From: Michael Haggett (email@example.com)
Date: Tue Feb 05 2002 - 10:54:37 EST
The recent spate of posts and the continual changing of the
thread subject means that things are beginning to get a little
confused (certainly for anyone trying to follow this thread in
the archives). I had hoped that the presentation of evidence in
the form of the Gemellus letter would move the discussion
(mainly between Randall and myself) onto firmer ground.
But his next post shifts back to the same sort of
assertions as before.
If we wish to convince the rest of the subscribers to this list
about the validity of either of our positions, we would do well
on concentrate on what the evidence may properly be taken to
MEAN. We can then build on this to say WHY we advocate
particular pronunciation systems and let others decide for
themselves which position is more cogent.
So let's look at this letter. Randall says that it's easy to
read if we use his particular pronunciation system. That isn't
saying much. He has based his pronunciation system on
manuscripts with spellings such as these, so by definition it
will sound natural. It's a truism. The inference is that it
will not make any sense to someone using a more conventional
pronunciation. This is not true. Speaking for myself, it makes
good sense to me.
I believe the manuscript is a phonetic representation of the way
that particular writer speaks Greek. It might equally have been
written using the conventional spelling, but the difference
between Randall's pronunciation system and a conventional one
seems to be this. Both will sound identical when spoken by
Randall because he naturally uses an Egyptian accent. However
the two will sound different when spoken by others: the Gemellus
letter will have a distinctly Egyptian sound, but the
(hypothetical) conventionally-spelt equivalent would be
pronounced in a conventional way.
The distinction I am making can be paralleled in English. There
are two ways of representing dialogue, and both are equally
valid. One might see this sentence:
"I've seen these three films this month."
The sentence as I've written it uses conventional spelling but,
because we come from different parts of the world and speak with
different accents, we will each speak this sentence in a
different way. Now suppose I was writing a novel and the
character speaking these words was Irish. I could do one of two
things: I could either spell the sentence in this conventional
way and leave my reader to imagine this being said with an
appropriate accent, or I could spell it in a way that
represented how that character would actually pronounce it.
"Oi'v seen deez tree filums dis munt."
(By the way, a film is a movie to you Americans ;-) My position
is that the Gemellus letter is an example of the latter. It is a
representation of Koine Greek spoken with a distinctly Egyptian
accent. In saying this I am not seeking to contradict Randall,
or imply that he didn't say or mean this. I am simply stating
what I think is the most likely reason for it to use the
spelling it does.
My conjecture is based upon this premise: the assumption that
the writing system is phonemic, i.e. that there is a direct
correlation between graphemes (the written letters and marks)
and phonemes (the sounds they represent). If we don't assume
this, we can't draw any meaningful conclusions about
pronunciation from this, or any, document at all.
Randall seems to be saying that, in terms of pronunciation, the
Gemellus letter is identical to its (hypothetical) conventional
equivalent. Therefore if both are phonemic, the letters in the
words that are spelt differently must be pronounced the same
way. I won't deny that this has a certain logic. However it is
not the ONLY logical conclusion that can be drawn. One can
equally postulate that both are phonemic, but represent two
different ways of reading the text. Let me explain what I mean:
Many of us grow up in regions, towns or even individual
neighbourhoods where people speak with distinct accents or
pronunciation systems. It's fine to continue speaking that way
if you never move out of them. But, if you do, you soon learn
that there is a more conventional way of speaking which is used
and can be understood by people across a wider range of local
In England, there used once to be something known as "the King's
English" - a received pronunciation that transcended local
variations. This would be the type of language used in
broadcasting, for example, and for this reason was sometimes
called BBC English. Generally speaking, the better educated you
were, the more likely you would be to speak it in situations
other than your own region, neighbourhood or family. Now,
please don't think I am advocating any kind of elitism or
snobbery . I certainly do NOT think that any one form of English
is better than any other. However each wider geographical area
develops a standard of this sort based on the principle of
understandability. For example, if Tony Blair were to make a
speech, he would use this conventional English, even though
there is a vast range of accentual variation in Britain (he was
born in Scotland, by the way, although you'd never be able to
tell). The same is true in America: George W Bush, when
addressing his nation, will use a more conventional form of
English than a Texan drawl. Bill Clinton did not get elected by
speaking with an Arkansas twang. Of course there will be traces
of accent in anyone (although the natural perception is always
that YOU don't have an accent ... everyone who speaks a
different way does!) but anything that obscures
understandability will tend to go.
