Friday, November 11, 2016


Lars Nordgren, Greek Interjections: Syntax, Semantics and Pragmatics. Trends in linguistics, 273. Berlin; Boston: De Gruyter Mouton, 2015. Pp. xii, 270. ISBN 9783110339369. $140.00.

Reviewed by Coulter H. George, University of Virginia (

Version at BMCR home site


In their authoritative reference grammar of Greek, Kühner–Blass devote only a little over a page to interjections, dismissing them as "blosse Empfindungslaute und … daher für die Grammatik bedeutungslos" ("simple sounds of emotion and … therefore of no importance for grammar") (§326, pp. 253–4). But anyone who has ever read Greek drama knows that they are ubiquitous, with nearly 1700 examples in the plays and fragments of the three tragedians and Aristophanes, which serve as Nordgren's corpus. Thus, especially in light of recent theoretical work (e.g. that of Felix Ameka), the time was ripe for a new investigation of the subtle differences, in semantic meaning and pragmatic use, between such exclamations as αἰαῖ, οἴμοι, and φεῦ. Michèle Biraud's Les interjections du théâtre grec antique (Peeters, 2010) was a significant step towards filling this lacuna, with important discussions of many interjections, but as it did not receive much attention outside the Continent (L'Année philologique records only one review in an English-language journal) and is not structured as a reference work, Nordgren's comprehensive study of these most slippery of words is timely and welcome. While not everyone will agree with all of his assessments—in the end, it is probably impossible to avoid a certain amount of subjectivity in deciding whether, say, pain or surprise is the key emotion expressed by a particular instance of an interjection—his book will nevertheless serve as the first port-of-call for anyone wishing to understand the most distinctive nuances of these words.

Nordgren begins by defining exactly what he means by interjections. As is often the case with linguistic categories, the boundaries are not watertight, but certain guidelines allow one to establish at least a prototypical view of an interjection: an indeclinable word that expresses a mental state, attitude, or reaction on the part of the speaker, and stands alone syntactically. They thus differ from onomatopoeic words, which generally describe physical phenomena, and from particles, which are integrated into their host utterance. Secondary interjections, like ἄγε, which are transparently derived from other parts of speech, are excluded from consideration here as well. Nordgren then sets out the three main classes of interjections, categories that are central to the structure of the rest of the book. First, there are the expressive interjections, exclamations like αἰαῖ, οἴμοι, and φεῦ that directly express the emotions of the speaker. Second, interjections like εἶα and ὠή—which Nordgren, following Ameka, labels conative, but which might more intuitively be classified as directive—express a desire on the part of the speaker that the addressee should do something; they often resemble imperatives. Third, phatic interjections, such as εἶἑν and ναί, express the mental state of the speaker towards the discourse itself.

In keeping with the subtitle of the book, Nordgren then divides the body of the study into three main chapters: syntax, semantics (by far the most substantial part of the book), and pragmatics. At first glance, given that interjections are defined, in part, through their failure to interact with a host clause, one may well wonder how syntax is relevant at all. The answer: just because interjections are not dependent on words higher up the syntactic chain of command doesn't mean that other words cannot in turn be analyzed as dependent on them. Most of the chapter, accordingly, discusses the sequence of words that is often found following expressive interjections of his first category, analyzing the interjection as the head of a phrase with up to five elements governed by it, but with the complete set found only outside his corpus, in Homer: ὤ μοι ἐγὼ σέο, τέκνον, ἀμήχανος (Od. 19.363): the first slot is occupied by the dative of the enclitic first- person pronoun, the second by a nominative (or vocative) nearly always referring to the speaker, the third by an ablatival genitive of exclamation, and, occasionally, a vocative in position four, with the speaker again referenced in the fifth slot. The schema may well be more rigid than is called for; for example, of the four examples of Position 5 (p. 66), one does not follow his pattern—in E. Ph. 373 (οἴμοι τῶν ἐμῶν ἐγὼ κακῶν), the nominative is actually nested within the genitive phrase that ostensibly occurs in the third position—and Diggle suggests emending away a second (Supp. 805) on the basis of parallels in Menander.1 One also wonders whether it is always right to consider the interjection as the syntactic head of the phrase, since many of the elements said to be dependent on it can occur independently of interjections as well. But in detailing the sheer number of interjection phrases that follow some variation of this pattern, Nordgren has usefully drawn attention to this particular area of Greek word order. In addition to brief discussion of the syntactic behavior of the other two categories of interjections, morphology is also covered in this chapter: while interjections do not inflect, they do undergo non-standard morphological variation, being extended through the use of repetition (παπαῖ can be extended to παπαπαῖ); and ἰ- may be prefixed, and -άξ suffixed, to certain interjections, although their exact force remains unclear.

Next comes the core of the book, on the semantics of interjections. In each case, Nordgren follows the principle of moderate minimalism; that is, he aims to describe a single basic meaning of each interjection, which may be abstract and wide-ranging, but also has the virtue of being unified. It is an ambitious goal, considering that we do not have access to intonational particulars that would otherwise help distinguish between various uses: think, for instance, of the difference in English between the Oh! of surprise, with its high, rising intonation, and the lower-key Oh. of resignation. That said, one could presumably extrapolate even from purely written data that Ouch! is, essentially, an expression of pain, while Oops! signals the recognition of a mistake.2 Thus, we may well expect a similar situation in Greek: some interjections of more circumscribed use will be amenable to definition with reasonable certainty, but others will remain semantic potpourri. To illustrate how this plays out, it will be helpful to turn to specific examples.

