This paper consists of two parts. The first (§1 etc.) is a special commented mini-chrestomathy: I present grammatically classified Shenoutean passages, briefly commenting on their structure and analytic implications. Thereafter (§2 etc., “postliminaries”), I will share with the reader, at some length, reflections on issues arising from consideration of these texts, beginning with a discussion of the meaning and significance of the anacoluthia concept, in a language such as Coptic.

Fifteen years on, we now have the second volume of fourth-century documentary texts – mainly letters – from Kellis (present-day Ismant el-Kharab, in the Dakhleh oasis), editing seventy-five new documents, added to the forty-five published in 1999.1 In fact, these are “two halves of a single work” (p.4).Given the syntactical and dialectal peculiarities of “dialect L*”, we by now have a corpus well worthy of its own systemic grammar, with the impressive second installment also serving as control, to evaluate the impressions given by the first.2 Here too we have an admirable edition, textual apparatus, translation and commentary – this reviewer would be grateful for a more intensive grammatical annotation. The edition, classified mainly by provenance and sender, follows an introduction (a brief one; that of Kellis I serves both volumes), including dating of texts (p.5f.), and is followed by exhaustive reasoned indices.I shall dwell here briefly on syntactical highlights, remarkable or striking constructions, taking up a few points of grammar, as well as a few critical comments on analysis and translation. This elegant work and the rare privilege of “discovering” a “new”, extensively documented dialect, and at the same time a rich trove of grammatical features in so early a source are any linguist’s and philologist’s wistful vision – to say nothing of such enviable collaboration of leading scholars.

The paper assigns, in a ‘pointillistic’ structural profile, narrative functions to dyma and dyna, formal presentatives, in syntactic detail and macrosyntactic patterning, on the database of Kate Roberts’s short stories and novellas.1 The extensive distribution and rich functional range of these elements matches their formal complexity and narratological significance. This presentative pair, expanded by verbal, substantival or pronominal presentates, form six narrative tenses, distinct formally and functionally, in complex interplay with their environment.In fact, however, dyma and dyna comprise doubly two homonyms: dyma/dyna presentatives, and dyma/dyna referential pronouns, typically rhematic or focal.Following a descriptive breakdown of the syntactic properties of the presentatives, the Presentative Narrative Tenses (PNTs) I to VI are discussed.Functionally striking and statistically prevalent is (PNT I) # dyma + noun phrase/personal pronoun + yn-converb2#, where we encounter two homonymous sub-tenses: the first with specific scenic or theatrical (‘dramatic’, narratologically scene-setting) semantics; the second non-scenic, but tagmemically functional. It is noteworthy that the entire presentative clause is high-level, narratologically rhematic or focal to the preceding text: it contains the key event. The presentative signals immediacy between narrator, reader and narrated character.Two presentative narrative tenses are non-verbal: adverbial presentates (dramatic presentation of motion) and scenic presentation of nouns.Another major issue treated here concerns the anaphoric pronouns dyna and dyma, rhematic in Nominal Sentence and Cleft Sentence patterns.

Following an early brief attempt at a formal-and-functional resolution of the pre-verbal elements fe- and mi- in narrative (Shisha-Halevy 1995: Excurse II), these two discourse-function converters are examined again, as part of a comprehensive narrative-grammatical study of Kate Roberts’s fiction.

This paper ponders analytically the Circumstantial and Relative Conversions in Coptic (CC, RC), seen especially as satellital, in the [nucleus — satellite (expansion)] dependences. I wish to present here some progressions of thought about central topics and vexed questions concerning the CC, which is arguably among the “most Egyptian” of Coptic grammatical features, familiar as they may be, as a basis for a typological profile. The issues considered are presented in sequences which, I believe, are pertinent, with connections that appear to me instructive. The examples given are usually minimal and representative only. The hidden agenda of this paper aims, inter alia, at demonstrating the descriptive effectiveness of structural syntactic analysis. I submit that we do not yet properly understand the CC, and contest the conventional way of approaching it. The CC differs interestingly from the other conversion. Not only is it the earliest of converters in Egyptian diachrony — it is the earliest “completely formed” converter. Its structural tension with the RC is an informing feature of Coptic syntax. (The RC is but half-way to converterhood, ⲛⲉ- is arguably not a converter at all, deposed by Polotsky in the 1987 Grundlagen from converterhood, and the Focalizing Conversion is of restricted distribution, morphologically overlapping the RC and the CC and (in Bohairic at least), giving sometimes (in the Preterite) impression of a base-conjugation form.

