Klaas Bentein, Verbal Periphrasis in Ancient Greek. Have- and Be- Constructions. Oxford: Oxford University Press. 2016.
The development of periphrases constitutes a well-examined central phenomenon for the diachrony of the grammar of Greek as well as crosslinguistically. Bentein (hereafter B) adds new perspectives to the discussion of change in periphrases: transitivity is considered another parameter that can reveal aspects of periphrases. B also analyzes synchronic characteristics of periphrastic constructions in various registers, demonstrating a variation with regard to these constructions. This makes B’s book a valuable addition to other works on the diachrony of periphrases.
The study provides an in-depth examination of the development of the Ancient Greek verbal system by investigating the diachrony of constructions with εἰμί ‘be’ and ἔχω ‘have’ and a participle (see 1). The constructions under investigation belong to an area of some debate, mainly with regard to two questions: whether both εἰμί ‘be’ and ἔχω ‘have’ form periphrases in Ancient Greek, and whether the analysis of adjectival periphrases, such as πρέπον ἐστί ‘it is fitting’, is similar to that of the (typical) verbal periphrases. B shows continuities as well as discontinuities between Archaic/ Classical Greek and Post-Classical/ Byzantine Greek. He organizes the rich data that he offers into relatively small subperiods: Archaic Greek (8th–6th cent. BC), Classical Greek (5th–4th cent. BC), Early Post-Classical Greek (3rd–1st cent. BC), Middle Post-Classical Greek (1st–3rd cent. AD), and Late Post-Classical and Early Byzantine Greek (4th–8th cent. AD).
Chapters 1 and 2 of the book offer a discussion of the main concepts: periphrasis, aspect, grammaticalization, and verbal–adjectival periphrasis. Chapters 3–5 present the diachrony of periphrastic constructions with εἰμί ‘be’ and ἔχω ‘have’ from the Archaic to Early Byzantine period. Their development is analyzed based on three basic functional domains: perfect aspect, imperfective aspect, and perfective aspect. The main motive in all sections (offering coherence to the analysis in a systematic way) is a focus on grammaticalization paths and on semantic transitivity (following Hopper & Thompson 1980). This decision leads to a creative way of attempting a synthesis of grammaticalization and transitivization. The book successfully raises strong criticism against many previous proposals. For instance, B deconstructs Dietrich’s (1973ab) idea of a relationship between Erzählungsmanier (‘narrative pace’), periphrasis, and Christian texts; he shows that it is not the case that the authors of a genre follow the same narrative pace (p. 272).
The whole study is heavily integrated into a theoretical model, as one can realize with the sections on the noun-to-verb continuum, for instance. According to this approach, adjectival participles are located on a continuum ranging from most noun-like to most verb-like. The perfect participle is least verb-like, and the present participle is more verb-like than the perfect participle but less than the aorist participle, which is the most verb-like. In this way, B follows the prototype model of linguistic categorization, according to which there are no sharp boundaries between categories and some category members are more prototypical than others; periphrastic constructions can be located on a “periphrasticity” continuum in which the most prototypical periphrastic constructions are syntactically contiguous, paradigmatically integrated, and semantically idiomatic.
Moreover, the book contains an exhaustive description of all subcases/interpretations of each of the constructions, as attested in the texts. Consider, for example, εἰμί ‘be’ with the present participle (pp. 242–243): B discusses (a) the case where the particular periphrastic construction can be related to a foregrounded event; (b) examples where a foregrounded event serves as the “focalization point” for the periphrastic construction under examination; (c) other instances where the periphrastic construction is coordinated with another clause, one of them serving as the focalization point and the other as the backgrounded event; and (d) examples where the focalization point for the periphrastic construction is pragmatically provided.
B draws clear conclusions, even though his starting point is a huge amount of data and the close examination of a large number of passages and quantitative results:
(a) εἰμί ‘be’ with the perfect participle in the domain of perfect aspect and ἔχω ‘have’ with the present participle in the domain of imperfective aspect underwent a process of transitivization, meaning that they were used in increasingly transitive contexts. For instance, εἰμί ‘be’ with the perfect participle was used increasingly more often in the active voice, with a volitional first participant and an affected second participant.
