MikeS
posted on Jul 08, 2003 - 09:26 PM
Nofri ouon niben,
I seem to recall that there was a question on one of the topics that dealt with the definite article in Coptic and why the forms `v and `; were used in front of certain letters.
I had commented back that I would look into it and see if there was a reason for this (I think the original question was to that effect).
Anyway - I did a little study of the definite article in Coptic and
here's what I've been able to come up with on the usage of the definite article. Though I don't have a definitive answer regarding the variations found in the Bohairic, I've put together a short synopsis of what I have found and offer an explanation, though I'm not 100% convinced myself, it's just offered as a possible explanation regarding the various usage in Bohairic.
Lets' go way back for a moment... Ancient Egyptian did not have articles, but it did have demonstrative pronouns. Like most languages (English included), the articles developed from the demonstrative pronouns. This development seems to have occurred in the Late Egyptian period. Whether or not this development was a natural one, or whether it was influenced by other languages, or even a combination of both will probably never be known. Suffice it to say that by the time the language developed into what we call Coptic today, articles were already widely used in the language.
I think there may have been two stages of development in Bohairic ñ one that happened as the natural development of the Egyptian language and its dialects, and one which happened quite later.
Let's look at Sahidic first to see what I mean.
In Sahidic, we see the following definite articles:
p pe
t te
`n ne
How they're used is quite simple; if the noun begins with a single consonant or a vowel, the shorter forms (p, t, `n) are used. If the noun begins with two consonants (remember this also includes those double consonants written with a single letter ñ in Sahidic, theta, phi and chi are t+h, p+h and k+h, not as in GB), the fuller forms (pe, te, ne) are used. The fuller forms are also used with certain nouns denoting periods of time (e.g. terompe, pehoou).
Sahidic is that simple! Note also that if the noun begins with a 'p' or 'm', the 'n' in `n and ne will become an 'm'. This is just a simple case of place assimilation. Also, if the noun begins with 'h', the 'p' and 't' are combined with the 'h' and are written as phi and theta (e.g. hwr ñ vwr). You will find that some scribes wrote phwr, etc., but the combined forms were more common.
The forms for "the God" and "the mother" in Sahidic are Pnoute and tmaau, not `Vnoute and `;maau as you would expect with Bohairic definite article.
So, as I hope you can see, Sahidic articles are rather simple and straightforward.
Let's take a look at Bohairic now.
We see the following forms in this dialect:
p, pi, `v
t, ], `;
nen, ni
Here's where it gets interesting. I think the original development in Bohairic just included p, pi, t, ], nen, and ni and that the forms were employed just as they were in Sahidic.
It's also possible that the forms `v and `; came about just as they did for Sahidic as well (i.e. combining p and t with nouns beginning with 'h'). Why then is the modern use nothing like the Sahidic use?!
Yes, it's true, you have two different dialects here, but the only thing dialectally different between the two is that Sahidic has 'e' where Bohairic has 'i' in the fuller forms (remember t+i = ]). It's also been said by some that Sahidic 'e' for Bohairic 'i' might have been an attempt to write a "shwa" sound in a language which had no other way of writing it (the supralinear stroke ñdjinkim- might have been another way).
The "modern" usage shows us something quite different though.....the rules say that there are actually two forms of the definite article; the so-called "strong" and "weak" forms. The weak forms are the older short forms and in modern usage, these are used if the noun is abstract, generic, one of a kind, when you're not specifying a noun in an exact way. The strong variations, which come from the old full forms, are used when the noun is specified in an exact way or a specific thing or person is referred to.
This usage, I suspect, was a much later development in the language as it's not, at least as far as I know, evidenced in Sahidic. I have to wonder if this development didn't take place after Coptic ceased to be an everyday spoken language. Is it possible some confusion arose in the proper usage of definite articles? It is quite possible that over time simple aspects of everyday grammar can be forgotten. Again to use English as an example ñ our indefinite article rule is quite simple (or so it seems!) ñ if a word begins with a vowel, it's "an", otherwise it's "a" (though some will use "an" before 'h' as well). But what happens if you're a scribe writing a document and can't remember if it's 'an apron' or 'a napron'? Both are pronounced exactly the same. This word in English (along with many others) became "apron" ñ but that's not actually the correct word; it should be "napron" (viz. 'napkin' ñ a little 'napron' - -kin is the diminutive suffix in English)! English actually has a number of these "errors" which have become standardized over time.
I have to wonder if this didn't happen in a later period with the copying and re-copying of Coptic texts ñ i.e. just scriveners errors happening over time? I'm not suggesting the writing of Coptic was careless in anyway, but being an amateur calligrapher myself and having learned the hard way from experience, copying texts is very tedious work and errors can very easily be made.
What I'm suggesting that may have happened is with the pronunciation particularly of `p, `t, pi, ] ñ this sort of "overuse", if I may use the term, of applying the djinkims may have resulted in the deterioration of the traditional use of the definite article possibly because people simply couldn't remember the correct form, especially if they sounded almost the same in pronunciation.
