Bryn Mawr Classical Review 1998.10.03
Egbert J. Bakker (ed.), Grammar as Interpretation: Greek Literature in its Linguistic Contexts. Leiden: Brill, 1997. Pp. 262. ISBN 90-04-10730-4. $108.00.
Reviewed by Stephen Colvin, Yale University
Word count: 1984 words
This collection of essays is loosely themed around the notion that familiarity with the linguistic approach now generally known as pragmatics can assist in the critical interpretation of ancient Greek texts. The methods and results are demonstrated by application to a series of practical problems in ancient Greek authors, leaving the reader to infer general principles from a succession of case-studies, which, pace Plato, is easier than trying to digest pages of theoretical definition and explanation; and it is easier to gauge the possibilities and limitations of a new approach by reference to material with which one is familiar. Although some of the methods may be new and the conclusions occasionally startling -- I shall mention some trivial reservations below -- there is much that will be familiar to the alert classicist, who is by now at home with narratology and the "new rhetoric" in the explication of ancient texts; as the editor puts it in his introduction, the essays reflect the shift in interest from the "what" to the "how?" in the production of meaning. As a sampler the collection works well, offering a variety of areas of Greek language and literature for study; and for this purpose the fact that such collections invariably include texts in which a particular reader is uninterested is less important, since the method of investigation will still be of interest.
There are just six essays in the volume, which makes it possible to give a brief overview of each contribution. The house-style has an English translation after each extract; there is an Index Locorum and a General Index at the back of the book. Egbert Bakker has an elegant introduction which includes a brief summary of the six contributions. His own essay constitutes Chapter 1, and is entitled "Verbal Aspect and Mimetic Description in Thucydides" (pp. 7-54). The essay concerns Thucydides' use of the aorist and imperfect tenses: Bakker argues that this can be connected with the authorial perspective at a given point in the narrative. There is much in the piece that is deeply interesting and insightful, but I had some difficulty swallowing his principal hypothesis, which is that a new layer of pragmatic analysis must be superimposed to understand this admittedly complex alternation of narrative tenses. He begins (sect. 2) by rehearsing the standard explanations for the alternation, which boil down to (i) the traditional punctual (or closed) action (= aor.) versus durative (or open) action (= impf.), and (ii) the increasingly accepted distinction between "background" verbs (impf.) which describe context, motivating circumstances, etc., and "foreground" verbs (aor.) which move the narrative forward; not unreasonably, Bakker describes this as an extension of the traditional view from the level of the single sentence to the level of discourse. He proceeds (sect. 3) to draw an interesting and valid distinction between two types of language use: the "discourse of the knower", an abnormal mode of communication, much beloved by traditional grammar, in which facts are stated; and the "discourse of the observer", a mode of communication which is closer to everyday language use. This is by way of preparation for the argument that Thuc. frequently aims at pretended immediacy, the presentation of events as if they are seen on the spot. No one, I suppose, would deny the ability of Thuc. to grip his audience with his powers of description: the question is whether this can be "grammaticalized" by reference to the use of the aor. and impf. tenses. There is a further distinction to be drawn, however, before one can get to grips with the hypothesis. Bakker posits two types of discourse in Thucydides, viz. the diegetic (sect. 4) and the mimetic (sect. 5) modes: the explanation of an aor. or an impf. will differ according to which of these two categories it falls in. In diegetic mode the tenses will be explained along the lines of "standard" explanations (i) and (ii) above. In mimetic mode, however, the impf. is to be interpreted as a marker of "displaced immediacy", i.e. the internal point of view of the characters rather than the authorial voice (which presents "facts" in diegetic mode). Thus, for example, at Th. 7.71.4 αἰεὶ γὰρ παρ' ὀλίγον ἢ διέφευγον ἢ ἀπώλλυντο ("they were constantly on the verge of being saved or lost") the non-completeness of the action is a natural consequence of the internal perspective adopted by the author. In this mode an aor. will be used for authorial intrusion, or for various other, slightly complicated reasons. This is a densely-written and thought-provoking piece; my problem is the level of complication that it adds to the interpretation, for it is hardly repaid by the benefits when one has taken into account all the special categories and explanations which are needed to bring recalcitrant verbs into line, the theory begins to look (in the words of another contributor to the volume) unfalsifiable. I have no doubt that Thuc. employs a complex narratological strategy; but it seems to me that the interplay of past tenses in his writing can be accounted for within existing categories.
Chapter 2 ("Interpreting Adjective Position in Herodotus," pp. 55-76) by Helma Dik is a persuasive paper, with the virtues of brevity, clarity, and an admirable internal organization. Dik argues that the traditional, semantically-based view of adjective-noun order is mistaken, and she proposes a pragmatically-based correction. This would perhaps be the best paper for a novice or casual browser to turn to for an impression of the approach to text and language that the collection (and in a wider sense the discipline) adopts. Dik notes that a widely-held view on the subject divides adjectives into two classes, determining and qualifying, and then predicts that a determining adj. will follow its noun and a qualifying adj. will precede. Her thesis, which is based on data from Herodotus, is that noun-adjective order is the default ordering in Greek, and that a departure from this is a marked usage which needs to be accounted for in pragmatic or contextual terms.
