Akhmimic

akhmimic. The Akhmimic dialect (siglum A), along with the Sahidic (S) and the various vernaculars of the Lycopolitan dialect group (L), is one of the Upper Egyptian dialects of Coptic. Its range of distribution extends from Aswan to Akhmim (Coptic smin or ,min; Arabic, ’A mîm; Greek, Panos or Panopolis). Thebes is considered the point of origin as well as the center of the Akhmimic-speaking region (Kahle, 1954, pp. 197-203; Vergote, 1973, Vol 1a, p.4). However, according to the former view of Kasser (1982a), the proto-Sahidic dialect (a reconstructed entity, symbol *pS, very like dialect P) also began to develop in Thebes after having infiltrated from the north as early as the second half of the third century A.D. As a result, in the fourth century A and S were in concurrent use in the Upper Thebaid. While Akhmimic had not been uniformly standardized, it became a medium of writing as early as the fourth century and reached its zenith in the fourth and fifth centuries. Beginning in the fifth century, it was gradually displaced by the Upper Egyptian koine (S), although for its part A (besides L?) both influenced S—as seen in the Nag Hammadi texts and Shenute—in the fourth and fifth centuries and left traces in nonliterary texts from Thebes dating from the seventh and eighth centuries. These observations imply that even after being displaced by S, A was still in use as a spoken language.

1. General Characteristics The dialectical features typical of A may be sumarized in contradistinction to S (if necessary also to P; cf. Kasser, 1960) and to the group of dialects included under L (i.e., L4 = Manichaean texts; L5 = London Gospel of John, ed. Thompson, 1924, as well as the Dublin fragment of the Gospel of John and the Geneva fragment of the Acta Pauli, the latter two being unpublished; L6 = Nag Hammadi texts of codices I, X, and XI, as well as the Heidelberg Acta Pauli, ed. Carl Schmidt, 1904, 1909).

The textual citation below refer to editions and studies of the documents noted in the bibliography as follows: APocSoph = Steindorff, 1899; I Clem. = Schmidt, 1908; I Clem.(R) = Rösch, 1910 (pp. 1-87); Elias = Steindorff, 1899; EpAp. = Schmidt, 1919; Ex. = Lacau. 1911 (pp. 45-64); Herm. = Lefort, 1952 (pp. 1-18); Jn. = Rösch, 1910 (pp. 119-60); MinProph. (Minor Prophets) = Till 1927b (for Obadiah 1-13; Amos 1:1-2:11; 6:3-10; 7:10-16; Habakkuk 1:7-2:2, 2:11-3:19; Haggai; Hosea; Joel 1:1-14, 2:19-3:21; Jonah 4:2-11; Malachi 1:1-7, 2:9-3:24; Micah 2:11-5:8, 6:1-7:20; Nahum 1:1-3:8, 3:14-19; Zephaniah; Zechariah 1:6-4:5, 7:14-14:21); MinProph (Minor Prophets) = Malinine, 1950 (for Obadiah 13:21; Amos 8:11-9:15; Habakkuk 1:1-7, 2:2-17 (sic); Joel 1:14-2:19; Jonah 1:1-4:2; Micah 1:1-2:11, 5:8-6:1; Nahum 3:8-14; Zechariah 1:1-6, 4:6-7:14); Luke = Lefort, 1953; Ost. = Till 1931 (text A); Prv.=Böhlig, 1958; Sir. = Lacau, 1911 (pp.64-67); P (Papyrus Bodmer VI) = Kasser, 1960.

1.1 Grapheme Inventory. Beyond S and L, A ads the alphabetic symbol |, the postpalatal spirant /x/ (or h, [x2] from the Egyptian or h); in P, the symbol for this is q.

1.2 Phonetic Characteristics.

1.2.1. A, L /a/, S /o/ in a closed stressed syllable before a nonlaryngeal sound, e.g., A, L can, S con, brother.

1.2.2. A, L /e/, S /a/ in a closed stressed syllable, e.g., A, L ren; S ran, name.

1.2.3. A /I:/; S /e/ before a laryngeal in the medial and final position, e.g., A mieise, mise, S m//se, L m/se, crowd; A ouieibe, S, L ou//b, priest; A ri, S, L r/, sun; A ou|i, S, L ous/, night.

