This is the first study that empirically investigates preposition placement across all clause types. The study compares first-language (British English) and second-language (Kenyan English) data and will therefore appeal to readers interested in world Englishes. Over 100 authentic corpus examples are discussed in the text, which will appeal to those who want to see 'real data'
AUTHOR: Bortone, Pietro TITLE: Greek Prepositions SUBTITLE: From Antiquity to the Present PUBLISHER: Oxford University Press YEAR: 2010
Coulter H. George, Department of Classics, University of Virginia
In this book, Bortone (henceforth B.) traces the development of the Greek prepositional system over the course of its entire 3000-year history. The first third of the book deals with general theoretical issues about the syntactic and semantic description of prepositions, cases, and the relationship between them. B. devotes particular attention to the localistic hypothesis (the idea that spatial uses are primary), a concept that provides a consistent orientation to the remainder of the book, which takes synchronic snapshots of four different periods of Greek (Ancient, Hellenistic, Medieval, and Modern) in order to establish the bigger diachronic picture.
Part 1: Background to Greek prepositions
Chapter 1: On the function of prepositions (pp. 3-32)
In the first chapter, B. looks at the syntax of prepositions, in effect to establish the boundaries of his study. On the one hand, he must deal with the relationship between prepositions and cases (and, to a lesser extent, between prepositions and postpositions). From the practical standpoint of work on a mainstream Indo-European language, the two categories can be kept apart fairly easily (e.g., prepositions occur only once at the start of a noun phrase, whereas case endings are repeated on each relevant element), but B. is well aware that such diagnostic criteria often apply less well in other language families. He must also deal with the fact that, especially in Ancient Greek, prepositions can have considerably different meanings depending on the case that they govern. In his view, this is not much of a problem, as one can simply describe the combination of preposition plus case as a single discontinuous morpheme that functions in parallel to either a simple case or a simple preposition. Finally, because of the particular structure of the Modern Greek prepositional system, in which there are few ''proper'' prepositions, with much of the functional load carried by compound prepositions (forms analogous to English _out of_, in which a first, adverbial element narrows down the meaning of the second, prepositional element), B. argues in favor of treating such compounds as true prepositions in their own right.
Chapter 2: On the meaning of prepositions (pp. 33-85)
Like cases, prepositions vacillate between having a clear meaning (_he fled to Paris_ vs. _he fled from Paris_) and seemingly being semantically empty (_different to_ vs. _different from_). At the same time, as B. shows with examples from an impressive number of languages, certain patterns recur: locative expressions are often related to comitative ones, which in turn are frequently bundled together with the instrumental; the locative is also close to the allative, which in turn often expressed the recipient. To create some descriptive order out of this apparent chaos, B. turns to the localistic hypothesis: as he shows quite convincingly, one gains a lot by positing that the spatial meanings are primary. After all, these are the meanings that children acquire first, the meanings that are most likely to be expressed consistently from language to language, and, in what is one of the main points of the book as a whole, they are the first meanings to arise diachronically when a new preposition is born. Only later, along various cross-linguistically common semantic pathways, do the more abstract senses develop.
Chapter 3: On the development of prepositions (pp. 86-106)
In the next chapter, B. takes a step back to look at how prepositions are created in the first place. Members of just about any word class can turn into them, as B. details: verbs (_regarding_, _considering_), adjectives (_near_), univerbations (_among_), borrowings (Spanish _hasta_ < Arabic _haṭṭa_). But B. devotes the most space to the prepositions that develop from nouns. While the precise etymology of the most basic Greek prepositions must inevitably remain somewhat speculative, there are very good reasons for viewing many of them as ultimately derived from inflected nouns: the _-i_ on the end of _peri_ ''around, about'' (found in other prepositions as well) gives it the appearance of the locative case of a Proto-Indo-European (PIE) noun. What is more, Greek renews its prepositions in the same way: Classical _peri tou kēpou_ ''around the garden'' later gives way to _kuklōi tou kēpou_, where the first element is transparently the dative case of the noun for ''circle''.
