Annamaria Bartolotta (ed.), The Greek Verb. Morphology, Syntax, and Semantics. Proceedings of the 8th International Meeting on Greek Linguistics. Agrigento, October 1–3, 2009. (Bibliothèque des Cahiers de l’Institut de Linguistique de Louvain—128). Louvain-La-Neuve: Peeters, 2014, Pp. i–ix + 342

The volume under review contains the reworked papers presented at the 8th International Meeting on Greek Linguistics held in Agrigento (Italy) on October 1–3, 2009 and hosted by the University of Palermo. This is a well-established and reputed biennial scholarly meeting which has produced a number of very interesting and useful collections of the studies presented in the preceding seven meetings on a variety of themes, such as dialectology and linguistic variation, phonology, syntax, semantics, etc. The theme of this volume (as of the conference itself) is the Greek Verb, and, as stated in the editor’s introductory note, the contributions deal with phonological, morphological, syntactic and semantic phenomena of the verbal system of Ancient Greek, whereas the topics covered present a wide variety, “[…] but the approach which unifies the volume is that of challenging traditional divisions and rigid boundaries between different levels of analysis […].” (pp. v–vi). In line with this general ‘philosophy’ of the studies in the volume, a wide variety of theoretical approaches are applied, “[…] both synchronic and diachronic, from cognitive (prototype theory) to formal (Distributed Morphology) to pragmatic-functional and historical-comparative approaches” (p. vi). The back matter of the book (pp. 329–342) is occupied by a word index, an index of subjects and an index locorum.

Thus, the nineteen papers cover a wide variety of topics, such as morphology, semantics, pragmatics, aspect, prepositional semantics, whereas in two studies a synthesis of semantic, pragmatic and syntactic factors is taken in order to tackle the phenomenon of tmesis; the theoretical approach of Distributed Morphology is applied by S. Reed in the analysis of the verbal paradigm of Classical Greek, one paper discusses the periphrastic perfect, whereas another paper deals with phonological problems, namely the one by A. De Angelis on the irregular use of the ending -σσω (instead of the expected -ζω) in the present of some verbs, in the framework of cognitive theory; the concept of transitivity is dealt with by J. Kavčić, using as case study the verb διαφθείρω in Classical Greek; finally, one paper is the sole study on etymology.

After summarizing the main points of each of the papers, in our brief presentation the focus will be on some studies which, to the reviewer’s mind, seem to be of special interest.

M. Benedetti discusses a special and rare use of the verb κλύω in a ‘copulative function’, as she says, a usage which is more common in tragedy, and which seems to reflect an Indo-European (henceforth IE) relic as far as the meaning of the root *ḱlew- ‘hear; be famed’ is concerned. This double meaning of the verb is a matter of diathesis, with the first meaning ‘hear’ used in non-middle structures and that of ‘be famed’ in syntactically middle structures. As pointed out by the author, this idea could lead to a more general study of other verbs not only from Greek but also from other Indo-European languages like, for instance, the root *weyd- ‘behold; perceive’, with interesting results that may affect the overall interpretation of ancient texts.

The phenomenon of tmesis seems to draw quite some attention in recent years (cf. for instance Duhoux, 1998, Giannakis 2001, Watkins 2002, Boley 2004, Hajnal 2004, Priestley 2009, Haug 2011 and 2014, to limit ourselves to the last 15 years or so). This is the topic of N. Bertrand’s paper, who discusses the issue of tmesis in relation to word order (with particular emphasis on the hyperbaton) and noun incorporation in the language of Homer. Two types of tmesis are distinguished, the ‘external’ type where preverb and verb are connected on semantic grounds, and the ‘internal’ tmesis whereby verb and preverb constitute a Verbal Complex. Some typological evidence is brought in to show that similar phenomena, especially with respect to noun incorporation, are also attested in different (not only IE) languages.