Speaking metaphorically, the array of distinctive linguistic traits is a portrait or profile, not a check-list or catalogue. This means that we are considering, not a list but systemic co-occurrence and/or combination and/or hierarchy of features that is distinctive. This, however, is difficult or near-impossible to depict in a simple presentation, and in the following lines I will also particularize or list after all. Twenty-five years ago, in the Coptic Grammatical Categories (Rome, 1986), I attempted to present a system of systems, focusing on adverbials, that might serve as basis for identification. It goes without saying, that a precise, sensitive high-delicacy descriptive work is a sine qua non in authorship studies, with the central query being to what extent we can detect the typical, and to what extent can the typical be misleading. Authorship statements are not infallible,1 and can only be as confident as the linguistic description is sensitive and broad-based. The difficulty of authorship proof in a dead language, and, besides, one which we are still trying to get the measure of, should not be underestimated. And yet, ideally and with careful and considered application, I would suggest linguistic attribution is even more conclusive than explicit “philological” one.Not unlike forensics in general, the logic of cumulativity is based on systemic configurativity. (This logic is exponential: the more numerous and high-ranking the symptoms, the exponentially higher the certainty of attribution.) Few of the features here presented by themselves are exclusively Shenoutean, but any of them in combination with others are conclusively so. The number of features “necessary” for establishing a Shenoutean “identikit” depends on their critical value, which is scalar (lexical features differ in indicativity from phraseology, from morphology, micro- and macro-syntax); on the other hand, the greater the number of traits, the more confident the attribution. An instance of a very high criterion is the rich syntactic range of quotation manipulations; low-value traits are morphological features, including morphophonological ones such as “Akhmimoid” (or Southern) ⲁ for “normal Sahidic” ⲉ, or unreduced prenominal infinitive allomorphs (e.g. ⲟⲩⲱⲙ-), or unreduced thematic pronouns in the Interlocutive Nominal Sentence (e.g. ⲛⲧⲱⲧⲛ-). The theoretical aspects of authorship studies (familiar especially from study of Biblical corpuses), as against the practical aspect, on which I shall focus here, regards internal relations, such as those between ϣⲁⲧⲛⲁⲩ and ϣⲁⲛⲧⲉ-ⲟⲩ ϣⲱⲡⲉ, or between the jussives ⲙⲁⲣⲉϥ- and ⲉϥⲛⲁ, the positions of ⲉⲧⲃⲉⲟⲩ and ⲛⲁϣ ⲛϩⲉ, also such issues and calculi as the cumulative probability of a specific authorship, the absence of occurrence as an identifying factor, statistical features and scales of typicality. The practical angle concerns features occurring in the texts, and aims at assessing them cumulatively, with rising confidence of attribution. While less-than-typical characteristics are ubiquitous, they are usually interspersed with features of diacritical value. A practical principle, of the type of “the dog that did not bark at night”, would conclude non-Shenoutean authorship from a consistent and total absence (in a text of considerable length) of Shenoutean traits, or absence in Shenoute of specific features (cf. Crum, Dictionary 544a, ϣⲁⲓ “festival” not found in Shenoute). Of course, this “identity kit” is as dynamic as it is systemic, in the sense that new texts introduced into the canon, texts removed from the canon, new forms and interpretations, all may modify the critical syndrome.The stylistic tones of Shenoute’s work are familiar, mostly summed-up as passionate rhetoric, and have been pointed out in various, often (but not always) more or less derogatory descriptions, since Johannes Leipoldt, De Lacy O’Leary, K. H. Kuhn and Bell. This biased and impressionistic view of Shenoute at his most typical, which, however, is of limited use in less than typical, less rhetorical, texts or passages in texts, is simplistic;, Shenoute, who can be quite pedestrian, occasionally surprises us with gentle, emotional, even poetic turns as well as register changes. But his consummate rhetorical craftsmanship is much more sophisticated than that, and his authorial fingerprint accordingly very complicated.

On the following pages, we propose to present and discuss our database for some features of Coptic, deviating from the “canonical” picture as seen in the grammars, from L. Stern’s to B. Layton’s, and in the grammatical literature generally. These are Lesefrüchte, and the treatise more of a work-note than a conclusive and systematic discussion; it is meant to attract attention, but also a description of environment and function. A historical dimension is of the essence in these cases, and will be addressed in some detail, for a diachronic cycle may here be in evidence, and an appeal to pre-Coptic Egyptian linguists is envisaged; also, a methodological perspective – pointing out the flimsiness of our comprehension of Coptic grammar, as well as its “canonical” nature, which is the main reason for the impulse for editorial condemnation and emendation. Finally, this essay is an homage to the syntactical sensitivity and analytic intelligence of W. E. Crum, not to be eclipsed by his lexicographical genius. In Ludwig Stern’s words, Coptic cannot easily be “erlernt”: of its terra incognita patches, our notes pick one verbal, one non-verbal feature.