Part 2: The history of Greek prepositions
Chapter 4: Prepositions and cases in Ancient Greek (pp. 109-70)
In the second part of the book, B. turns to the description of the Greek prepositional system, starting with the situation in Ancient Greek. First, he gives a general overview, outlining the Greek case system and its PIE origins, and listing and defining the so-called proper prepositions. In short, these are the older stratum of prepositions, which form the core of the system (the Greek equivalents of ''to'', ''from'', and the like) and share the syntactic property of also being used as verbal prefixes. But B. also argues strongly against neglecting the improper prepositions: the stepchildren in the prepositional family, these correspond to e.g. English ''following'' or ''opposite'', and are often sidelined in grammars as newer and somehow less central to the system--unjustly, since many of them, over time, come to replace the proper prepositions. Indeed, one general lesson to be drawn from B.'s work is the importance of liminal categories like compound and improper prepositions: easily passed over in synchronic studies, they are frequently the interesting transitional categories in diachronic development.
After this initial survey, B. examines the use of cases and prepositions in Homer, the earliest Greek corpus substantial enough to allow a syntactic study of this sort (generally dated to the eighth century BC, but containing many archaisms as well). In keeping with his general interest in marginal features, B. also gives a full treatment of relic endings like ablative _-then_ and allative _-de_, which sit awkwardly at the boundary between cases and adpositions. A cursory discussion of the syntactic ambiguity of the Homeric prepositions follows (it is often unclear whether they are adverbs, adpositions, or preverb), but for a full discussion of this issue, one should still consult Horrocks 1981. What is primarily important for B. is the fact that the prepositions are largely facultative at this point: the case system still bears the brunt of distinguishing between allative, locative, and ablative, and the prepositions often serve simply to add dimensionality to the construction. For example, the dative case on its own can mark location, so adding the preposition (or is it an adverb?) _hupo_ simply specifies further that the location in question is the one underneath the landmark noun in the dative. In pp. 153-66, B. summarizes the main diachronic trends that can be traced as one moves into the period of Classical Greek (5th-4th centuries BC). Probably the most fundamental change is the shift of functional weight from the case ending to the preposition itself: the dative case is on its way out, and there is a blurring of the nuances conveyed by the different cases after several of the individual prepositions.
Chapter 5: Prepositions and cases in Hellenistic Greek (pp. 171-94)
The next stop on B.'s diachronic tour is Hellenistic Greek--broadly speaking, the Greek that spread throughout the Eastern Mediterranean with Alexander the Great's conquests in the late 4th century BC. Also familiar under the label Koine (the ''Common'' dialect; B. uses the terms more or less interchangeably: p. 171 n. 2), this variety of Greek is most famously associated with the language of the Septuagint, the Greek translation of the Hebrew Bible, and the New Testament, and these are the texts that B. uses as the basis for his snapshot as well. They are problematic texts for linguistic research, however, because their language has clearly been influenced by Hebrew and Aramaic--though the exact extent of that influence remains a matter of debate (see George 2010). Still, there are some diagnostic tools that can help determine whether or not prepositional usages in the Septuagint and New Testament reflect the general development of Hellenistic Greek, such as their presence in other texts of the period not subject to interference from Semitic or indeed their continued life in Medieval and Modern Greek. What remains, after accounting for the effects of the substrate, is the continuation of the changes already underway in the Classical period: the decline of some of the older proper prepositions (e.g. _ek_ ''out of'' is increasingly replaced by _apo_ ''from''), the increase in use of the new improper prepositions (e.g. _epi_ ''on'' gives way to _epanō_ ''on top of''), and further decline in the dative case.