(b) Periphrasis in general became more transitive over time: it appeared in the domain of perfect aspect first (εἰμί ‘be’ with the perfect participle and ἔχω ‘have’ with the aorist participle); in a later stage, it appeared in imperfective aspect (εἰμί ‘be’ and present participle), which is more transitive than the perfect aspect because it can express an action rather than the state which may be the result of an action (as a perfect aspect does) (cf. the kinesis parameter in the transitivity scale of Hopper & Thompson); and it appeared in perfective aspect in a final stage: unbounded imperfective aspect is less transitive than bounded perfective aspect (cf. the aspect parameter in the transitivity scale of Hopper & Thompson).
B’s attempt to use transitivization as a criterion of how grammaticalized a construction with εἰμί ‘be’ and ἔχω ‘have’ and a participle is makes this part of his analysis innovative. The basic argument according to this perspective is that the periphrastic use of a construction is much higher in transitivity. Bentein uses a concept such as transitivization to replace the contrast between verbal and adjectival periphrasis in a way that includes the risk that the described increase in transitivity is actually identified with an increase in the verbal characteristics of a construction (that is, the periphrasis becomes more verbal than adjectival, in terms of B’s scale). B states “sentences containing ‘adjectival’ periphrases are typically low in transitivity” (p. 306). Transitivization is actually seen as one additional result of a change in the manner (through periphrases) in which tense and aspect are grammaticalized; but it can indeed mean a more verbal than adjectival periphrasis in terms of B’s scale. Therefore, the conclusion on transitivization can be re-interpreted as integration of periphrasis into the verbal system. In this respect, the perspective followed does not allow B to exclude some examples as cases of clear adjectival usage of the participle (for instance, if the participle is coordinated with a typical adjective; see p. 150 or p. 185). B considers these examples important for the origin of the construction, but these examples may also add some cues on the synchronic nature of the particular participles and (nonperiphrastic) constructions in the particular stage.
The idea to include transitivity as a basic criterion of the examination of periphrases probably has its starting point in an observation about a similarity between Pustet’s (2003) parameters of prototypical verbhood and Hopper & Thompson’s (1980) parameters of prototypical transitivity (p. 100). However, transitivity is one of the three parameters (together with “dynamicity” and “transience”) that Pustet uses. Moreover, transitivity for Pustet is identified with valency and does not take the form of transitivity of Hopper & Thompson, which includes ten component parameters: participants, kinesis, aspect, punctuality, volitionality, affirmation, mode, agency, affectedness of the object, and individuation of the object.
We should notice that B’s study does not focus on an obvious link between his claim that εἰμί ‘be’ with the perfect participle developed from a resultative perfect to an anterior perfect and his observation that this particular construction became used in more transitive contexts. As transitivity for B is semantic transitivity, it is evident that the new meaning of anterior perfect is absolutely more semantically transitive than the meaning of the resultative perfect.
The possible connection between transitivization and grammaticalization paths is not supported in the study with an examination of the diachrony of transitivity. For this reason, it seems that transitivity is considered in several sections as a rather stable parameter. However, there is a common consensus in relevant crosslinguistic studies that transitivity undergoes several important changes. This holds true especially if transitivity is seen as consisting of several parameters, as in B’s analysis, which follows Hopper & Thompson’s perspective. For instance, a change in cases can reflect a new direction for transitivity, or the development of intransitive verbs into transitive verbs can be related to a transitivization process, or changes in voice, which marks the transitive and intransitive constructions, can also show an innovated transitivity system.
Accordingly, the study does not emphasize and does not include as a main part of its analysis a change that has affected both the system of aspect and transitivity: the clear transition in Greek (and a common development in many Indo-European languages; see, for instance, van Gelderen 2004) from lexical aspect (Aktionsart/ inner aspect) to grammatical (outer) aspect; the Greek verbal system developed from the expression of oppositions of lexical aspect/ Aktionsart (states—telic activities—atelic activities) to the expression of oppositions of grammatical aspect (perfective—imperfective) and tense (Moser 2008). Two significant changes mark the transition from the stage of lexical aspect to the stage of grammatical aspect: (a) a tendency toward transitivization, with an increase in the number of transitive constructions for verbs (not only periphrases) that were used only in intransitive constructions (Lavidas 2009) and (b) the grammaticalization of the contrast between the active and passive voice morphology (for instance, more verbs can be passivized in later stages of Greek; Luraghi 2010).