In the course of an exploratory study of ‘Shenoute’s rhetorical syntax’, a comprehensive investigation of the syntactic poetics of rhetorical complexes (the grammatical high-order signifiers, for which the signified ‘added-value’ is ‘rhetoricity’), I have encountered a textemic set of rhetorical narrative structures which, I believe, provides important new insights on the grammatical nature, texture, and properties of narrative in general.2 In this, a pilot study, I shall offer a brief overview of this set, attempting a cursory formal-and-functional description of the individual textemes, and present representative and selective token documentation (usually not more than a single example for each category; more, sometimes many more are attested).3 Statements made here have no claim to be universally valid, but are meant to describe the Coptic situation. For reasons of space, I have left out most of the secondary literature, whether literary, narratological or linguistic stricto sensu; the types discussed are selective, out of the numerous types in my files: I wish here to draw attention to this rich vein of syntactical and stylistic data, to their variety and intricacy, to offer a provisional typology and observations on distinctive grammatical properties, and perhaps to stimulate debate. The writings of Shenoute (c. AD 348–c. AD 465) are the most extensive authentic (i.e. untranslated) corpus of Sahidic Coptic and Coptic in general, a corpus which (although always appreciated for its high stylistic, literary, and rhetorical sophistication), has in the last decade of the last century gained in scholarly attention, and is currently being re-edited and retranslated as a joint international project.

The converb, in its least specific and sharp resolution, is used to mean ‘adverbial verb form’, or ‘verbal adverb’ (see the subtitle of Haspelmath and König 1995). Mostly and for long it has been known, in the description of various languages, as ‘gerund’. Definition of the converb reveal an underlying blurredness: Haspelmath (1995: 3ff.): ‘Non-finite verb-form whose main function is to mark adverbial subordination’; Nedjalkov’s (1995) is more sophisticated: ‘a verb-form which depends syntactically on another verb-form but is not its syntactic actant, that is does not realize its semantic valences’. (This is surely unsatisfactory, for the converb is arguably actantial in cases like ‘start walking’). Probably the worst is the definition in Himmelmann and Schultze-Berndt (2005: 60): ‘we use the term converb for ‘participles’ which are used primarily as adjuncts’. As Grønbech (1979: 35) says of Turkic postpositions and gerundial forms, the converbs are ‘fluid and hard to hold on to’, which for a ‘cross-linguistic valid category’ (Haspelmath and König 1995, in which see Haspelmaths’s and König’s own contributions), is not an ideal condition. […]

The term and its diffusion. The converb, in its vaguest and least critical, also least specific resolution - cf. the notorious conceptual muddle involving -ing forms and constructions in English - is used as meaning “adverbial verb form”, or “verbal adverb”; see the subtitle of Haspelmath and König (eds.) 1995.; mostly and for long it has been known as “gerund”. Definitions reveal the underlying blurredness: Haspelmath (1995:3ff.): “Non-finite verb-form whole main function is to mark adverbial subordination”; Nedjalkov’s (in Nedialkov 1995) is more sophisticated: “a verb-form which depends syntactically on another verb-form but is not its syntactic actant, that is does not realize its semantic valences”: this is surely unsatisfactory, for the converb is arguably actantial in cases like “start walking”. Probably the worst is the definition in Himmelmann and Schultze-Berndt (eds.), 2005:60 “we use the term converb for ‘participles’ which are used primarily as adjuncts”. As Grønbech 1979:35 says of Turkic postpositions and gerundial forms, the converbs are “fluid and hard to hold on to”, which, for a “cross-linguistically valid category” (the title of Haspelmath and König (eds.) 1995, in which see Haspelmath’s and König’s own contributions), is not an ideal condition.