First, consider Nordgren's Category 1 interjections, those expressive of some sort of mental state, to whose semantics he devotes some seventy pages (pp. 93–164). He subdivides these according to the general emotion they signal, namely pain (645 examples), lamentation (551 exx.), surprise (231 exx.), and joy, with only eighteen examples, in a distant fourth place —unsurprisingly, given the corpus. For each of these categories, he goes through the individual interjections in turn, exhaustively considering how commentators have handled them, and arriving at a single core meaning; these are then summarized in a final section for each emotional category, such that, for expressions of surprise, we learn that, while ὤ is the unmarked item in this group, with the 'informational equivalent' "I am surprised", others are more specific, e.g. αἰβοῖ "I am surprised (because of the quality of this)", βαβαί "I am surprised (because of the quantity of this)". Sometimes these assessments are grounded in details of the specific passages Nordgren cites. His examples of αἰβοῖ, for instance, by and large suggest that it is used when the speaker is confronted by something completely different from what he expected (Ar. Ach. 189, Nu. 829, Av. 610, 1342), whereas βαβαί (E. Cyc. 156, Ar. Ach. 806, Pax248, Lys. 1076) is uttered not so much to signal a complete incongruity as the realization that something is better or worse than expected (see the first and last examples respectively).

In other cases, however, one would have liked more argumentation in favor of one reading of the interjection over another. Most fundamentally, it is not always clear why one emotion is viewed as primary, another as secondary to the sense of an interjection. Consider the very first interjection covered, ἆ, whose informational equivalent, Nordgren concludes, is "I am surprised (and I dislike what I notice)" (p. 100). The wide emotional range attributed to it may be seen from even just a handful of the commentators Nordgren quotes: Barrett calls solitary ἆ "a sharp cry of protest", for Kannicht it expresses either "körperliche oder seelische Schmerzempfindung" on the one hand, or "Erstaunen … oder Protest" on the other; Dodds is similarly vague: "It can be the gasp of astonishment … or a groan of pain … but often it expresses urgent, protest, 'Stop!'." Introduce doubled ἆ ἆ, and the picture becomes even more complicated: Parker calls it "a tragic cry of surprise, often accompanied by pain, or … anger," and Mastronarde "an exclamation often expressive of surprise or distress"—and Sandin distinguishes between single ἆ, which largely expresses pity, and the doubled interjection, which expresses alarm, pain, or protest. Given these descriptions, where should one position ἆ in Nordgren's schema? Perhaps it really is a cry of surprise, but couldn't it just as easily be pigeonholed as an exclamation of pain—or, even, with its frequent use in protests, in the second category, the conative interjections, that, like directives, attempt to influence the behavior of the addressee? Nordgren is of course well aware that multiple different uses of ἆ are in play; but when he himself writes, "The emotive core meaning of ἆ is dislike" (p. 100), and never explicitly takes a stand on what difference, if any, there is between ἆ and ἆ ἆ, one is left uncertain where the boundary lies between an interjection of surprise felt at something painful and an interjection of pain caused by something surprising. It is also regrettable that Nordgren does so little to engage with Biraud's monograph in this section: she is not cited at all in the discussion of ἆ, even though she devotes considerable space (pp. 80–8) to analyzing the various uses of this interjection. Given that Biraud, for her part, pays little attention to the rich Anglophone scholarship on drama, it is a pity that the opportunity to unify the two strands of scholarship was missed.

When he moves to the second and third categories of interjections, Nordgren generally returns to firmer ground: ὠή, for instance, can convincingly be summed up as meaning "I want you to pay attention to me"; and the most charming interjection in the book, the ψύττ' of E. Cyc. 49, is indeed equivalent to "I want yousheep to move", as can be ascertained from the words that follow: οὐ τᾷδ', οὔ; "Won't you go this way?" Similarly, in the phatic domain, the position that ναί indicates agreement—and εἶἑν compliance—with the preceding utterance will meet with little opposition.

In the final body chapter, Nordgren turns to pragmatics: the consideration of how the context of the utterance can inflect the core semantics of an interjection. In particular, he looks at five ways in which secondary uses can become conventionalized as regular functions of an interjection, with the first perhaps the most important: what begins life as an expressive interjection can develop a secondary use as a directive (e.g. ἰώ, originally an expression of lamentation, comes to be used as a cry for help). While there is a short conclusion, the book's real closural device is an alphabetical lexicon of interjections, citing every example of each word's use in drama, along with relevant collocations, metrical data, and textual issues. It is a useful list, although it is a shame that Nordgren did not sort the citations by usage: thus, one cannot tell from the lexicon e.g. which examples of ἰώ he regards as expressions of lamentation and grief, as invocations, or as appeals for protection. This should not, however, detract from the chief accomplishment of Nordgren's book: anyone who wants more information on the difference between interjections like οἴμοι, ὤμοι, and φεῦ now has an obvious first place to start.


1.   Since all these phrases involve the pseudo-enclitic placement of ἐγώ after the possessive pronoun ἐμός, it would be worth comparing the examples in Helma Dik's 2003 article "On unemphatic 'emphatic' pronouns in Greek" (Mnemosyne 56(5): 535–50).
2.   The sense of "Oops" would, one imagines, be as retrievable from a transcript of this debate as from the video.

No comments:

Post a Comment

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