Reviewed by James Clackson, Jesus College, Cambridge (jptc1@cam.ac.uk)
Pinkster reveals in his preface that the publication of his new Syntax of Latin (hereafter OLS) was originally intended to be published in 2014, a century after the publication of the second volume of Kühner-Stegmann's Satzlehre.1 Volume 1 of OLS was published in 2015, and we await news of Volume 2. The comparison with Kühner-Stegmann reveals the ambition of the project, and it is undoubtedly true that this book will become the point of reference for the next century of Latin scholarship as Kühner-Stegmann has been for the last. For those who think that we already knew enough about Latin grammar, think again. This work is the culmination of decades of research on Latin, enriched by the discovery and publication of new texts and a century's advances in linguistics. Furthermore, the Latin grammarian of the twenty-first century has resources that outstrip anything available to Kühner and Stegmann: not only digitised texts, but also the Thesaurus Linguae Latinae (and the Oxford Latin Dictionary), and Pinkster has made excellent use of all of these (see, for example, the discussion of medius, pp. 1048-1050, which relies heavily on the Thesaurus article). Some parts of OLS parade the grammarian's fancy new clothes and the benefits of research on digitised corpora. For example, one pie-chart shows the frequencies of Latin cases and prepositional phrases based on a survey of over six thousand nouns and pronouns occurring in a selection of five prose and five poetic texts (p. 1180); another on the next page gives the figures for the occurrences of different types of syntactic constituent in the same texts (i.e. subjects and objects of the verb and suchlike), and a bar-chart on p. 1182 cleverly combines the two. At other times the presentation seems reassuringly familiar. The subtitle of this volume of OLS recalls that of the first volume of Kühner-Stegmann, Syntax des einfachen Satzen. Like its predecessor, OLS is structured as a series of chapters with numbered sections, each of which is devoted to a specific aspect of syntax, bearing its own title. Each section is illustrated by the presentation of a series of well-chosen examples with English translations; further untranslated examples and problematic passages are sometimes included in a supplement. All examples in the main text (but not those of the supplements) are listed in an Index locorum (pp. 1361-1390). Another index combines the subject and Latin word indexes of Kühner- Stegmann, allowing the reader quick access to relevant discussions. For many Classicists OLS will serve only as a fall-back in times of need and the superficial similarity between the two works may lead to the impression that OLS is an updated translation of Kühner-Stegmann, with some new terminology, pie-charts and tables thrown in for spice. Such users may not even notice that OLS has chapters with unfamiliar titles such as Verb frames or Sentence type and illocutionary force , but instead they may lament the absence of a self-contained section devoted to the grammatical category of number or the reduction in space given to prepositions, from a hundred and seven pages in Kühner-Stegmann to just over seven (pp. 1227-1235). In order to get the most out of OLS it is important, however, to appreciate its profound methodological and structural differences from its predecessor. OLS sets out to give a description of Latin within a framework which can be broadly described as functional, an approach which takes the primary purpose of language to be a communicative tool. 'An adequate grammar must take this communicative aspect into account and pay due attention to the contexts and situations in which utterances are produced.' (p. 7). The aim of OLS is to describe and explain language in use, rather than to lay down sets of formal rules. Consequently, OLS allows space to show how phrases and sentences fit into larger units of discourse and often relies on sampling of texts of different genres to reveal overarching patterns of use. Furthermore, although this does not aim to be a historical work, there are discussions of some of the major changes in Latin syntax throughout the language's history (see for example pp. 473-481 on some of the changes in the verb system, or pp. 1236-1242 on the rise of prepositional phrases in competition with bare case forms). Since Pinkster is interested in the communicative function of language, texts which he terms 'interactional' have primacy in his description of Latin. The impact of this can be shown by a rough calculation of a sample of authors featured in the Index locorum (citations were counted to the nearest half-column, and the figures given express this as a percentage of the total; for comparison a similar calculation is given in brackets from the much larger Index locorum to both volumes of Kühner-Stegmann):2 Augustine 0.6% (0.025%); Caesar 3.3% (4.3%); Cicero 27.8% (33.8%); Horace 0.6% (1.8%); Livy 4.4% (9.2%); Ovid 1.1% (1.9%); Plautus 32.3% (9.4%); Tacitus 1.1% (3.6%); Terence 5% (3%); Tertullian 1.1% (0.1%); Vergil 1.6% (2.6%); epigraphic and documentary texts 2.2% (0.25%). These figures reveal the major part played in OLS by our best source of interactional Latin, comedy; passages from Plautus feature in almost a third of all translated citations. This comparison also shows the greater attention to non-literary and later Latin texts, and the consequent demotion of historians and poets. On the above figures, the Augustan poets and Livy occupy proportionately half the space they did in Kühner-Stegmann. The framework of functional grammar brings with it new terminology and concepts. Some of this may be initially off- putting, but the explanations are usually clearly written for the neophyte. For example, section 6.24 has the foreboding title Interrogative sentences with an indirect directive illocutionary force (p. 345). The opening sentence immediately explains what this means in a clear and approachable manner: 'Interrogative sentences can be used to order or invite someone to do something, or to stop doing something.' Despite the subtitle of this volume, OLS also pays greater attention to passages of text longer than the clause or sentence in order to present how a grammatical feature functions in a wider context. Thus Chapter 7, on tenses and moods, (pp. 379-671) and the section of Chapter 11 on pronouns (pp. 1118- 1176) contain citations of twenty lines or more in order to illustrate how tenses and pronouns contribute towards the structuring of discourse. In another departure from the general practice of his predecessors, Pinkster is not afraid to leave some matters open, and to present the reader with some of the scholarly discussion. There is a