1.2.4. A, (L5) /u:/, S, L /o:/ in a long final position or before a laryngeal, e.g., A, (L5) kou, S, L4, (L5) kw, L6 kwe, to place (<Egyptian ); A, (L5) [ou, S, L4 (L5) [w, to remain (<demotic g3, g‘ < gr); A prou, S, L4, L5 prw, winter; A, (L5), L6 jou; S, L4, L5 jw, to say.

1.2.5. */ew/ in stressed final syllables: A /o/, L /eu/, S /au/, e.g., A, L4 (ManiH=Manichaean Homilies) no, L neu, S nau, to see; A, L4 (ManiH) etmmo, L etmmeu, S etmmau, that (literally, which is there).

1.2.6.1. Typical of certain A texts (but not orthographically standardized throughout in A), insertion of a nasal bfore /t/ following /u:/ in open stressed syllables: A in I Clem. Herm., Prov. (minority), Jn. (minority) mounte, S, L, etc., and elsewher in A moute, to call; A in I Clem., Herm., Prov. (majority) nounte, S, L, etc., and elsewhere in A noute, God.

1.2.6.2. In contrast, the inserted nasal (sonant) is missing in A, L4, L5 (L6 oscillating) in the posttonic syllable: /mt/, A |amt, L4, L5 samt, L6 samnt, S somnt, three; A, L4, L5 hamt, L6 hamnt, S homnt, copper; A, L4, L6 camt†, S comnt†, to be stretched.

1.2.7. Anaptyctic vowel [ ] as well as /e/ is written as e at the end of a word following a closed syllable of the pattern /voiceless consonant + voiced consonant or son(or)ant/, by which A and L4 form a group distinct from L5, L6, e.g., A, L4 cwtme, S, (L4), L5, L6 cwtm, to hear; A, L4 tahme†, S tahm†, to be invited (versus S, A, L tahme, to invite you [f.]), A |tartre, L startr, S stortr, to be amazed. Naturally, te consonant can also be /’/, e.g., A ouaabe†, L4 ouabe †, S, L5, L6 ouaab †, to be holy (<w‘b). Note that while the representation in spelling of the anaptyctic vowel e in the configuration /consonant + son(or)ant/— not /laryngeal + son(or)ant/—is not standardized in L4, the spelling with e predominates by far. Consistent spelling occurs only in caune, to know (also P cooune). The anaptyctic vowel noted here by / / is phonemically relevant according to Hintze (1980). 1.3 Morphological Features.

1.3.1. Second present tense in A, B(!) af-, S, L ef- (in affirmative sentences, it is homophonic with the circumstantial verbal prefix).

1.3.2. Imperfect: A B(!) naf- (pe), S, L nef- (pe).

1.3.3. Second perfect: A naf-, S, L ntaf.

1.3.4. In the “etah group” (see 3.2.1.1), the relative morpheme of the perfect remains invariably etah- if the subject of the relative clause is identical to the antecedent in the main clause. 1.3.5. Homophony exists between the subject pronoun of the conjunctive tense and that of the first present, except the form of the first-person singular: first present S, A, L ]-; conjunctive A ta-, S, L (n)ta-.

1.3.6. Causative prefix A te-, S, L5, L6 tre- (in L4 the prefix is not standardized, te- occurring alongside tre-).

1.3.7. The qualitative form t- causative verbs in A almost entirely ends in -aeit.

1.3.8. The negation of the nominal sentence and the bipartite conjugation pattern in A is realized without n, i.e., only with en (S n ... an): A (L) efcwtme en, S enfcwtm an, while he does not hear; A etcwtme en, S ete nfcwtm an, who does not hear.

1.3.9. Definite article in A, including those preceding double consonants and expressions of time, are p-, t-, n- and (without e). 1.3.10. The second-person feminine singular possessive articles in A, B are pe-, te-, ne-, S pou-, tou-, nou-.

