Monday, September 2, 2019


Albert Rijksbaron, Form and Function in Greek Grammar: Linguistic Contributions to the Study of Greek Literature. Edited by R. J. Allan, E. van Emde Boas, and L. Huitink. Amsterdam studies in classical philology, 30. Leiden: Brill, 2018. Pp. xviii, 428. ISBN 9789004385771. €127,00.

Reviewed by Coulter H. George, University of Virginia (

Version at BMCR home site

Publisher's Preview
[Chapters are listed below.]

Where better to begin than the book's epigraph? "[T]ake languages seriously. Whenever there is some overt difference between two constructions X and Y, start out on the assumption that this difference has some kind of functionality in the linguistic system." This principle, taken from Simon Dik's Theory of Functional Grammar, underlies the scholarship of Albert Rijksbaron, many of whose most important contributions to Greek linguistics have been conveniently gathered together in this volume (with Rutger Allan, Evert van Emde Boas, and Luuk Huitink serving as capable editors). Readers will probably be most familiar with Rijksbaron's work from his Syntax and Semantics of the Verb in Classical Greek and his linguistic commentary on Plato's Ion, but there is also much to be gained from bundling together this greatest hits collection, as it brings out especially clearly the fruitfulness of the Functional Grammar approach. On a wide range of topics—particularly the tense and aspect system of the Greek verb and the discourse particles—time after time, Rijksbaron's careful attention to the contextual circumstances that differ when two competing forms are used reveals previously unnoticed patterns that can help all classicists better understand the nuances conveyed by an imperfect in Hesiod, or an aorist imperative in Plato. To illustrate how this plays out in greater detail, what follows will first focus on three representative chapters from the collection, then skim more briefly over the rest.

Consider first Chapter 2, which discusses the Greek perfect. First published in 1984, it does an especially good job of questioning the existence of the so-called resultative perfect, but because it has only now been translated from the original Dutch, it should finally receive the attention it deserves. About a century ago, both Wackernagel and Chantraine had noted a shift in the usage of the perfect between Homer and Classical Attic: transitive perfect actives, often marked by a kappa, start to gain ground relative to earlier, mostly intransitive perfects, with e.g. πέπεικα "I have persuaded" coming into existence alongside older πέποιθα "I trust". According to these heavyweights of Greek linguistics, all these perfects indicate a state consequent upon some previous action, but, whereas the older forms highlight the state of the subject, the newer ones are said to emphasize the resulting state of the object, and were thus dubbed resultative perfects. But while a shift in the center of gravity of the perfect is undeniable, Rijksbaron rightly has reservations about the Wackernagel–Chantraine model and attacks it in two main ways. First, he shows that some of the perfects that Chantraine had adduced as object-oriented are in fact just as subject-oriented as ever. For instance, when Herodotus writes Ἀχελῴου, ὃς ... τῶν Ἐχινάδων νήσων τὰς ἡμισέας ἤδη ἤπειρον πεποίηκε "... Achelous, which ... has already made half of the Echinades islands to be mainland" (2.10.3), one might, in isolation, take this as justification of the resultative reading, as the state of the islands would seem to be more important than that of the Achelous. But, as Rijksbaron points out (p. 47), the words that precede—εἰσὶ δὲ καὶ ἄλλοι ποταμοί, οὐ κατὰ τὸν Νεῖλον ἐόντες μεγάθεα, οἵτινες ἔργα ἀποδεξάμενοι μεγάλα εἰσί· τῶν ἐγὼ φράσαι ἔχω οὐνόματα καὶ ἄλλων καὶ οὐκ ἥκιστα Ἀχελῴου... "There are also other rivers, not so great as the Nile, that have wrought great effects; I could declare their names, but chief among them is Achelous"—show that this is a passage about the standing of rivers, not that of the Echinades, and so it is the subject, not the object, of πεποίηκε whose state is more relevant here. Second, whereas the traditional account sees the perfect as falling together with the aorist remarkably early, Rijksbaron shows that the difference between the two tenses remains intact well into the Hellenistic period. In Sophocles' Philoctetes 923–4 and 940, for instance, Chantraine had regarded the verb tenses in Philoctetes' τί μ', ὦ ξένε, | δέδρακας; ("Stranger, what have you done to me?") and οἷ' ἔργ' ὁ παῖς μ' ἔδρασεν οὑξ Ἀχιλλέως ("the wrongs Achilles' son has done to me") as functionally equivalent. But, as Rijksbaron points out (p. 49), the contexts differ: the perfect is used in the former, when the agent, Neoptolemus, is still around to be blamed in the present, whereas, once Philoctetes is alone again, the aorist simply registers the occurrence of the event in the past. As a coda, Rijksbaron goes on to show that the distinction is still active as late as the Rosetta Stone (196 BC), where the aorist remains the main narrative tense, with the perfect reserved for events considered in light of their present significance.