Published in: Chronique d’Egypte, 84, pp. 136–152.This work of Jean Winand’s aims at providing an account of Tense and Aspect (or rather Aspect and Tense) systems in Egyptian: this (notwithstanding the focus on Old and Middle Egyptian, with Late Egyptian rather thinly treated, and Demotic and Coptic virtually absent) is a staggeringly ambitious undertaking. It implies a confidence in our comprehension of Egyptian, synchronic and diachronic, which this reviewer must admire, but cannot share. And yet, it is almost a blessing that the Later Egyptian systems are only lightly touched upon, for this renders the in-depth treatment of OE and ME virtually monographic, which would be hardly feasible for the whole of Egyptian history — all the more so, since the joints or seams between “successive” phases are fictive and indeed fallacious. On the other hand, one would wish for an extended application of the author’s hypotheses to LE (and Demotic), for the cryptic nature of the earlier phases of Egyptian renders any judgement made regarding their imponderables both subjective and irrefutable. It is easy to pass speculation on O/ME as descriptive statement, which would never do in the more “transparent” later phases. Be that as it may, the reader gets occasionally the eerie, unsettling feeling that it is a transcendental, panchronic (or panoramic) Egyptian that is here under typological scrutiny. (I cannot see, for instance, the soundness of a combined statement [p. 197] on the Stative and sḏm.n.f forms on the basis of Sinuhe and the Late Ramesside Letters).Uploading this article to the Internet Archive is done with the author’s permission and by his request.

Published in: A. Giewekemeyer, ed. Liber Amicorum: Jürgen Horn zum Dank. Göttingen. Göttingen: Göttingen Seminar für Ägyptologie und Koptologie der Universität, pp. 113–129.The following are notes taken in the course of an ongoing long-term study on “Shenoute’s Rhetorical Syntax”, mapping the grammatical (mainly syntactic) poetics of Shenoute’s published and unpublished work, with a focus on rhetorical value and effect of forms and constructions. I wish to present here textual and descriptive materials on some not unimportant rhetorical devices which, beyond having rhetorical functions (such as lusis), also signal tonal and emotional nuances, in particular bearing on Shenoute’s often underplayed sardonic sense of humour, irony and sarcasm. Note that this is a mere outline: the observations basic, brief and often laconic, discursive and occasionally repetitive, the theses tentative and often, I fear, half-baked, the bibliographical referencing minimal, the illustration no more than representative.Uploading this article to the Internet Archive is done with the author’s permission and by his request.

Barring the Nominal Sentence, Egyptian grammatical study of the heroic age, from Old Egyptian to Middle Egyptian, from K. Sethe to H. J. Polotsky, was mainly concerned with the verbal system and verb syntax. What has been stated about nominal syntax beyond the very basics would not exceed, all grammarians told, a few pages of print and very little individual variation based on real original research. One cannot help feeling this is due to the absence of “orthodox” affixed articles, as if these are anchoring points for syntactical observation of the noun. (Terminologically, of course, “articulum”, Greek ἄρθρον, means a metaphoric “linking joint” — Gelenk — revealing no less than a realization of its prime environmental role). And yet, the absence of bona-fide definite and indefinite articles in written Old and Middle Egyptian, somewhat like the absence of graphemic notation of vowels, which, in H. J. Polotsky’s conception of verbal category, sets us free from la superstition de la forme (De Boer) and encourages us to resort to the structural definition of linguistic identity, this “deprivation” too must be taken as a blessing in disguise: it forces our attention off the noun — temptingly “adequately” translatable in isolation into a European-style language — onto its environment, where much signalling information regarding (non-)specificity and (non-)particularity is to be found. The difficulty of seeing clearly in the matter of noun determination stems inter alia from looking for a “copy” correlation with what we have grown used to feel as Indo-European (or rather European) articles; but also from the generally implicitly accepted dichotomy of grammar and lexicon, a dichotomy more leaking than most other linguistic models; and especially from our being so to speak mesmerized by the article(s), which impairs our peripheral vision (yet another metaphor) and obscures our view of co-signals of determination. Here, incidentally, the trap of ethnocentricity is particularly ready for the unwary, the more so since it is, by easy terminological transference, the article — where present — that is conceived of as “definite” or “indefinite”, and not the noun and its environment. Moreover, in ignoring environmental determination, the typological significance of a definite article (and as a matter of fact, the article is but marginal in the overall phenomenon) can easily be exaggerated.The commonly — indeed conventionally — erroneous synchronic view of article function can also flaw a satisfactory resolution of article-less states. For instance, the proportion of (macro)syntactic — anaphorical or cataphorical — and exophorical or intrinsic functioning of the articles may vary dramatically between narrative, dialogic, expositive, legal or ritual textemes. Finally, the continuity fallacy, of chronologically successive written phases seen as real succession in linguistic diachrony, distorts our view of article evolution.