1.3.11. The third-person plural possessive articles in A, B are pou-, tou-, nou-, S peu-, teu-, neu-.

1.4 Lexical Features (Akhmimic Isoglosses).

1.4.1. Significant function words:

1.4.1.1. A aou, S, L auw, and.

.4.1.2. Directional preposition to or toward (<Egyptian r) A, L a-, S e-, including the adverbs containing the formative a/e, A, L abal, P abol, S ebol, out of; A a|oun, L ahoun, P aqoun, S ehoun, B eqoun, into.

1.4.1.3. Negative imperative A mn-, S (L) mpr.

1.4.1.4. Lexical and functional distinction between conditional particles formed from -pe and -je: A ei|pe, |pe, if (in the sense “supposing if it true that”) is to be differentiated from ei|je, |je, if (in the sense “as if it were”), the equivalent of esje in S. This distinction is found only in L6 of those texts belonging to the L group: eispe as opposed to eisje (Funk, 1985).

1.4.2. A number of nominal and verbal lexemes specific for A (cf. Till, 1928,pp. 276-78; Kasser, 1979a). While the vocabulary of Akhmimic has been treated throughout in Crum (1939), Kasser (1964), Westendorf (1977), a fully documented lexicon of Akhmimic has not yet been compiled.

1.5 Syntactic Features.

1.5.1. Connecting objects.

1.5.1.1. Nonreduction of the stressed vowel of the infinitive occurs also when the object is attached directly to it. In such cases, the infinitive retains the form of the status absolutus. In the bipartite conjugation pattern this construction is possible only with objects not modified by a determinative (Shisha-Halevy, 1976).

1.5.1.2. There is a tendency toward placing a pronominal indirect object (dative) before the direct object (accusative) without an accusative particle, in the case of ], to give; kto, to bring back; and jpo, to bring forth, which then take the respective forms te ne= accusative, kte ne= accusative, and jpe ne= accusative. 1.5.2. As in L, the affirmative final clause following je is almost always the second future (S third future, or futurum energicum). As is apparent in sections 1-5 above, not all of the described features are exclusively characteristic of A. Further, it is the totality (or, if not all features appear in a given text, the combination of sufficient individual features within a text) that assigns a document to the Akhmimic dialect.

The following sections will include a discussion of the phonemic inventory (2) and the conjugation system (3) of A, out of which the criteria for group classification (4) of Akhmimic texts will be derived. At the end (5), problems associated with a number of texts traditionally assigned to A will be treated.

2. The Phonemic Inventory of Akhmimic As is traditional, consonants and vowels will be treated separately in this inventory.

2.1 Consonants 2.1.1. Consonantal phonemes and graphemes are shown in Table 1. The eighteen consonantal phonemes of A correspond to seventeen graphemes. The laryngeal stop /’/ does not have its own sign but is expressed, or may be recognized, by the following:

graphic vowel doubling (or “breaking” of vowels) (e.g., kaa=f, to place him) syllabic structure /voiceless consonant + voiced consonant + / (cf. 1.2.7: ouaabe†, ouabe†, i.e., /wa’b( )/) vowel narrowing / / > /i:/ (r/, ri; cf. 1.2.3), / / > /u:/ (kw, kou; cf. 1.2.4) the postconsonantal first person singular suffix pronoun (kaa=t, to place me, i.e., /ka’t/) The phonetic articulation of / / b is disputed; Vergote assigns it to the bilabial category, but see his Grammaire copte (1973, Vol 1a, sec. 28).

g And d appear only in Greek loanwords. The replacement of k by g in the unstressed syllable ng (e.g., S mounk, moung, to form) is foreign to A. Except in Greek loanwords, z is not represented in A, since anz/be (or variously anc/be), school, is not attested in Akhmimic.

v, y, and , are, as in S and L, monographic characters of the phoneme combinations /p + h/, /t + h/, and /k + h/. Only in Bohairic do aspirants corresponding to /p/, /t/, /k/ occur. The symbols ' and x represent the phonemic combination /p+ s/ and /k + s/, e.g. 'ic, nine, and texe (S ta[ce, takce), footprint. The graphemes (o)u and (e)i also serve to indicate those vowels, [u:] and [i:], which are homorganic with the voiced spirants.