generous but discerning citation of relevant books and articles, and some users of OLS might be surprised to learn just how much has been written about even quite obscure points of grammar. Sometimes, however, this practice may lead the reader into the dark about what Pinkster himself thinks, or needing to make a further trip to the library to seek the full explanation of a passage (see, for example, the discussion of subjunctive tense forms in result clauses on p. 574, or the discussion of Example (d) on p. 646). The new approach to Latin syntax in OLS brings with it many benefits. The emphasis throughout is on explanation, rather than simple categorisation and labelling. This can be seen clearly in the treatment of the Latin cases and prepositions. Where Kühner-Stegmann lists functions sequentially according to each case or preposition, in OLS the discussion of case and prepositions after verbs is mostly encompassed in a wide-ranging chapter entitled Verb frames (pp. 71-229), which is organised around the number of arguments occurring with various verbs. Arguments are here understood to mean the elements which are obligatory to fill out the meaning of a verb, and which may be expressed by case or by prepositional phrases. Latin do 'I give' and pono 'I put' both have three arguments: the first two are subject and object for both verbs, but for do the third argument, the recipient, is in the dative, and for pono the third argument, the location, is expressed by a prepositional phrase or adverb. This chapter also includes revealing discussions of the difference between different constructions such as the choice between the accusative and dative or the accusative and ablative after the verb dono (p. 152). The reader sometimes will have to work harder, and usually will not be aided by a handy but meaningless grammatical label, but perhaps the gain in understanding will be worth the effort. Chapter 7, The semantic values of the Latin tenses and moods, marks a number of departures from Kühner- Stegmann, most notably in the section on the indicative tenses. As Pinkster states explicitly on pp. 391-392, other grammars build up their theories of tense and aspect by close attention to narrative texts, principally Roman historians, but the consideration of a wider body of texts enables Pinkster to put forward different conclusions, bolstered by frequency counts on select genres, which are presented in tabular form on p. 406. One striking claim is that the historic present, rather than being specifically marked as a stylistic device, is 'the narrative tense par excellence in Latin stories' (p. 405). Elsewhere in this chapter, he casts doubt on the view that there are two distinct values of the Latin perfect, one like a Greek perfect or English have past, expressing the resulting state of an action, the other like a Greek aorist or English simple past, referring to actions in the past. For Pinkster a speaker using the perfect 'asserts that an event has taken place before the moment he is uttering the assertion' (p. 442). This may lead to the special interpretation of the perfect that it describes a resulting state, but this interpretation is rarely found with active perfects, and then only with certain verbs, although this meaning is widespread with the perfect passive. As a concomitant to this, Pinkster does not see the examples of primary sequence with a perfect main verb as indications that the perfect here has its 'resulting state' meaning, but rather as evidence that 'adhering to the sequence of tenses is basically an optional strategy' (p. 559). The disadvantage of the functional approach is that of any new paradigm: those brought up using a different terminology have to go back to the classroom. The copious index attempts to steer the reader familiar with the old set of terms to the right places to learn what is new, but this is not entirely successful. Take, for example, the term accusativus Graecus. A reader of Virgil might find a reference to this term in a commentary, and want to look it up. The index refers to three pages, p. 244, p. 267 and p. 1076; the first two passages address the topic of accusatives with passives, the third accusatives with adjectives. The index does not give any indication of where to find the best general discussion of the topic, in the section entitled 'Respect adjuncts', in Chapter 10, Satellites, pp. 915-917, a location that will not be obvious to most classicists (although there is a cross-reference to this passage at p. 244). Pinkster's argumentation shows clearly that several completely different phenomena are sometimes lumped together as accusativus Graecus, and he separates accusatives after the passive of verbs of dressing, such as induo (which he terms an 'autocausative'), from those after verbs of hitting or wounding. It is a shame that many readers will not have the patience to find out these insights. Perhaps a comprehensive index to the whole grammar, once it is completed, might take special care to highlight the best entry points to the discussion of those topics that appear at several different places across the work. Similarly, OLS is not recommended as a recourse to those who want to remind themselves quickly of Roman dating formulae, which feature variously at p. 835, p. 841 and p. 1229, even though there is useful information at all of these places. The proofreading and cross-checking of this volume is generally of a high standard, and few readers will be troubled for long by any of the remaining slips (which are largely, as far as I could see, on the level of the choice of type face, such as p. 17 example (b) where Ostium should be Ostium and angiporto should be angiporto, or p. 74 example (d) where facere should be in bold). Any grammatical work of this size and scope will inevitably contain some minor errors and places where it is possible to disagree with the explanation or interpretation of a particular passage. These should not detract from the massive achievement that OLS represents. English is possibly the only other language for which there is a comparable modern description of syntax in scale and insight to match OLS (Quirk et al. A Comprehensive Grammar of the English Language,3 a work cited by Pinkster as an inspiration, p. xxxi). OLS will undoubtedly and deservedly be a source of pleasure and pride for Latinists for years to come.
Notes:
1. Raphael Kühner and Carl Stegmann, Ausfürliche Grammatik der lateinischen Sprache. Satzlehre. Revised by Andreas Thierfelder, 3rd ed. (Hannover, 1955).
2. Gary S. Schwarz and Richard L. Wertis, Index locorum zu Kühner-Stegmann "Satzlehre". (Darmstadt, 1980).
3. Randolph Quirk, Sidney Greenbaum, Geoffrey Leech and Jan Svartvik A Comprehensive Grammar of the English Language. (London, 1985).