[Editorial note: [...] indicates use of Coptic text. Original script is available for viewing in the PDF format of this article.]
(CE:A186a-A188b)
PHONOLOGY OF THE GREEK OF EGYPT, INFLUENCE OF COPTIC ON THE. The main source for the Greek language in Egypt is the mass of nonliterary papyri, ostraca, and inscriptions from the Ptolemaic, Roman, and Byzantine periods, a total of almost fifty thousand documents. An analysis of the orthographic variations in these documents indicates that the pronunciation of the Greek koine spoken and written within the confines of Greco-Roman Egypt reflects to a large extent a transitional stage between that of the classical Greek dialects and that of modern Greek. But there is also extensive evidence of bilingual interference in its phonology by Coptic.
As regards consonants, there is some evidence from as far back as the early Roman period for the shift of the classical voiced stops /b/, /g/, and /d/, represented by [...], [...], and [...], to fricatives, as in modern Greek. But there is abundant evidence from documents of the same period and place that these sounds were still stops, for [...] and [...] interchange very frequently, and [...] occasionally, with the symbols for the corresponding voiceless stops [...], [...], and [...], respectively. Similarly, [...], [...], and [...], the symbols for the aspirated stops /kh/, /th/, and /ph/, also interchange frequently in the same documents with [...], [...], and [...]. This confusion, found extensively only in Egypt and paralleled in the spelling of Greek loanwords in Coptic, has no satisfactory explanation in terms of Greek phonology, for although both the voiced and aspirated stops have shifted to fricatives in modern Greek, they have never merged with those of another order but have remained distinct to the present day.
In Coptic, however, there was no phonemic distinction between voiced and voiceless stops in any dialect. But the sound represented by [...] occurs as a distinct phoneme, pronounced during the Greco-Roman period as a voiced bilabial fricative [[...]]; hence, the symbols for the labial stops are not so frequently confused. Similarly, the unconditional interchange of aspirated and voiceless stops is caused by bilingual interference. In Coptic, aspirated stops were phonemic only in the BOHAIRIC dialect, where the opposition occurred only in accented syllables and the aspirates were lost in late Byzantine times.
In addition, the voiced bilabial fricative quality postulated for Greek [...] especially when it interchanges with [...] /w/ or [...] /y/ coincides with that of Coptic [...], and the fricative quality of intervocalic Greek [...] in connection with rounded back vowels may represent the labiovelar fricative quality of the Coptic [...] /w/.
There is also widespread confusion of [...] and [...]. Although in Greek the phonetic quality of these liquids varied considerably, nowhere outside Egypt was there an identification of the two sounds. In thc FAYYUMIC dialect of Coptic, however, from which area most of the documents showing this interchange come, there may have been only one liquid phoneme /l/, for most words spelled with [...] in other dialects show [...] in Fayyumic, although [...] is retained in many words.
The final nasal is frequently dropped in pronunciation, a tendency that has continued in spoken Greek to the present day. In addition, medial nasals are frequently lost, especially after stops. This is also the result of bilingual interference, for in Coptic a voiceless stop had a voiced allophone following a nasal. This fact, combined with the underdifferentiation of voiced and voiceless stops, made [...], [...], [...], and [...], for example, simply orthographic variants of the same sound /t/.
Initial aspiration is frequently dropped. This represents a phonetic tendency within Greek itself, in which aspiration was generally lost during the period of the koine. Aspiration was also lost in some Coptic dialects in Byzantine times.
In vowels, the classical long diphthongs were reduced to simple vowels by the end of the first century B.C. The short diphthongs in [...] became identified with simple vowels, [...] with [...] in /i/ already in the third century B.C., [...] with [...] in /[...]/ in the second century B.C., and [...] (and [...]) with [...] in /y/ by the first century AD.
The short diphthong [...] had become a simple vowel /u/ before the beginning of the Ptolemaic period. In the Roman and Byzantine periods, it interchanged occasionally with [...] and o, both representing /o/. Since this interchange was rare elsewhere in Greek but was paralleled in Greek loanwords in Coptic, it may rest on bilingual interference. In Coptic, [...] is a reflex of [...] and [...] and [...], and it has been proposed that [...] after [...] represented the same sound; but a phonemic opposition between /o/ and /u/ seems well established.
By the second century B.C. the short diphthongs [...] and [...] were showing evidence of the reduction of their second element to a consonantal sound [w], which closed to a bilabial fricative [b] in Byzantine times. This corresponds to the known historical development of these diphthongs from original /au eu/ to /av [...]v/ or /af [...]f/ in modern Greek. Parallel orthographic variations in Coptic suggest that Greek [...] and [...] may have been identified with Coptic [...] and [...], both arising frequently from contraction from [...] and [...], which also represented a vocalic plus a consonantal element.
The simple vowels for the most part preserved their classical Greek pronunciation, but itacism was more advanced because of the nature of the Coptic vowel system, in which there were only three front vowel phonemes corresponding to the four Greek front vowels. In addition, [...] seems to have been bivalent, since throughout the Roman and Byzantine periods it was confused sometimes with the /i/ sound represented primarily by [...] and [...], and sometimes with the /[...]/ sound represented by [...] and [...], as well as frequently with [...]. In Coptic, [...] occurred only in accented syllables and was bivalent. In all dialects it represented an allophone of /i/ before or after sonants. In Bohairic, it also represented an allophone of /ae/.
The simple vowel represented by [...] was particularly unstable. In the koine where the diphthong [...] came to represent /u/, u apparently represented the Attic value /y/, until it finally merged with /i/ about the ninth century A.D. The interchange of the symbols for /y/ and /i/ possibly indicates the unrounding of the /y/ and its merger with /i/ in Egypt during Byzantine times. But the constant confusion of [...] with other vowel symbols, especially [...], suggests underdifferentiation of phonemes through bilingual interference, since Coptic had no /y/ sound. There are parallel interchanges of [...] with [...] and [...] in Greek loanwords in Coptic.
There is also a frequent interchange of [...] with [...] and o, mainly in unaccented syllables but occasionally in accented syllables as well. This is also the result of bilingual interference, for in no dialect of Coptic were there more than two phonemes corresponding to the three Greek phonemes represented by [...], [...], and o.
Finally, all quantitative distinction has been lost. This in turn reflects a change in the nature of the Greek accent from pitch to stress, which came about in Egypt, as generally throughout the koine, through the transfer by nonnative Greek-speakers of their own accentual patterns to their Greek.
The possibility of the influence of Coptic on the phonology of the Greek of Egypt has long been recognized but usually not invoked to explain more than isolated phenomena in documents clearly emanating from the Egyptian element of the population. But the evidence of bilingual interference in the nonliterary papyri, ostraca, and inscriptions, especially from the Roman and Byzantine periods, is so extensive that Coptic influence must have fairly permeated the Greek language in Egypt.
FRANCIS THOMAS GIGNAC, S.J.