If you look at hymns or texts and try to apply the rules for what I think may be the original usage (short vs. full forms), it simply doesn't work ñ from this point of view, it's a mess! In Pekethronos, for example, you see "pi`sbwt" and "`psbwt" used in the same sentence ñ if you apply the older usage, you'd expect to see just the fuller form (pi`sbwt). As someone else had pointed out, you can see things like Pinou] `nten `vmyi and `Vnou] nai nan - `vry and piry (in which one would actually expect to see `vry)!
Again, I have to wonder about those djinkims ñ if `p can be pronounced like pe, ep, pi ñ particularly the later (same with `t as et, ti), I can see where confusion would arise over whether it's written as `t, `p or ], pi!
It's possible that as the usage deteriorated, a new rule had to be given to account for the changes. In all fairness, it could also be argued that these variations in usage are the actual natural changes of the usage in progress right before our eyes, i.e. this switch from short/full forms over time to becoming week/strong forms as witnessed in various texts. Or, by the same token, as I mentioned, you could also argue that there was so much deterioration, some sort of new rule had to be posited to account for the differences in usage. The week/strong rules as illustrated above, can sort of account for all these changes rather nicely; if the wrong form were present, it could be argued that the writer simply wanted to add more emphasis/"definiteness" on the word (or not, depending on the case).
Again, I'm not saying this was the case of what happened, but I think both explanations are plausible. I think to properly argue either way, you'd actually have to do a small study of early Bohairic texts and see if the usage of the definite article conforms to that of Sahidic (i.e. the use of a short/full form as shown above). If it could be shown that Bohairic followed these same rules, you could argue that the week/strong thing was a much later development in the language.
This still doesn't explain why we se the forms `v and `; before the so-called "blemner" consonants (b, l, m, n, & r)!
First some general comments about this group of consonants.
I don't know if Bohairic does this or not, but with Sahidic, if a word has a blemner consonant and the djinkim is used, the blemner consonant is doubled (`rro ñ king, b`lle ñ blind). 'e' is used regularly instead of a djinkim only when a consonant preceding the one that would take the djinkim is a blemner, so "mokmek", not "mokm`k".
Also just a quick comment about the pronunciation of `v. Some texts will give "ev" in transcription, while others give "ef" ñ so you see the forms "efnouti" as well as "evnouti". Both are actually correct. The underlying form is "efnouti" and is the form you'd hear in careful, well, annunciated speech. In everyday conversation, however, people just tend to talk fast and what happens here is that the "f" undergoes what's called a voicing assimilation and becomes it's voiced counterpart, 'v', in anticipation of the voiced "n" in "nouti". I suspect, therefore, that both ev and ef are acceptable pronunciations as they appear to be in what's called "free variation", i.e. either one may be used without changing the meaning of the word. Some speakers, in fact, may use both pronunciations in the same sentence for the same word!
OK ñ I've rattled on long enough about /p/ and /t/. The main question was WHY are the forms `v and `; used before this particular group of consonants??!!
To look at this group of letters we have; b, l, m, n, r, i, and ou. Note that I write these sounds as phonemes by using the "/" symbol, so when I write /w/ and /j/, it's really just the 'w' in aWay and the 'y' in Yell. These sounds are written in Coptic by ou and i and occur when followed by a vowel (e.g. Peniwt).
Regarding the last two - `v and `; are only used if the i and ou serve as glides, that is, if they receive the consonant sounds of 'y' and 'w' respectively, otherwise the regular forms of `p, `t are found (e.g. `pouro, because ou here is /u/, not /w/, pIcrayl because I here is /i/, not /y/, BUT `Viwt, because i here is a consonant 'y', written in phonetic transcription it would be /efjÛt/, e;ouab, because ou here is a consonant 'w' (phonetically /ethwab/.
Coptic seems to have a neat little rule that says before consonants that are not obstruents (stops, affricates, and fricatives), i.e. in front of resonants (nasals, liquids and glides), you can't have the sounds /t/ and /p/. So, in other words, b, l, m, n, r, w (ou) and j (i) are resonants (OK b isn't, but I'll get to that), the rest are not. Our rule, therefore is that p, t > v, ; /_X [+ resonant] ñ that is p, t, become v, ; in the environment where they precede a phoneme ("X") which is a resonant (/b, l, m, n, r, w, j/).
This is just a guess on my part, but I think what we're seeing here is what's called a "place assimilation", that is, p and t are what are called stop consonants; to make them, the flow of air through the mouth gets completely stopped, as you go down the hierarchy of sounds from air being completely stopped to air simply resonating different ways in the mouth, but never being stopped (the hierarchy for Coptic is stops, affricates, fricatives, nasals, liquids, glides). So what I think might be happening is that when what may be an underlying p, t come before b, l, m, n, r, ou and i, these sounds are actually "pulling" the p and t down the hierarchy to make them sound more like resonants ñ they don't make it all the way down (just two "steps") and turn into their corresponding fricatives; f/v and th.