The second paper in this area is by A. Pompei, and it deals with the process of grammaticalization of the Verb-particle construction to the univerbation phase. This process tmesis-to-univerbation is completed in five successive steps almost all of which can be reconstructed for the language of Homer, with the exception of the fifth phase which “[…] is establishing a system of oppositions through different prefixes applied to the same verbal base” (p. 273). I think it is worth offering a little more attention to Pompei’s argument. I give here the five grammaticalization phases in the order they apply as given by her (p. 268):

View Table

The above schematic representation of the process of univerbation (= grammaticalization/lexicalization) of the P-V complex is understood as following (see also 272–273): In the first phase of co-occurrence, we have the simple syntactic arrangement of the two dislocated constituents, with P being simply an adjunct to the verb with locative meaning; P and V begin to form a complex in the second phase when P shifts from its originally locative function to an aspectual element, although it can still maintain its locative meaning; in the third phase P and V seem to be interpreted as forming a unified semantic complex, although the univerbation is not yet complete, something that is going to happen in the subsequent fourth phase of compounding. These four stages of the process are attested or can be documented by the Homeric data. The fifth phase is the derivation phase, or the lexicalization phase, whereby the univerbated whole is interpreted as a thematized stem upon which new Ps can be added as a sort of reinforcement of the locative sense of the first P, creating thus ‘overcompound’ structures of the type {P₂[P₁V]}, e.g. προκατάκειμαι ‘lie down before’, etc. In some cases, especially later on in the history of the language the PV complex is completely thematized so that we have imperfect ἐκαθήμην from κάθημαι (= κατά + ἧμαι), where the prefixed augment ἐ- precedes the preverb (against the general rule according to which the augment is infixed between the preverb and the verb).

The discussion of aspect of the Greek verb continues to be one of the most popular themes in linguistic studies. The stated aim of R. Blankenborg’s paper under the title “The grammarized suggestion of proximity or distance: aspect in Ancient Greek” is “[…] to suggest a clearer formulation of the difference between especially imperfective and aorist aspect with reference to pragmatics” (p. 31), something that is done in the paper focusing on the infinitive and the imperative of the Ancient Greek verb, although I admit I do not really understand the neologism ‘grammarized’ used in the title here. C. Bruno discusses the formation of the periphrastic perfect with ἔχω, which has eventually taken over the older synthetic form, a process that is common, found in many other IE languages. R. Faure argues that the future oblique optative has a narrative function in embedded sentences after past tense, a usage which is reminiscent of albeit quite different from the historical present.

Prepositional semantics has always been a good topic for analysis as afar as aspectual and other similar usages are concerned (see, among others, Purdie 1897, Meltzer 1901, Brunel 1939, to refer to older studies), and this collection contains two studies on the topic. The first is by C. Citrano who discusses the meaning of compound verbs mainly with the preverbs ἀνά, ἐπί and secondarily with ὑπέρ in Homeric Greek within a cognitive framework. The second paper is by A. Revuelta Puigdollers who in a richly documented and well-argued study discusses verbs compounded with the preverb περί.