Note that in causative verbs such as jpio, to cause to be ashamed (< dj- pj), j may be interpreted in A as biphonemic /t + /; compare S, L jpo, A t|po, P t po, to bring forth (< dj- pr).

table 1. Consonants of Akhmimic

bilabial labio- dental pre-palatal post- laryngeal

dental palatal

Voiceless stops /t/ / /k/ /’/ /p/p t /,/c/j,[ k e.g.,aa Voiceless /f/ /s/ / / /x/ /h/ f c s | h spirants Voiced spirants /w/ (o)u / / b /j/ (e)i Nasals /m/ /n/ m n Lateral/ /l/l,/r/r vibrants based on: Vergote, 1973, Vol 1a, pp. 13, 15.

2.1.2. Phonetic alterations of consonants.

2.1.2.1. Assimilation: n before p > mp is not standardized, e.g., n-p appears side by side with m-p (|n p/ei as well as |m p/ei [not in EpAp.], en pe side by side with em pe [rare, I Clem.]); n before m > mm (rare), tmmouh, and we fill (Prov. 13:1); s before > , cse appears side by side with sse, it is fitting.

2.1.2.2. Dissimilation: mm > nm, mma= (status pronominalis of n-) A (standard) versus nma= (EpAp.); mp > np: npe (negative first perfect) A (standard) versus npe (EpAp.) 2.1.2.3. Partial depalatalization: k’ /c/ before s > ks, S, L nou[c, A noukc, to be wroth; S fo[c, A fakc, leap; S to[c, A ta[c, seat.

2.1.2.4 Metathesis: pwhc side by side with pwch, to bite; w|c side by side with wc|, to reap.

2.2 Vowels.

2.2.1 The vocalic phoneme inventory. A new and comprehensive system of Coptic vowel phonemes, especially modified for the separate dialects, has been proposed by Vergote (1973, Vol. 1a, sec. 60-65, A sec. 62). According to the phonemic system developed by Satzinger (1979),vocalic phonemes appear always as carriers of the stressed syllable. “All vocalic articulation outside of the stressed syllable may be explained as consonantal phonemes or as anaptyctic vowels which emerge according to specific rules” (ibid p. 344). While Satzinger’s system has the advantage of greater clarity, it ignores morphophonological connections. The result is that all morphemes that in the co-text do not function as the main stressed syllables remain unconsidered. The following summary relies on Vergote’s analysis but does not treat all possible phonetic articulations. [ ] as an anaptyctic vowel in closed syllables, with or without a sonorant, is not considered a phoneme (otherwise Hintze, 1980; cf. 1.2.7).

2.2.1.1. Short vowels:

/a/ a: bate, abomination; can, brother. Articulated before laryngeals and in stressed final position as [ ] (o), as in |o, thousand; teko, to destroy; |oop†, to be; ooh, moon; but ouaabe†, to be holy.

/e/ e: pese, half; ren, name. Articulated as [ ] before a sonorant concluding a syllable or before continuant, as in blje, shard, pottery; cbbe, to circumcise; tcto, to bring back.

/ / e: in unstressed initial, medial, and final sounds, but not as an anaptyctic vowel: eau, glory; [ec/[, rwme, man, as opposed to |wtbe /x t ( )/, to kill; tahme† /tahm /, to call you (fem. Sing), as opposed to tahme† /tah ( )/ to be invited.

2.2.1.2. Long vowels:

/ / i, ei: eine, to bring; mice, to bear; ji, to receive.

/ / /: k/me, Egypt; b/k†, going; n/tne, to you (pl.); /rp, wine. Articulated before laryngeals and at the ends of words as [i:]: mie, truth; ou|i, night; /h p/ h/p†, to be hidden, as opposed to /k ’/ kei†, to be laid.

Note that according to Vergote, / [ :] is an allophone of [i:] before and after sonorants ([er/[, n/tne).

/ / w: kwt, to build; pwt, to run; rwkh, to burn; thereto the allophone [u:] (ou) after /m/ and /n/ before laryngeals and when final, as in mou(n)te, to call; nou(n)te, God; kou, to place; prou, winter; but mmwtne.