Chapter 6: Prepositions and cases in Medieval Greek (pp. 195-237)
With this chapter, B. is able to shed light on some particularly murky territory. Medieval Greek has not received nearly as much scholarly attention as Ancient or Modern Greek, largely because of the difficulties of establishing a corpus that reflects real developments in the language. Nearly all Greek of this period is subject to varying degrees of interference from the Classical standard, so it is particularly challenging to work out whether a particular prepositional use is still a living part of the language or just a learned habit of the educated: diglossia, par excellence. Still, the situation is not hopeless. B. shows that there is much to be gained by looking at some of the more colloquial works, like the twelfth-century Ptochoprodromos poems, as well as ''translations'' of higher-register works into more vernacular language, such as a thirteenth- or fourteenth-century metaphrase of Anna Comnene's _Alexiad_ or Sophianós' fifteenth-century metaphrase of Plutarch's _Peri paidōn agōgēs_. The same trends continue and are now even more advanced. The dative case has fallen into even greater disuse: the indirect object is now often marked by _eis_ (earlier illative ''(in)to''). Even the distinction between the genitive and accusative after prepositions is now being lost, with the accusative gaining ground as the default case after all prepositions. Further semantic mergers between the older prepositions take place (_en_ ''in'', _pros_ + accusative ''towards'', _epi_ + accusative ''(on)to'', and _eis_ ''(in)to'' all collapse together as _se_, the phonological outcome of _eis_), and new prepositions (like _konta_ ''near'') are created in their place. Also important: most of the older prepositions increasingly have only non-spatial uses, while the newer prepositions are primarily spatial.
Chapter 7: Prepositions and cases in Modern Greek (pp. 238-301)
After discussing two theoretical problems with doing linguistic work on Modern Greek (continued diglossia, and the difficulty of determining whether a given shift in usage should be attributed to the Balkan Sprachbund), B. first looks at the Modern Greek case system, the main feature being the complete loss of the dative (apart from a few fossilized phrases). The bulk of the chapter, however, covers the prepositions, in particular the proper analysis of the numerous compound prepositions, as Greek has shifted from a system in which case marks direction (''to'', ''from'', ''at''), and prepositions specify dimensionality (''on'', ''next to'', ''under'') to one in which all but the most basic spatial relations are expressed with a compound, in which the first, adverbial element indicates dimensionality, and the second element, a proper preposition (usually _apo_ ''from'' or _se_ ''at, to'', but also sometimes _me_ ''with'' and _gia_ ''for''), plays a role in some ways similar to that of case in the ancient system. Some first elements occur exclusively with one second element (just as some of the ancient prepositions occur with only one case), but others can occur with different second elements depending on the nuance expressed (e.g. _panō se_ ''(up)on'' vs. _panō apo_ ''above''). What is more, these compound prepositions, originally purely spatial, have taken on abstract meanings in Modern Greek. Finally, B. considers the fate of the ancient prepositions that have mostly died out in the modern language, but occasionally still survive in limited contexts, showing that the spatial uses are entirely defunct (and giving the modern equivalents) and that it is only in figurative expressions that they survive at all.
All in all, this is an extremely impressive book. B.'s main point--that new prepositions are created to mark spatial relations, gradually acquire non-spatial meanings, then lose their original spatial meanings--is very convincingly argued, and anyone interested in the grammaticalization of prepositions (and case) will find it an indispensable source of examples. Particularly noteworthy is the support for unidirectionality: in all his copious data, B. finds no evidence that a preposition ever acquired spatial meanings after starting out as a marker of a non-spatial relation. B.'s linguistic breadth is also remarkable: it is clear, especially from the the first half of the book, that he is extremely comfortable with scholarly literature on cases and prepositions in a wide range of languages and time periods. Citations from ancient Greek grammarians are to be expected, but Rabbinic writings and the views of Leibniz rather less so.