By the same reckoning, I believe there is a conventional
standard for Koine Greek. It is a standard represented by the
predominant spelling of documents which have a wider importance
than the merely local. Subject to all the provisos of Textual
Criticism, this is what has been copied down the ages and is now
printed in the Greek texts we use.
Of course the words in these documents can be pronounced with
any one of hundreds of local accents, but a conventional
pronunciation will be readily understood by everyone in a wider
area, whereas a specific local pronunciation will usually be
understood only by those who use it, or those that have become
familiar with it by means of travel.
If I apply this theory, reading out the conventionally-spelt
Greek text will sound DIFFERENT from reading out the same words
from a text in which the spelling represents a local dialect
OTHER THAN MY OWN.
Of course what I read from a conventionally-spelt document will
be spoken with MY particular accent, and if someone else were to
read out the same document it would reflect THEIR accent.
However, if I read out the Gemellus letter as it is written, the
listener will hear a rough imitation of an Egyptian accent - and
the same distinction will be true of anyone who does not usually
speak Greek with an Egyptian accent. ONLY if you normally speak
Greek with an Egyptian accent will both sound the same.
Under my theory both the conventionally-spelt text and a
regionally-spelt text are phonemic, but the greater range of
distinctions represented by conventional spelling can be
maintained - the equivalents are valid, but only in particular
circumstances. Under Randall's theory, distinctions are
eliminated and the equivalents are universally applied. True,
Randall's theory allows for different accents, but only within
the bounds of across-the-board equivalence.
Now, which is more likely to be true? For this we have to look
more closely at how equivalence applies to real language. We
need to see how well theory applies to a language for which we
have actual aural evidence. I'll use English because it's the
one language everyone on the list will understand
Let's take a word like plough. The American spelling is
different, presumably because plow more accurately represents
the way it is pronounced. I can hardly argue with that - plow
now rhymes with cow! However a future researcher (without any
aural evidence) trying to figure how things were pronounced in
our era would see that the spellings plough and plow were
interchangeable, and therefore might quite reasonably conclude
that "ow and "ough" were generally interchangeable. That is what
Randall is doing and, in this parallel, he'd be right.
However, let's think just a little more deeply before we move on
and apply this conjecture in an across-the-board manner. Does
it apply in other situations?
Both plough and plow rhyme with cow ... and how! But plow does
NOT rhyme with snow or blow. Conversely, a word with the same
spelling can be pronounced in two distinct ways. Vincent van
Gogh's name is pronounced to rhyme with snow in America but
cough here in Britain. Moscow is pronounced to rhyme with cow
in America but snow in Britain. Real language contains anomalies
like these (and more, such as colour/color being equivalents,
but complete words such as our/or not being equivalents) which
you would never be able to figure out from the written evidence
alone. I believe you can only apply across-the-board
equivalence based on what is written if you know there are no
exceptions in the various ways words are actually spoken. It is
unsafe to do anything else. I suppose it is just logically
possible that there was a complete equivalence over all
accentual variations in the Koine-speaking world. But how
anyone can conclude that so remote a possibility is actually
proved beyond reasonable doubt is quite beyond me!
Now I move on again, because the ultimate aim of this exercise
is to come up with an teachable pronunciation model. Any such
model is conjecture. I believe we can only determine how useful
it is by how consistent it is with the principles on which it's
supposed to be built, and how consistently it can be applied in
practice. Any theory might look great on the ground, but can it
The Erasmian position that I (among many others) advocate
maintains that each letter or letter combination represents a
distinct sound which is not shared with another letter or letter
combination. More bluntly: "What you see is what you'd hear"
or "You read the document the way it's spelt." Please do not
think that by "Erasmian" I mean any one particular pronunciation
model. I don't. What I mean by Erasmian is a system of
pronunciation that is phonemic without redundancy. Any
individual pronunciation system can be classified as more
Erasmian or less Erasmian to the degree that it conforms to
this. The systems that are called Erasmian broadly have
separate sounds for separate letters or letter combinations.
Some of the actual pronunciation systems that are taught are
less than perfect (I'm thinking of those that don't maintain a
distinction between EI and H or EU and HU) but the difference
between the various Erasmian systems is so minor that they can
be grouped together under that name.