/ / ou: |oun, interior; coure, thorn; tounoueiet= (abal), to reveal (this last is different in Vergote, 1973, Vol. 1a, sec. 56).

2.2.1.3. Contraction vowel:

/ew/ o, w only in final sounds after /m/ and /n/: mo, mw mother; no, to see; mmo, there. The written variations mo, mw do not indicate the neutralization of a supposed opposition *o versus , since in the A vocalic system [o] does not appear as a phoneme, but exists only as an allophone of /a/.

Note that w for o occurs occasionally as the final sound /-a’/ of the causative verb tasw, to increase (I Clem. 59:3, p. 77,9; Elias 33:9).

2.2.2. Vocalic opposition dependent upon syllabic structure (long-short opposition).

/ /, /a/: /k t/ kwt, to build, /katf/ katf, to build it (masc.). /p rx/ pwr|, to spread out, /parxf/ par|ef, to spread it (masc.) out. /m nk/ mounk, to form, /mankf/ mankef, to form it (masc.). /m r/ mour, to bind, /marf/ marf, to bind it (masc.). Before laryngeals: /x p / |wpe, to become. /xo’p/ |oop†, to be.

Note that in the status nominalis, /a/ before a sonorant is reduced to [ ]: mr, to bind someone/something: pr|-, to spread out someone/something.

/ /, /a/: /nh t=/ nhout=, to trust, /nhat/ nhat†, trusting (there are no further examples).

/ /, /e/: /n t ( )/ n/tne, to you (pl.), /nek/ nek, to you (sing. masc.). /c r c/ [er/[, hunter, /c rec / [ere[e, hunters.

/ /, /e/: /m rit/ , beloved (one), /m ret / merete, beloved ones). /m s / mice, to bear, /mestf/ mectf, to bear him. Before a sonorant, / / disappears and the sonorant becomes sonant and syllabic: /c n / [ine, to find, /c t f/ [ntef, to find him. Vocalic opposition is summarized in Table 2.

3. The Conjugation System The summary of the system is based on Polotsky (1960) and Funk (1981). Except in special instances (e.g. conjunctive), the form cited here is only the third-person masucline singular and the corresponding prenominal form (nom. = before nominal subject). The entire paradigm is not attested in all conjugations.

Unless specifically mentioned, the form is affirmative (neg. = negative). Every basic tense (abbreviated hereafter to “basic”) is followed (if attested) by its satellites, after “And”: cir. = circumstantial, rel. = relative, pret. = preterite, II = second tense. Forms between brackets [ . . .] are reconstituted from very similar forms (zero = no verbal prefix, no particle, etc.) 3.1 Bipartite Pattern:

3.1.1. Present (basic) f-, nom. zero. And circ. ef-, nom. e- or ere- (cf Polotsky, 1960, sec. 55); rel. et- or etf- resp., nom. ete- or etere-; pret. naf- . . . (pe; also nef-, Ex. 1:5, by influence of S?), nom. na- (Jn. 12,2; na- see also second perfect) or nare . . . (pe); II af- (with je, ja@n/u, I Clem. 48:2), nom. a- or ar- or are- (cf. Polotsky 1960, sec. 55).

3.1.2. Future (basic) fna-, nom. zero . . . na-. And circ. efna-, nom. e- . . . na-; rel. etna- or etfna-, nom. ete-. . .na-; pret. nafna-. . .pe (in the apodosis of the hypothetical form = Irrealis); II afna-, nom. a-. . . na- or are-. . . na- (second feminine singular ara-, I Clem. 20:7, cf. Polotsky, 1960, sec. 59; second masculine singular also ekna-, EpAp. 23,4; with je, jafna-, first plural janna-, I Clem. 58:1; second masculine singular also jekna-, Prv. 5:2; nom. ja. . . na-, Prv. 3:22, but je a-. . . na-, Prv. 3:10; nom. a(re)-. . . na-; cf. Polotsky, 1960, sec. 55).