The use of tenses in the Homeric text often presents a real crux, especially the poet’s choice between the aorist and the imperfect. Quite often the choice seems to have no motivation, some other times meter plays its role, whereas other cases may be motivated by pragmatic, psychological or other considerations. This is the topic of E. Crespo’s interesting study “A rule for the choice of aorist and imperfect”, where the author examines several Homeric passages from the Iliad in which the aorist and the imperfect of the same verb seem to co-occur with reference to a single telic event. The solution offered by Crespo in these cases is by means of a rule that seems to govern this tense shift in the Iliad and which is: “when in a narrative context referring to the past time the narrator repeats the same verbal lexeme in the singular to refer to an action that was carried out first by an agent and then by a different one, all the verbs appear in the aorist save the last one, which can be in the imperfect or in the aorist” (81). This is an important observation, since the tense alternation in such contexts would normally and traditionally be explained mainly by the ‘conjunction reduction’ rule. My own studies on similar topics show that there is also an interesting point to be observed here: in most cases, the verbs that alternate between aorist and imperfect are verbs that form the present stem with reduplication, e.g. (ἔ)τικτε(ν) : aor. ἔτεκε/τέκετο, ἐδίδου : aor. δῶκε, (ἐ)τίθει : aor. κατέθηκεν. There seems to exist some connection between present-stem reduplication and the aorist stem (in some other cases the perfect stem as well) in such ‘tense’ shifts where we have a neutralization of the morphological opposition and the creation of what I termed ‘homotropy’, i.e. similar aspectual usage (see Giannakis 1997: 51). The same holds true in other such oppositions between imperfects and aorists not necessarily of the same verb and/or of the reduplicative type. The solution in these instances relies on the close contextual reading of the text rather than on strictly morphological or grammatical grounds. As for the pair (ἔ)τικτε(ν) : aor. ἔτεκε/τέκετο, the solution may be pragmatic: in the majority of cases the aorist is used in order to render a purely historical (thus more distant) fact, whereas with the imperfect the birth of the last member in a genealogical line was rendered, perhaps expressing the sentimental attachment that exists between the child and his mother (see Giannakis 1997: 201–206; also 1993). In the rest of the cases with other verbs, there might be more than just one single general rule that is in operation, one of them being the ‘conjunction reduction’ rule. However, Crespo’s suggestion seems to hold as a more general principle, though not, I would say, as a rule in the sense understood in linguistic literature, like for instance Wackernagel’s Law for the second position placement of enclitics and the like. In this way, a variety of factors would be allowed to operate within this general principle (for a similar study with some statistics, see Crisafulli 1969).

In a detailed study with rich documentation, J.L. García Ramón tackles the issue of the morphosyntax of the Greek middle/passive aorists in -η- and -θη-, trying to detect the model according to which these formations took place in the language. His claim is that the passive aorist morph -η- goes back to the stative morpheme *-eh₁- of Indo-European, and it was integrated into the inflectional system of the Greek verb as the result of a grammaticalization process, with its beginnings in a few verbs and the subsequent spread to other verbs as well. One function of this morpheme was to create aorists for stative roots which originally lacked this tense but were confined to the present stem formations only. It is also argued that this suffix was added to two types of verbs, transitive or intransitive telic-transformative and stative, i.e. non-telic verbs, associating thus these formations with aspect and aspectual characteristics of the verbs in question.

J. Kavčić discusses the differences as far as meaning is concerned of the verb διαφθείρω when used with agent or patient in the aorist and the perfect, as seen in a number of Classical Greek authors. She locates differences between the two tenses, which she attributes to the gradient nature of transitivity, with the aorist being highly transitive, whereas with the perfect the degree of transitivity is found to be much lower, a fact that also shows that the theory of the merger of the aorist and the perfect in the history of the language cannot really rely on the syntactic similarities between the two tenses, at least as far as the data with the verb used here as case study is concerned. My feeling, however, is that conclusions based on the study of a single verb cannot really allow us to draw any reliable conclusion and any generalizations are premature, a fact also admitted by the author.

Using Vedic, Hittite and Greek data, R. Lazzeroni shows that the second singular imperative ending -i has nothing to do with the deictic particle -i that we find in the primary endings of the IE verb conjugation and which expresses the hic-et-nunc meaning of the verb. L. Melazzo discusses the κ-aorists and perfects, suggesting that the -κ- originated in laryngeal roots in *h₁ or *h₃ when coming in contact with the first singular ending *-h₂e of the perfect, therefore its origin is phonetic. Furthermore, he considers that originally the aorist form in -κ- of the verbs δίδωμι, ἵημι, τίθημι, i.e. ἔδωκα, ἧκα and ἔθηκα respectively, pertained to the perfect as far as their meaning is concerned. As stated by the author, P. Poccetti’s aim in his article entitled “Some thoughts about ‘delocutive’ verbs in Greek: Functions and semantics” is to apply Benveniste’s idea and definition of this very productive class of verbs and see how they behave in Greek. Furthermore, he establishes certain cross-linguistic criteria for their classification, saying that both colloquial and pragmatic concerns are involved in their handling and examination, a fact best exemplified with many hapax delocutives that occur in Aristophanes.