Furthermore, one could re-interpret the findings on the transitivity parameter as a possible broad interrelation of finiteness and transitivity and their development. This, of course, would demand the existence of parallel characteristics of development of participles with other nonfinite types. This option, however, cannot be excluded because the differences between participles and the relevant periphrases and other nonfinite types were not aims of the book.
We should note that the book contains a rich number of figures, tables, an appendix (about the texts included in the corpus), indices, and a glossary. On the other hand, the structure of the book can be observed as quite unbalanced, in that one-third of the book focuses on the background concepts to periphrasis and aspect (chapters 0–2). The other part of the book consists of three main sections based on the functional domains of perfect aspect, imperfective aspect, and perfective aspect. English translations of the Early Greek words and passages are given systematically in the whole book (with one possible exception: Table 0.1). Information about the source of the translation of the passage is always provided. After each example, B discusses, in detail, the context of the example, with regard to the meaning and the function of the construction under examination. In many cases, a significant use of remarks of text commentators assists the advancement of the linguistic analysis, which shows a path of cooperation between philological commentators and the linguistic description (see, for instance, p. 238). In many cases, B successfully compares the Greek data with data and relevant developments in other languages (for instance, p. 221), offering a very significant cross-linguistic perspective in the analysis. One should also note the detailed but easy-to-read presentation of Tense/Aspect and the essential presentation of subfunctions, according to the context. For instance, with regard to perfective aspect, B describes the main features of a number of subfunctions: “constative [perfective] aspect,” “ingressive [perfective] aspect,” “generic [perfective] aspect,” and “habitual [perfective] aspect.” Very recent theoretical analyses are also taken into consideration for the description of the examples from Early Greek. B’s criticism is presented not only against early analyses but also explicitly against many recent analyses of the verbal system of Classical and Post-Classical Greek.
On the other hand, glosses are not given in any example. This, in combination with the examples consisting of long passages, the relevant constructions not being underlined in all the examples (especially in the first part of the book), and there being no constructions underlined in English translation, can make some sections of the book less easy to read to a part of the general public. The English translation of the passages is not always close enough to a literal translation, causing difficulties in the presentation of the usage of the construction in some cases.
The book includes considerable quantitative data. The quantitative results are presented in the form of a normed rate of occurrence (with the number of instances per 10,000 words) and of a type-token ratio (the ratio of the number of different nonfinite verbs to the total number of nonfinite verbs appearing in each construction). However, there is no discussion of the statistical significance of the results. The corpus of the study is based on two online lemmatized databases: the Thesaurus Linguae Graecae (TLG) and the Duke Databank of Documentary Papyri (DDBDP). Information would be necessary on the corpus—whether parts of the databases or the whole databases are used and why—and the corpus study methodology—which also is related to the theoretical background of the study on whether “stereotype expressions,” such as ὁμολογούμενόν ἐστι ‘it is acknowledged’ or ἔστω περιεχόμενον ‘let it be encompassed’, are included in the quantitative data. We can further observe that no quantitative or in-depth analysis of the parallel absence/presence of synthetic forms (e.g., of the perfect) is offered, even though B considers them an important factor for the development of periphrases (for instance, he refers to the hypothesis that the developments in the synthetic paradigm can be closely related to the changes in periphrases (p. 113)).
Particular emphasis is put on register and genre (see, for instance, section 1.3), but it would have been useful to have a sociolinguistic classification of the texts in the corpus—as well as their regional dialect. Even though the author mentions that the “tenor vector” (concerning the social relation between the interactants) is an important factor since the mode of discourse (written communication) is stable and the field of discourse influences the lexical preferences, and not the grammatical choice, the “field,” instead (in the form of macro-genres), is discussed in detail: epic poetry, drama, scientific prose, etc. In parts where the book refers to the “tenor vector,” in the form of lower versus higher register, the sections appear to be inconclusive or mainly indicative of how difficult a task it is to have quantitative results in this respect. For instance, B comments (a) on scholars who consider Plato’s dialogues—especially the early ones—as a source for everyday language, whereas Plato also adopts higher-level language in other passages of his works, and (b) on the documentary papyri, which do not constitute a homogeneous corpus with regard to the high-versus-low register question (see p. 23).