3.2 Tripartite Pattern.

3.2.1 Tenses with special negation (if not II). Independent (sentence) conjugations.

3.2.1.1 Perfect (basic) af- (second feminine singular ar-), nom. a-; neg. mpf-, nom. mpe- (in EpAp. a dissimilative npf, nom. npe-). And cir. eaf-, nom. ea-; neg. empf-; rel. etaf-or etah-, nom. eta- (to the etah-group belong EpAp.; Elias; Ex 2:14, :11; I Clem.; Jn.; and Ost.; nevertheless, the inflected form eta= also appears in these texts under identical syntactic conditions; pentaf-, Prv. 18:22 and ntaf-, Elias 22:11, are to be considered as influenced by S); II naf-, nom. na- (Polotsky [1937 and 1944] is to be credited with the discovery of the Akhmimic second perfect; it is found with a derivative of the second degree only in the protasis of the hypothetical form = Irrealis, enaumahe, if they had gone, Prv. 2:20), also eaf- (I Clem. 31:3), [etaf-] (I Clem. 32:3, ne@ [e t/rou etauji eau aou aujice abal hitootou en / hitn nouhb/oue ndikaiocun/ etaueue alla hitm pefouw|e, “Now all of these are glorified and elevated, not through themselves nor through their works of righteousness which they did, but through his will”; cf. second perfect of B).

3.2.1.2 Completive (basic) (it is unknown whether the expected affirmative substitute *afouou ef- is simply not attested or actually does not exist in A; the passage cited in Crum, 1937, 373b, Hos. 13:2, is not pertinent: au(ou)ou, they perished, is a finite verb form in the first perfect; on the completive as a formal category, affirmative substitute S afouw ef-, has already . . ., neg. mpatf-, has not yet . . ., see Funk, 1981, pp. 191-94); neg. mpatf- (in EpAp. a dissimilative [npatf-], third plural npatou-), nom. mpate- (EpAp. [npate-] not attested). And circ. neg. empatf-, nom empate-.

3.2.1.3 Aorist (basic) |aref- (second plural) |aretetn-, Hg. 2:16), nom. |are-; neg. maf-, nom. mare- or ma- (cf.

Polotsky, 1960, sec. 55). And circ. e|aref-, nom. e|are-; neg. emaf-; rel et|aref-, nom. et|are-; neg. etemaf- (Prv.

14:23); pret. #ne|aref-$ third plural ne|arou-; II a|aref-(Elias 38:13), nom. a|are- (Prv. 11:10, 19:14, also e|are-, Prv. 19:15 by influence of S?) 3.2.1.4 Futurum energicum (or third future) (basic) afa-, nom. a-; neg. nef- (also enef-, Sir. 22,19), nom. ne-; with je, jafa-, nom. ja-; neg. #jnnef-$ (e.g. third feminine singular jnnec-, Lk. 18:5), nom. jnne- (Elias, I Clem.).

3.2.1.5 Imperative e.g., eno, see! (Ex. 4:13; for imperative with preformatives, see Till, 1928, sec. 147d); or infinitive; or ma + t-causative (no attestation among t-causative verbs of an imperative constructed simply from an infinitive); neg. mn- + infinitive; also mpwr a-, ApocSoph. 10,6. (this form is also common in L4 and L6 [Nag Hammadi]; mpwr hin/b, do not sleep (pl.), A Ost. A 10, is highly questionable, probably an erroneous writing of mpwr a-).

3.2.1.6 Causative imperative maref-, nom. mare-, absolute maran (Mi 4:2); neg. mntf-, nom. mnte-.

3.2.2 Tenses with neg. tm(n)-. Subordinate (clause) conjugations.

3.2.2.1 Conjunctive (singular first, second masc./fem., third masc./fem., plural first, second, third) ta-, k-, te-, f- (also ntf-, EpAp. 2,14); c-, tn-, tetn-, ce- (or cou-, EpAp. 6,7), nom. te-.

3.2.2.2. Future conjunctive taref- (second plural taretetn-) nom. tare-.

3.2.2.3. Temporal: Group I ntaref-, nom. ntare- (Jn.; Herm.; Prv.; ApocSoph. 11,2; Elias; EpAp. 3,14,19,10); Group II taref-, nom. tare- (I Clem., x., EpAp., MinProph.).