The cognitive linguistic framework is used in the explanation of the creation of a paradigmatic irregularity in a class of present forms in -σσω (/-ττω) in Greek as opposed to the general pattern in -ζω discussed by A. De Angelis. The explanation offered by the author is by means of a paradigmatic analogy with other verbs in -s-, for instance the prototype ἔφραξα : φράττω works as the basis for other formations like ἔσφαξα : σφάττω (next to the regular forms φράζω and σφάζω), already pointed out by Schwyzer. The basis of the analogical creation is triggered by two formal similarities of these verbs, namely (a) their disyllabic paroxytone form and (b) their radical vowel /a/ with a consonantal onset.

S. Reed applies the Distributed Morphology theoretical model in presenting the verbal system of Ancient Greek, showing the advantages of such an approach, among them its predictive power for the entire corpus of the data, its explanatory benefits for the analysis of the semantic and morphological components of the (tense-aspect) categories of the verb of the language, as well as to “[…] account for semantic isomorphism across different tenses, moods and aspects” (277).

Aspect has for a long time been a fruitful area of investigation. S.E. Conti explores the aspectual opposition of present and aorist imperative, with the final result confirming the common interpretation of imperfective vs. perfective aspectual contrast between the two tenses. Considerations of Aktionsart are also playing an important role in the final interpretation, for instance the categories of telicity and definiteness. The final paper of the volume is by M. Zinzi, which deals with the peculiar formation of the ‘Doric future’ as used in Attic poetic texts. According to the author the traditional explanation that these formations are motivated by metrical considerations in these texts does not stand; the meanings expressed by these futures are rather epistemic and behave rather like subjunctives and not like ‘true’ futures, therefore their formative basis seems to have a semantic motivation.

Etymology is the least represented in the collection, this time with the study by M. Vernet on the etymology of Gk. βάλλω ‘throw’, Old Irish a-t _˙ baill ‘die’, and Lat. ualleō ‘die’, concluding that all three items derive from the same etymon, namely the IE root *g^welh₁- ‘to throw’. As for the Greek and Old Irish items, it is known from etymological works in the past that they may be associated; the contribution of this study is that it adds to the same family of words the Latin word, a rare presence in Latin, although of uncertain reading (once in Accius and in a gloss of Paulus Diaconus). The short note is well argued and with a good line of thought, and the end result could be rewarding in that a (minor) etymological crux (mainly of Latin) is resolved, despite the uncertainties that may still remain.

The papers make a good and interesting collection of essays that cover many issues of the verbal system of Ancient Greek. Most of them are well organized, with good documentation and reliable results, some even present innovative interpretations and/or explanations. Aware of the risk of being characterized by subjectivism, among them I could distinguish a few papers for their scholarship, handling of data, analysis and innovative suggestion, such as Crespo’s original and interesting explanation of the distribution of the imperfect and the aorist tenses in Homer, García Ramón’s rich in all respects account of the morphosyntactic features of the -(θ)η- morpheme of Greek in the formation of the middle/passive aorist, and the two papers by Bertrand and Pompei who tackle, from a different angle, the phenomenon of tmesis in Ancient Greek epic poetry.

A final point is in order here: many papers needed an editorial attention, especially with respect to the English language. In many instances ‘italicisms’ have crept in several texts, whereas in other instances wrong syntax and/or usage of English words are apparent. This mishap notwithstanding, I think that as a whole the volume is a worthwhile effort and a good contribution to the study of the Greek verb. Once again it becomes clear that despite the long tradition in the study of the Greek verbal system, there still are issues to be readdressed and others to see for the first time; in general, the verb is a very fertile ground for still more productive and original work.