Gender, number and case

Coptic nouns have inherent grammatical gender but normally are not marked morphologically for this category, contrary to earlier stages of Egyptian, where the ending -.t unambiguously encoded the feminine grammatical gender.³⁴

The morphological structure of the Coptic nouns may also show the traces of a formal difference between masculine and feminine lexemes, but from a synchronic point of view, there is no indication to find out the gender of nouns like rwme ‘man’ and swše ‘field’

that have similar syllable-structure and vowel qualities and yet differ in gender specification (1). Lexemes ending in an atone -e are generally feminine, but this ending may be as well the consequence of the loss of a final consonant different from –t, cf. nTr >

noute ‘god’. (For more about the interrelationship between syllable-structure and gender, see Vergote’s classification of all types of Coptic noun lexemes according to their phonological (syllabic) constitution: Vergote 1973: §§96-99)

33 For the DP hypothesis, see: Abney (1987), and for a good summary thereof Bernstein (2003). Although I am in full agreement with this approach, I will use the more general term ‘noun phrase’ throughout the thesis for any kind of phrases whose lexical nucleus is a noun.

34 There is some evidence that the ending -.t was much longer kept in orthography (as a kind of graphical gender-marker) than spelled out in spoken language.

The grammatical category of gender usually becomes visible only by means of the agreeing determiners such as, for instance, the definite article (2), or cross-reference performed by personal pronouns.

(1) rwme swše

‘man’ ‘field’

(2) p-rwme t-swše

DEF.SG.M-man DEF.SG.F-field

‘the man’ ‘the field’

There are, however, a few nouns that have two related forms corresponding to male and female biological sex, e. g. son ~ swne ‘brother/sister’ Rro ~ Rrw ‘king/queen’ (for a list thereof, see inter alia Layton 2000: §107; Reintges 2004: 52-53). Others are underspecified for this feature and can co-occur with both sets of determiners, that is to say, they can denote both masculine and feminine members of the set described by the common noun: e.g. xMxal ‘servant/maid’ (See more examples at Layton 2000: §106).³⁵

Normally, Coptic nouns are not marked morphologically for number either. The singularity and plurality of a noun phrase – like in the case of gender specification – become visible by means of determiners (see exx. 3 and 4), or cross-reference.

(3) p-rwme N-rwme

DEF.SG.M-man DEF.PL-man

‘the man’ ‘the men’

(4) t-swše N-swše

DEF.SG.F-field DEF.PL-field

‘the field’ ‘the fields’

Nevertheless, there is a relatively large set of exceptional nouns exhibiting a remnant plural form, e. g. son ~ snhu ‘brother/brothers’ (Layton 2000: §108(b); see also Vergote 1969;

Vergote 1983: §§113-114 on number of Coptic nouns in general Vergote 1983: §§115-120

35 I must note that there is fundamental difference between the view of Ariel Shisha-Halevy and mine concerning gender specification. Shisha-Halevy (1986: Chapter 5 and 2007: 341) when arguing for the nuclearity of the determiner (determinator in his terminology) claims that lexemes do not have an inherent gender in Coptic, and “the article is the concord motivant” (1986: §5.1.1. But see the criticism of this chapter by Polotsky 1989). I would also argue for the nuclear function of determiners, but from a syntactic

(structural) point of view supporting the so called DP-hypothesis (see note 33), according to which the projection of a noun (i.e. the noun phrase) is part of a larger functional complex (the DP), in which the determiner has a central role and D-head has a nominal complement. But the gender of the phrase as a whole is inherited from the noun lexeme. As for the gender feature that appears on the determiners, it is the result of a mere agreement operation, just the same way as other modifiers (adjectives, numerals) agree with the head noun in gender, independently of determination.

with dialectal variations). Remnant plurals correspond to three different morphological patterns, as it is summarized by Chris Reintges (2004: 53-54): they are formed (i) by adding a plural suffix (-au , -hu , -oou, -ooue, etc.),³⁶ (ii) by the alteration of the syllabic structure (so called broken plurals, e.g. eiwt ~ eiote ‘father/fathers’), or (iii) by the combination of these two strategies (e.g. xwb ~ xbhue ‘thing/things’). As it is already pointed out by Vergote (1983: §114), the remnant plural forms are used in parallel with the synchronically more regular singular forms to express plurality (5). In Crum’s dictionary (1939) these nouns are marked with the “& sg as pl” annotation.³⁷

(5) p-son ne-snhu N-son

DEF.SG.M-brother DEF.PL-brother DEF.PL-brother

‘the brother’ ‘the brothers’ ‘the brothers’

In Coptic, there is no inflection for case: the grammatical function of a noun (or better a noun phrase) is indicated by distinct sentence positions and functional morphemes.³⁸

Greek nouns were borrowed in their singular, nominative case form with their original gender if they were masculine or feminine. Neuter Greek nouns were re-classified as masculine ones (Girgis 1971-73: §91; Hopfner 1918: 14). As I will argue later in this chapter in a more detailed way, loanword integration was obviously conditioned by inner Egyptian structural constraints. Accordingly, the integration strategy of Greek nouns perfectly fits to the conditions of the Coptic grammatical system lacking morphological case distinction on the one hand, and making use of two grammatical genders only on the other. As for the expression of plurality, Coptic needed only a singular lexical entry of the foreign word, leaving the work of plural-forming to the syntax.

Occasionally, Greek nouns were adopted apparently in accusative or genitive case, but this phenomenon may be equally explained by language change in certain varieties of Greek itself, where a “tendency of simplification and unification” can be observed in the

36 Reintges mentions –ooue only, while Layton (2000: §108(b)) lists more suffixes, seemingly all that have the element -u in common. However, he fails to mention the alternation (-o ~ -oi) attested in xLlo ~ xLloi., and lists another ending type -ate, -ote which is probably not a proper suffix but part of the stem after having been reshaped in a plural form (see the second group of Reintges). Vergote (1983: §114) also mentions the use of the collective ending -h that sometimes can serve as a plural marker (e.g. toou ~ toueih ‘mountain/mountains’, moou ~ moueih ‘water/waters’.

37 For gender and number of the noun in general, see also: Stern (1880: §§199-207), Steindorff (1951: §§92-94), Till (1961: §75-86), Vergote (1973: §81) and Lambdin (1983: 1.1-1.2). For the various appearances of the article in the examples, see Chapter 3 on determination.

38 The only structural case is the nominative (for a possible formal syntactic analysis of its structural position, see Reintges 2001: 100), the object of the verb can be expressed either by a prepositional phrase, or by its forming a construct state with the verb (for an analysis of the ‘differential object marking’ in Coptic, see Engsheden (2006 and 2008)). All other relations within the sentence, i.e. the relation between the verb and its arguments, are expressed by prepositions.

nominal morphology, such as the fusion of certain declension classes or the tendency of re-categorization of inanimate nouns to be neuter instead of masculine. It might be the case that Coptic borrowed these “new” forms directly from Greek, thus there is no need for looking for an inner Egyptian logic in order to account for such deviant or nonstandard forms (cf. Girgis 1971-73: §§80-81; for further cases of genitive loans: §§82-85; for Greek plural form used as singulars in Coptic: §§86-89).³⁹