In all of this, the presentation is a model of clarity, and my only complaint is the unfair one that he did not occasionally go even further in pressing his advantage. For instance, in preparing the translation from the original Dutch article (which only cited examples in Greek), the editors have used Godley's translation of Herodotus' οἵτινες ἔργα ἀποδεξάμενοι μεγάλα εἰσί ("that have wrought great effects"), which obscures the fact that the finite verb in the relative clause is in fact the stative εἰσί, thereby providing additional evidence for taking πεποίηκε as similarly stative in the parallel ὃς ... πεποίηκε clause. One might also have noted that the marvelous personification inherent in making rivers the agent of ἔργα ἀποδεξάμενοι—a phrase so thematically important to Herodotus—further increases the extent to which this passage is organized from their perspective. Likewise, in discussing the Sophoclean near-minimal pair, one wants to know more: the perfect is second-person and in a direct question, the aorist third-person and in a subordinate clause; do either of these features correlate more broadly with the distribution of the two tenses, as one might expect given Rijksbaron's argumentation? Furthermore, with the hindsight of an additional thirty-five years of research, it is also tempting to suggest that the change in the perfect can be connected to the past indicative augment's having become obligatory. If the perfect is preferred in direct speech and with first- and second-person subjects, this would match the conditions under which the augment is favored with the aorist in Homer,1 and the growth of the transitive perfect could be seen as a response to the loss of a distinction formerly marked by the presence or absence of the augment in the aorist indicative.

While the other chapters in the volume mostly present material that is comparatively accessible already, unexpected insights arise from assembling it in one place. Scholars of Hesiod's Theogony will probably have come across Rijksbaron's 2009 chapter on discourse cohesion in the proem (Chapter 9 in the current collection), with its discussion, in particular, of the notorious imperfect στεῖχον in line 10, which M. L. West had argued was a timeless imperfect (for which there are no good parallels) before labeling it an injunctive (a form that is only functionally distinct in Sanskrit). Rijksbaron instead opts for the view that this is a focalizing imperfect—that is, it presents the Muses' approach as a habitual occurrence experienced by Hesiod the character, rather than as an omnitemporal activity recorded by Hesiod the omniscient narrator. Now, in the original chapter, which was presenting a linguistic commentary on the whole of the first 115 lines, Rijksbaron did not have enough space to lay out fully all the parallels that are available as supporting evidence. So it is especially welcome that readers can now also find, in the same volume, a 2012 chapter (here Chapter 7) that puts the Hesiodic passage in the context of a more fully developed account of other such imperfects (here dubbed imperfects of "substitutionary perception", a term that is more targeted, but unfortunately also more opaque to the narratologically uninitiated). The class of imperfects to which στεῖχον belongs thereby becomes clearer, and the argument as a whole more persuasive.

Since space does not permit a detailed examination of the remaining chapters, it will be best now to step back and outline some major themes that emerge from the rest of the collection. First, tense and aspect loom large elsewhere as well. Chapter 3, working with evidence from Herodotus, deals with imperfects that, at the start of a passage, create the expectation of a narrative to follow; Chapter 5 also focuses on the imperfect, arguing that many supposed historical presents in Sophocles and Euripides (like καλεῖ at S. OT 1245) should in fact be printed as unaugmented imperfects (κάλει), since the events they describe are not decisive enough for the historical present to have been the appropriate tense for them. Chapters 4 and 6 both look at the fundamental division of aspect encoded by the present and aorist stems: the former contrasts Plato's use of λέγε and εἰπέ (the present invites interlocutors to keep speaking; the aorist asks them to establish a particular point), while the latter advocates for the present infinitive ἀκροᾶσθαι (rather than aorist ἀκροάσασθαι) at Ion 530d9, since σχολή, upon which the infinitive is dependent, typically calls for the more open-ended present infinitive. The embeddedness of language in dialogue also emerges as important: Chapter 8 investigates how much information Euripidean messengers are allowed to take for granted at the start of their speeches; Chapter 10 distinguishes between ἔφη and ἦ δ' ὅς as inquit formulae in Plato's dialogues, arguing that the former continues a line of discussion, whereas the latter closes it off (note the obligatory δέ); Chapter 17 considers whether Ancient Greek has a word for "no" (spoiler alert: no, it does not); and Chapter 19 not only posits that the difference between a πεῦσις and an ἐρώτησις is that between a wh-question and a yes-no question, but also observes that Pseudo-Longinus' On the Sublime mimetically illustrates both types of question in its discussion of them (18.1).