This seems to explain most of what's going on kind of nicely, but there's one problem (sort of) ñ I'm basing this all on the Greco-Bohairic pronunciation, not on the OB. If I base it on OB, we have a problem ñ where Coptic writes `p, pi, `t, ], `v,`; ñ OB pronounces (as far as I know) ab, bi, ad, di, ab, and at, respectively. If we try and apply the same rules as outlined above, we run into a problem; in pronunciation, there's no difference between `p and `v in OB! Also, the only difference between `t and `; in OB is that the former is voiced and the later is not (d/t) ñ the assimilation is a place assimilation, not a voicing assimilation ñ So, what does this tell us??!! From looking at this, I would have to say that the use of `v and `; as definite articles in Bohairic, would seem to have had to come about at a later date ñ about the time Greco-Bohairic started becoming used.
Of course, I could be entirely wrong here!! I would almost want to see what happens in older documents, as I mentioned ñ if older documents just evidence `p, pi, `t, ] as the articles, two things would need to be looked at; what happens to a word like e;ouab? One would expect to see it written as *'etouab' in older documents, same with the example `viwt ñ one would expect to see *`piwt. the other question would be, what is/was the sound in older Egyptian? You couldn't do this with articles (older Egyptian didn't have any); you'd have to use words where in Coptic ; and v occurred internally within the word before b, l, m, n, r, w, j.
OK, so to summarize here
ñ I think it's possible the reason `v and `; are used for the definite article before b, l, m, n, r, ou, and i is because of a place assimilation with what is really an underlying p and t.
- It appears as if the use of this form of the article was a later development in the language as these forms are not attested in Sahidic.
- It looks as if Coptic may also do this with it's sounds in general, i.e. this particular assimilation may happen, as one would expect, "across the board ".
I've taken a look at a few texts (hymns) and noticed the rule does indeed seem to go across the board. Coptic doesn't seem to like the sounds /t/ and /p/ before resonants (b, l, m, n, r, w, j ñ last two written i and ou).
Some of the evidence from texts is as follows:
e;be
e;by]
e;picaji
`;recmacf
;yeymeh
`nta`vmyi
me;myi (possibly a Gk. loan ?)
This phonological feature seems to hold only in Coptic words and does not apply to Greek loans as we see the typical Greek combinations of /tr/, /pr/, and /pn/ being allowed (e.g. petro, pi`prodromoc, `pneuma).
I haven't come across an example in Coptic (with a Coptic word, not a Greek loan) where you'd have /t/ or /p/ occur before b, l, m, n, r, w, j.
But this kind of presents a problem ñ again, this rule is based on GB pronunciation, not OB. So, can we assume that at some point along the line Coptic underwent an orthographic reform at the beginning of the GB period? The rule above kind of suggests that the older spellings wouldn't have these spellings as the OB pronunciations don't fit the rule. I tend to think not ñ again, I may be wrong ñ you'd have to look at old texts and see how these words were spelt ñ did the Bohairic of, say three hundred, four hundred years ago, write in the examples above: etbe, etby], etpicaji, `trecmacf, ;yetmeh, `nta`pmyi??! If so, can we determine a point in time where manuscripts start showing the forms as they are today?! Also with the forms Pinou] and `Vnou], show the correct "modern" strong and weak forms respectively. We have seen the older form of the word in Sahidic is Pnoute. Is it possible in older Bohairic texts to find the form (and only this form) "`Pnou]"? Was it ever just written this way? The form "Pinou]", should be a fairly recent variation ñ is there evidence of this form in texts prior to say, 1750?
OK ñ I'm rambling here and thinking out loud ñ but the point is, this phonological rule works great for GB, but given OB pronunciation, the rules kind of don't make linguistic sense and simply wouldn't work (assuming my understanding of OB pronunciation is correct concerning the letters t and p 
).
- If this happens across the board, AND it appears to be a later development, can we postulate earlier forms of Bohairic which do NOT show this happening?! (see above comment)
Oh, yeah, what about 'b' ñ I was supposed to get back to this wasn't I? Here, again, there is a bit of a discrepancy. In GB it's /v/ sometimes /b/. If we use this pronunciation, the rule really shouldn't apply to the letter 'b'. If, however, we take the OB pronunciation of the letter, more or less /w/, the rule works great, since /w/ is a glide!
I really am not all that familiar with the history of the Bohairic dialect, but based on the "evidence", I would suggest that the letter 'b' must have been a hard letter for speakers of OB to "change over" to the GB /v/, /b/, or perhaps one of the last letters to change over (they may not have all changed at once and old habits die hard). I suspect there may have possibly been a sort of intermediate stage, somewhere between OB and GB (call it "early GB" if you want), where the 'b' was still pronounced as /w/, so the rule above applied. Again, though, I have no idea of a timeframe here. GB is supposed to have been implemented in the late 1800's. Is it possible that some were using this pronunciation before this time?!
OK ñ so this is probably more than you ever wanted to know about the definite article in Coptic (but were afraid to ask 8O ) I hope this makes some sort of sense and answers a few questions ñ It's actually raised a few for me!! I'd appreciate any other information, or comments you can think of. I don't mean to turn this into a "linguistics forum" on Coptic, but it would be nice to try and determine why Coptic works the way it does.
Respectfully,
Mike S
P.S. ñ this has got to be the longest post I have ever written anywhere!! I appologize if the Coptic fonts are not correct - I think I double checked them all!