Yet it is evident that people use the term Erasmian to mean
different things. Randall uses it to describe the pronunciation
model most commonly taught in North America (I think he once
called this North American 'Rasmian). Stephen Carlson uses the
term to mean the way things were once actually spoken when Greek
adopted the 24 letter alphabet. I don't want to argue about
terminology, I just want to clarify what I mean by the term.
To me it is far from clear what principles guide Randall's
actual pronunciation model. I thought he was saying that the
development of Koine in the Roman period was characterized by
the loss of long vowels ... that, for example, W had become O,
that diphthongs were being reduced to single vowels and that
long A had become short A. I still confess that I can't make
sense of the equation A = A except with that understanding
(unless he was quoting Aristotle ;-)
I could see some logic in that. I can well see how the
pronunciation of a word like BASILEIA could change to BASILHA
and then to BASILIA (Stephen's example) ... it fits my
particular pronunciation model perfectly. But Randall has
now clarified that he means not a one-way process, but a
two-way process. To me this is quite disturbing.
He seems to be saying that you can swap at whim
(although to him, IF both alternatives are
pronounced the same way, it can hardly matter).
Of all his pairings I'd like to concentrate on W = O. I choose
this pairing because it is inherent in the definition of these
letters that one represents a long "o" sound and the other a
short "o" sound (at least by any logic I can figure). On his
website we see a picture of a man eating an egg. Yet the
pronunciation, as I would transliterate what I hear, is:
W ANHR ESQIEI TWN WON
(= hO ANHR ESQIEI TON WiON using conventional spelling)
We hear some, but not all, the conventionally written Os spoken
as if lengthened to Ws. Why some but not all? By what possible
logic is the accusative ending of the definite article not the
same as the accusative ending of the noun? It appears to me
that this is a purely arbitrary choice, making a distinction in
practice that he maintains is not made in theory. If I were his
student, I would be confused. I'd have to rely on his aural
model for the pronunciation of every single word, I couldn't
rely on any simple principles, I certainly couldn't rely on what
was printed in the text before me.
Furthermore, what are we to make of the meaning of the sentence?
It is capable of meaning two different things. Is the man eating
the egg (accusative singular), or is he eating (some) of the
eggs (genitive plural)? Either one accepts that the means of
making that distinction have been lost (the theory) or one
maintains in speech the written distinction between vowels in
order to make the meaning clear ... yet doing it inconsistently
Again, one of the criticisms I've heard Randall make of other
people is the way that they pronounce U as if it were OU ... yet
isn't that exactly what he does when pronouncing ICQUS and
ICQUN? I guess he must be trying to stress that syllable, but
isn't that confusing stressing a syllable with lengthening it?
OK, perhaps a few sentences is not enough for me to get a fair
indication of the way he actually pronounces things, but the
little I can hear doesn't inspire me with much confidence.
So which is the best model for students to learn? You can guess
my answer. To me the advantage of postulating an
over-arching conventional pronunciation based on the
conventional spelling is that it allows you to modify it
to speak in any accent you choose.
If you have a wide range of sounds to choose from
you can always opt not to use all of them if you believe that
they were not all used in any particular pronunciation. If you
never learn the full range at the outset you will either have to
learn them later or run the risk of being misunderstood by
others who didn't buy into the particular system you chose.
I have always maintained that differences in the way we
pronounce New Testament Greek are both inevitable and healthy.
I do not wish to invalidate any of them, not even Randall's ...
it's just another variation that we have to live with.
But I hope to have shown that it is not the ONLY way
of interpreting the manuscript evidence.
Randall might well think that learning a more conventional
Erasmian pronunciation gives you a bear of a problem. But if I
(with tongue firmly in cheek ;-) write that his pronunciation
sounds like a cow with a cough, he'll only hear it as two
cows ... or two coughs.
B-Greek home page: http://metalab.unc.edu/bgreek
You are currently subscribed to b-greek as: [firstname.lastname@example.org]
To unsubscribe, forward this message to leave-b-greek-327Q@franklin.oit.unc.edu
To subscribe, send a message to email@example.com
This archive was generated by hypermail 2.1.4 : Sat Apr 20 2002 - 15:37:17 EDT