Chapter 3 ("Towards a Rhetoric of Ancient Scientific Discourse: Some Formal Characteristics of Greek Medical and Philosophical Texts," pp. 77-129) by Philip J. van der Eijk comprises twelve sections, which, as the title suggests, set out questions and problems thrown up by technical or non-literary texts from the ancient world without attempting to tackle any of them in great depth (in this context there is a strong temptation to protect oneself with constant scare quotes around key terms like "literary" but I shall desist for the sake of typographic simplicity: caveat lector). The paper nevertheless raises some important issues, at the heart of which lies the knotty question whether there can be a text which is not in some sense aesthetic. This is a question which connects in interesting ways with a host of other problems, such as who the intended audience of an ancient scientific text were, and what the relationship between the conventions of a technical text and those of, say, a Pindaric ode (e.g. first-person statements) was. His observation that scientific discourse, so far from representing a pure seeking after truth, comes from the context of a highly competitive market-place will strike a familiar chord in the modern academic reader.
In Chapter 4 C. M. J. Sicking and P. Stork tackle "The Grammar of the So-Called Historical Present in Ancient Greek" (pp. 131-168). This paper picks up and amplifies a strand of thought running throughout the volume, namely that the time has come to replace some of the traditional terminology of textual criticism, whereby anomalous usages might be described as "vivid", "dramatic", etc., with an interpretative apparatus more firmly based on wider contextual or narratological analysis (the notion being that the usages in question are anomalous only within the bounds of the sentence, the locus and limit of traditional Western philology). The authors argue that the historical present (HP) is to be regarded as a grammatical rather than a stylistic phenomenon. Since the appearance of HP in texts is notoriously unpredictable this in itself is an inviting hypothesis: the question is whether a convincing replacement for the catch-all "vivid" explanation can be found. In a tightly-argued paper Sicking and Stork propose a focussing function for the present tense within a historical narrative: HP may be used to highlight a superstructure within a narrative -- either picking out the essentials of an argument or possibly distinguishing a separate strand from the main body of a narrative. It thus constitutes an additional structuring device for the narrator over and beyond the interplay of aorist and imperfect (the authors take it for granted that readers will be happy or at least familiar with the focus function of the aor.). An obvious objection would be that if HP can pick out both the essential structure of a narrative and an excursus (Solon and Croesus, for example, in Hdt. 1), then the concept is too lax to be useful. I think, in fact, that this might be dealt with by more detailed attention to the different types of narrative which employ HP (it is not unreasonable to suppose that structuring devices might work differently in Herodotus, Lysias, and a tragic messenger speech), but readers will judge for themselves.
In Chapter 5 ("Figures of Speech and their Lookalikes: Two Further Exercises in the Pragmatics of the Greek Sentence," pp. 169-214) S. R. Slings, continuing earlier work on the subject, argues that certain features of Greek word-order and grammar in classical texts which are generally considered anomalous (and are thus interpreted as literary figures, or dismissed as corrupt) are in fact reflections of spoken Greek in a (still developing) literary language. His paper falls into two parts: in the first he deals with anaphora, antithesis and chiasmus, and in the second he turns his attention to anacoluthon in Plato. The author argues that the key to understanding devices such as anaphora is the desire to spread information evenly and digestibly over clauses on the part of the speaker (writer): if new information fails to appear, then (and only then) the ordering device may be considered a rhetorical or literary figure. In the second part he distinguishes two types of anacoluthon, but makes use of a similar notion of cognitive overload in accounting for them: (i) traditional syntactic bugbears such as the hanging nominative may in fact be an advance warning of the "theme" of the sentence which, as a signal to the listener (reader), is deliberately unintegrated into the grammar of the predicate, and (ii) structural complexity may lead to a "downshift" in the grammar of a long sentence (the thought that dumbing-down began with Plato is comforting).
The final paper in the collection, by Gerry C. Wakker ("Modal Particles and Different Points of View in Herodotus and Thucydides," pp. 215-250), is a useful overview of the complexities of oratio obliqua (where is the authorial voice?) followed by an analysis of the use of the attitude-bearing particles τοι, ἦ, μήν, που, ἄρα, δή in the context of reported speech. The author discusses what she rightly calls a "sliding scale" of authorial commitment to the reported words, ranging from the obvious (οἶδα guarantees the truth of the proposition from the reporter's perspective) to the ambiguous (whether ἄρα belongs to the reporter or to the embedded speech).
The book will of course be of interest to those interested in the language, style and syntax of ancient Greek texts. Given that all six contributions are primarily concerned with prose texts, and at least three with the interpretation of the classical historians, the collection also deserves attention from those concerned with the critical analysis of historical texts. The papers are on the whole short enough to keep attention from wandering, but all come with useful additional bibliography which reveals the wide range of disciplines (from linguistics to historiography) which underlie the present volume and are to a certain extent united in it.