3.2.2.4. Limitative (“until . . .”) satf- (first singular sa]-), nom. sate- (also sante-, ApocSoph. 18,5).

3.2.2.5. First conditional afsa- (also efsa-, I Clem.; second plural asatetn-, also I Clem., but once esatetn-, I Clem. 63:2), nom. asa- (also esa-, I Clem.).

3.2.2.6. Second conditional #af-$ (only second plural atetn-, Zec. 6:15. The conjugation here referred to as second conditional is that termed “simple” conditional by Funk [1981], p.197], in contrast to his “expanded” conditional constructed with sa (S san). That the “protatic” efcwtm belongs to the tripartite conjugation was discovered by Shisha-Halevy, 1974. Affirmative forms are notoriously rare in literary texts. The protatic afcwtme does not occur in clause beginning with ei|pe and is only to be distinguished from the second present when it may be defined by its syntactic behavior as a (tripartite) subordinate conjugation. Neg. first singular aitm-, Mi. 3:8; second masculine aktmn-, EpAp. 40,12,14; third plural autm-,Ex. 4:8).

3.2.2.7. Causative infinitive -tf-, nom. -te- (second plural -tetetn-, Mal. 1,7, or -tetn-, Mal. 2:17; aptetn-, Prv.

24:23; also -tretetn-; concerning the second plural, see Polotsky, 1960, sec. 56; unique in A third masculine singular atre#f-$, Lk. 12:49, influenced by L).

4. Categorization Within Standard Akhmimic The Akhmimic literary texts exhibit a high degree of standardization. Disregarding sporadic deviations which may occur within the same texts, four criteria may be taken for an attempt at classification:

1. dissimilation m/m and m/p à n/m and n/p: (1.1.1) mmo, there à nmo, EpAp. 12:13, 28:14, 29:7; (1.1.2) mma= à nma=, EpAp. 1:5, 13; 19:1, 12; 17:7; (1.2.1) mpe-, neg. perf. à npe-, EpAp. (mpou-, 25:3); (1.2.2) mpate-, neg. completive à *npate, EpAp. (npatou-, 36:4) 2. the uninflected relative form of the perfect etah- (as opposed to eta=) 3. The temporal conjugation tare- (as opposed to ntare-) 4. The variant lexical appearance of the conditional particle “if (it happens that . . .)” |pe, ei|pe The criterion noted first pertains only to EpAp. This phonetic feature is supplemented by the fact that the assimilation n/pàm/p very rarely occurs at morpheme junctures in this text, the standard for a morpheme juncture in EpAp. being the unassimilated form (hn p/ei, etc.).

The distribution of the remaining distinctive forms may be represented in Table 3. Where etah- and eta=, tare- and ntare-, occur simultaneously, the second form is to be considered as unmarked.

A special group is constituted by I Clem., Ex., EpAp., moreover, is distinguished by dissimilation in its labial features. In I Clem., a distinction is still to be made between the conditional particle in Old Testament quotations and its form outside of such quotations (see footnote to Table 3). At the opposite end of the spectrum is Proverbs, which is the one Akhmimic text characterized only by the conditional particle |pe while lacking etah-, tare-, and ei|pe. John and the Minor Prophets assume a middle position: ei|pe occurs in both, but John also employes etah- and ntare-, while the text of the Minor Prophets uses eta= and tare-. The position of Ost., which employs etah- and ei|pe, remains uncertain because of the lack of a form of the temporal. The most strongly neutralized document is I Clem. (R) in which no distinctive form (etah-, tare-, or |pe) appears.

table 3.

etah- tare- ei|pe |pe I Clem.(R) - - + -I Clem. + + +(a) +(a) Ex. + + + -EpAp. + + + -Elias + + + -Jn. + - + -Prv. - - - + MinProph. - + + -Ost. + Temporal not attested + -aThe attestations of ei|pe and |pe are distributed in the Berliner Handschrift for I Clem. as follows:

ei|pe: p. 36, 18 (I Clem. 27:7); p. 69, 8 (I Clem. 54:2).

|pe: p. 11, 18 (I Clem. 8:4 = Is. 1:18); 13, 29 (I Clem. 10:5 = Gn. 13:16); p. 14, 5 (I Clem. 10:6 = Gn. 15:5); p. 23, 12 (I Clem. 16:16 = Ps. 21:9 lxx); p. 23, 14 (I Clem. 16:17).