A quantitative analysis of the parameter of register by using all available texts can probably be misleading, whereas a close analysis of one text or a single author (distinguishing between his more and less high-level passages, for instance, narrative versus nonnarrative passages) can provide essential insights. Furthermore, a contrast between higher-register and lower-register language (instead of the contrast between the registers) could have been more easily supported by a larger list of texts.
B also examines significant questions on the role of language contact in the development of periphrases. He addresses these issues from a critical point of view and presents strong arguments against previous analyses. Moreover, the author proposes modifications for the progressive-imperfective drift schema, which was proposed by Bertinetto et al. (2000): the diachronic source for the imperfective cannot be restricted to the locative construction; a stative source construction can be evidenced, for instance, in the case of the development of εἰμί ‘be’ with the present participle. B observes that the habitual function is semantically closer to the (iterative) durative progressive function than the focalized progressive function. This fact makes him propose that the grammaticalization path for the progressive-imperfective drift should allow for multiple branching (for instance, in early stages, toward either stative or locative/durative or, in the last stage, toward either focalized progressive or habitual).
The book constitutes a perfect combination of qualitative and quantitative description and analysis. A clear example is the detailed presentation of the context of the passage from The Iliad on the story of Pandarus and Diomedes and what the passage can show about the use of εἰμί, “be,” with the aorist participle (pp. 294–295); the author also presents a clear description of the quantitative distribution of εἰμί ‘be’ with the aorist participle in several registers and texts of Classical Greek (p. 295). On the other hand, the investigation of the diachronic development allows B to argue in favor of some further significant remarks, which do not refer only to the particular constructions: (a) with regard to Greek, there is no “split auxiliarization,” comparable to what is found in many European languages; and (b) with regard to language change, the notion of the “grammaticalization path,” as a universal semantic path of the development of forms, is instrumental for periphrases, as proved with the examination of the functional domains of perfect and imperfective aspect.
All in all, the book serves its major aims successfully: it addresses a more general public, which can be helped with the long discussion of the basic concepts, the glossary, and the detailed comments on all passages; it contains an exhaustive analysis of a large corpus of text; and it offers an alternative analysis of the semantic paths of constructions (mainly) with εἰμί ‘be’ and ἔχω ‘have’ and a participle, from the 8th cent. BC to the 8th cent. AD.
Bertinetto, P.M., K. Ebert & C. de Groot. 2000. The progressive in Europe. In Ö. Dahl (ed.), Tense and Aspect in the Languages of Europe, 517–558. Berlin, New York: Mouton de Gruyter.
Dietrich, W. 1973a. Der periphrastische Verbalaspekt im Griechischen und Lateinischen. Glotta 51, 188–228.
Dietrich, W. 1973b. Der periphrastische Verbalaspekt in den romanischen Sprachen. Untersuchungen zum heutigen romanischen Verbalsystem und zum Problem der Herkunft des periphrastischen Verbalaspekts. Tübingen: Niemeyer.
Gelderen, E. van. 2004. Grammaticalization as Economy. Amsterdam and Philadelphia: John Benjamins.
Hopper, P.J. & S.A. Thompson. 1980. Transitivity in grammar and discourse. Language 56, 251–299.
Lavidas, N. 2009. Transitivity Alternations in Diachrony: Changes in Argument Structure and Voice Morphology. Newcastle: Cambridge Scholars Publishing.
Luraghi. S. 2010. The extension of the transitive construction in Ancient Greek. Acta Linguistica Hafniensia 42.1, 60–74.
Moser, A. 2008. The changing relationship of tense and aspect in the history of Greek. STUF-Sprachtypologie und Universalienforschung 61.1, 5–18.
Pustet, R. 2003. Copulas: Universals in the Categorization of the Lexicon. Oxford: Oxford University Press.