Finally, the linguistic means of establishing discourse cohesion form a strand common to several of the pieces. In addition to Chapter 9, covering the proem of Hesiod's Theogony, Chapter 11 teases apart the differences between Herodotus' use of preposed and postposed οὗτος with proper names (if δέ is present, then the former marks the topic as salient, the latter does not; if δέ is absent, preposed οὗτος marks both the topic and the proposition as salient); Chapter 15 shows that, in the combination καὶ ... δέ, it is καί that is the connective particle, with δέ indicating that the second element should be taken as a new entity in its own right, most often yielding the meaning "and for that matter"; and Chapter 16 shows the extent to which Xenophon only rarely gives Κῦρος an article in the Anabasis (since it is primarily a continuous narrative, there's less need to deploy the article in its strong anaphoric function to highlight the referent), but uses it considerably more in the Cyropaedia (where the anaphoric article is more at home, since the work is structured as a sequence of detached scenes).

The final chapter in the volume at first glance looks like the odd one out, insofar as it is an examination of the dozens of commentaries on Xenophon's Anabasis published over the last two hundred years. While its first half is indeed more or less what one would expect of such a piece—a look at broader differences in commentary publication in the US, UK, and Germany in (especially) the nineteenth century—the second half serves as a salutary cautionary tale that even so often-read a sentence as the very opening of the Anabasis was misunderstood by the vast majority of commentators, who fall demonstrably short in explaining both the genitive Δαρείου καὶ Παρυσάτιδος and the present tense γίγνονται. Thanks to Rijksbaron's efforts, however, classicists now have a much better understanding of such syntactic matters, and his elucidation of these particular points provides a fitting close to the volume.

Table of Contents

1. A Review of: H. Hettrich, Kontext und Aspekt in der altgriechischen Prosa Herodots
2. The Greek Perfect: Subject versus Object
3. The Discourse Function of the Imperfect
4. Sur les emplois de λέγε et εἰπέ chez Platon
5. On False Historic Presents in Sophocles (and Euripides)
6. Ἀκροᾶσθαι or ἀκροάσασθαι (Plato, Ion 530d9)?
7. The Imperfect as the Tense of Substitutionary Perception
8. How Does a Messenger Begin His Speech? Some Observations on the Opening Lines of Euripidean Messenger Speeches
9. Discourse Cohesion in the Proem of Hesiod's Theogony
10. On the Syntax and Pragmatics of inquit Formulae in Plato's Narrated Dialogues
11. Sur quelques différences entre οὗτος ὁ + substantif, οὗτος δὲ ὁ + substantif, ὁ δὲ + substantif + οὗτος chez Hérodote
12. Sur les emplois de ἐάν et ἐπεάν (à propos d'Euripide, Bacchae 50–51)
13. The Syntax and Semantics of Expressions of Sorrow and Related Concepts in Homer
14. The Meaning and Word Class of πρότερον and τὸ πρότερον
15. Adverb or Connector? The Case of καὶ … δέ
16. Sur l'article avec nom propre
17. Does Ancient Greek Have a Word for 'No'? The Evidence from οὐκοῦν (…) οὐ Questions
18. The Treatment of the Greek Middle Voice by the Ancient Grammarians
19. A Question of Questions: peusis, erôtêsis and [Longinus] Περὶ ὕψους 18.1
20. The Xenophon Factory: One Hundred and Fifty Years of School Editions of Xenophon's Anabasis


1.   For the tendency of the Homeric augment to occur more often in character speech than in narrative, especially with perfect-like aorists, see Andreas Willi, Origins of the Greek Verb (Cambridge, 2018), pp. 369–72.

No comments:

Post a Comment

Note: Only a member of this blog may post a comment.