The form |pe is used in four out of five passages that cite the Old Testament, whereas ei|pe occurs only outside of such quotations. The remaining form |pe on p. 23, 24 may have been attracted to the identical form on p. 23, 12. In I Clem., therefore, two levels of language can be recognized in the case of the conditional particle.

A summary by morphological characteristics appears in Figures 1 and 2. Both types of morphological classification lead to the same “extreme” groups: I Clem., EpAP., Ex., and Elias, on the one hand, and I Clem.(R) and Prv., on the other. Jn. And MinProph. have no distinguishing features in common with the other main groups, but form a class of their own.

5. Akhmimoid Texts 5.1. A number of literary and nonliterary texts (e.g., letters, magical texts) have traditionally been designated Akhmimic: 5.1.1 Literary texts. The Ascension of Isaiah (AscIs.) = Lacau, 1946. The Berlin Genesis fragment, P. 8773 (Gn. 1:18-2:5, fragmentary) = Leipoldt, 1904. Gal. 5:11-6:1 = Browne, 1979 (pp. 19-21). The Hymn [of Hierakas] = Lefort, 1939. Ps. 46:3-10 lxx, a pupil’s exercise on a wood tablet = Crum, 1934.

5.1.2 Letters. Listed by Simon, 1940, p.201, with footnotes 30-31.

5.1.3 Magical texts. Same as above, with footnote 32; Ernstedt, 1959, no. 70.

5.2. The literary texts AscIs., Berlin Genesis, and the Hymn were previously described by Kahle (1954, pp. 203-205) as “Akhmimic with Subakhmimic [that is, L] influence,” with AscIs. And Genesis forming a group of their own. The latter texts were shown by Kasser to be early forms of the dialect L (see especially Kasser, 1979b and 1982b, in which AscIs. And Genesis are referred to as i and i7, respectively; see also Funk, 1987). AscIs. and the Berlin Genesis have definitely to be eliminated from the body of Akhmimic texts, as does the Hymn of Hierakas, which corresponds more closely to i7 (and L) than to A (h> / / swpe: h> /x/ |n, without an anaptyctic vowel in the syllable /CR/ [= voiceless consonant + voiced cosonant or son(or)ant], second perfect nau-; vocalization of the stressed syllable as in L) or to Galatians (see Kasser and Satzinger, 1982).

5.3. Akhmimic Psalm 46, which is characterized by irregular orthography, is to be considered an early form of L rather than A, since none of the dialectical features of Akhmimic are distinctly marked: e.g., auw, and not aou (see 1.4.1.1); baab†, to be holy (for *ouaab†) and not baabe† (see 1.2.7); ouote (for ouhote*) and not ou(h)ate, (he is) fearful. In this connection, it is noteworthy that instead of the Akhmimic hnw(w)he, fear, the S (L) lexeme hote (hate) is employed.

5.4. The nonliterary textes were delineated earlier by Simon (1940) as Akhmimic with Sahidic influence, or As (for the letters) and “As vulgaire,” or vulgar Akhmimic with Sahidic influence (for the magical texts). While detailed evidence cannot be offered here, it should be pointed out that the Meletian letter Pap. 1921 (between 330 and 340 A.D.; ed. Crum, in Bell, 1924) clearly belongs to L, as does the letter from the John Rylands Library, no. 396, which was claimed by Crum (1909, p. viii) as an example of “a practically pure Akhmimic” text.

5.5. It may be concluded that the more or less Akhmimoid texts should no longer be counted with the corpus of texts written in Akhmimic dialect, not even with the mitigating addition of a small s, which is to indicate Sahidic influence. This means therefore, that the A dialect is only represented by literary texts (i.e., biblical, apocryphal, and patristic) and that, finally, “Akhmimic” is identical to “standard Akhmimic.” The Akhmimic texts are exclusively documents translated from Greek or Sahidic. Just for the most comprehensive texts (MinProph., Prv.) it has been shown that they represent interlinear versions of Sahidic (Till, 1927b, p. xxx; Böhlig, 1936, p. 35).