written Scots
tries to represent it - I have seen _that_ spelt as <tha>, for example.
With these vowels we are talking about the principle may be the same, in
that the traditional spelling may represent the original phoneme - the
problem is that the evidence (apart from the traditional spelling) is
not so clear cut.

Hmm. I hope this makes sense...

(BTW - I would suggest á - like Icelandic - rather than å, by analogy
with
other uses of the 'acute' diacritic in the rest of the script.
Scandinavian languages which use å don't use the acute on other letters
either.)

John M. Tait.

==================================END===================================
 You have received this because your account has been subscribed upon
 request. To unsubscribe, please send the command "signoff lowlands-l"
 as message text from the same account to
 <listserv at listserv.linguistlist.org> or sign off at
 <http://linguistlist.org/subscribing/sub-lowlands-l.html>.
=======================================================================
 * Please submit postings to <lowlands-l at listserv.linguistlist.org>.
 * Postings will be displayed unedited in digest form.
 * Please display only the relevant parts of quotes in your replies.
 * Commands for automated functions (including "signoff lowlands-l") are
 to be sent to <listserv at listserv.linguistlist.org> or at
 <http://linguistlist.org/subscribing/sub-lowlands-l.html>.
 * Please use only Plain Text format, not Rich Text (HTML) or any other
 type of format, in your submissions
=======================================================================