That said, there are some shortcomings--which, let me hasten to add, I discuss at length not because they represent serious flaws (they certainly do not), but because they raise interesting points. For the most part, they are the inevitable result of attempting a full-scale history of Greek prepositions, combined with a generous theoretical introduction, all in only three hundred pages. While his arguments for the localistic hypothesis, for instance, are on the whole well-grounded, occasionally some of the rough edges have been smoothed off a bit too much. In his table of the Ancient Greek ''proper'' prepositions (p. 117), he lists the commonest meanings of _dia_+accusative as ''all over; on account of''. This misleadingly overstates the localist position, however, as the spatial sense of _dia_+accusative is found only in poetry (LSJ s.v.). Another oversimplification is his statement that, in the Classical age, ablatival genitives are rare, whether with or without prepositions (p. 127, 146 n. 66). This is certainly true of the use of the genitive on its own, as well as with many prepositions (including those B. highlights in this connection, _kata_ and _hupo_), but certainly not across the board: _apo_, _ek_, and _para_+genitive are all used ablativally extremely frequently. Along the same lines, B., while absolutely right to rehabilitate the improper prepositions, probably goes too far in the other direction in downplaying the difference between them and their older counterparts. He writes: ''[T]he supposedly immense difference between 'proper' and 'improper' prepositions is essentially a matter of age'' (p. 141). Well, yes, but age matters: the difference between a case ending and a postposition--or, indeed, the noun from which that postposition developed--is also ''only'' a matter of age, but surely it remains useful to distinguish these categories.
Another problem inevitable with attempting so broad a study on a historical corpus is the necessary inclusion of texts from very different genres that can and do use prepositions in very different ways. Now B. is clearly aware of this pitfall and generally avoids it, but some misleading examples still slip in: example 4.109 (p. 157) uses a line from Sophocles' _Antigone_ (1123-4) to illustrate the fact that _para_+genitive (normally ''from'') could have the same locative spatial sense that this preposition also has with the dative and accusative. But this line is not only textually corrupt (some manuscripts have the accusative singular rather than the genitive plural), but also occurs in a choral ode--the sort of passage where the Athenian tragedians were particularly prone to stretch language beyond its normal semantic and syntactic limits. (For a general study of how the syntax of tragedy, including spatial use of the dative and accusative cases, is different from that of prose, see Bers 1984.) It is also strange that B. chooses to argue for the decline of _en_ ''in'' on the basis of its occurring less often in the Hellenistic epic poet Apollonius of Rhodes than in Homer (p. 193). True, Apollonius did model his language on Homer's, but often by exaggerating archaic features, so his less frequent use of _en_ could easily be due to his desire to avoid what in his eyes was a comparatively prosaic preposition. (For Apollonius' archaizing use, relative to Homer, of the preposition _hupo_+dative, see George 2005: 73-4.) This mixing of evidence from different genres has wider implications when one looks at the argumentation on pp. 147-9, where B. demonstrates the richness of the Ancient Greek prepositional system by listing all the different Greek prepositions that could express a single given prepositional phrase in English. At first glance, it is impressive that he can give no fewer than nine equivalents for ''out of fear''--and, indeed, his overall main point, that there was a lot of synonymy, ripe for simplification, is certainly correct. But one wonders whether it wouldn't also be possible to say the same of English (where one can act not just ''out of fear'', but also ''from'', ''in'', ''through'', ''with'', ''for'', or ''by fear'', depending on various contextual factors)--especially if we do not just restrict ourselves to 21st-century vernacular, but to all genres of poetry and prose from the past five centuries. For, of the examples given by B., five come from poetic texts, four from prose, and they range in time from the Iliad to Isocrates (4th century BC).
Along the same diachronic lines, the exact timing of some of the shifts in usage that B. notes might not always be clear to a reader who wasn't already familiar with the basic history of Greek. Comparing pp. 145 and 156, we find that B. takes the position that the semantic force of the cases governed by prepositions is vigorous in Homer and much weaker in Classical Greek; this is indeed eminently sensible. But it then seems odd that, in exx. 4.111 and 112 on p. 157, he uses constructions with _peri_+dative and +accusative from *Homer* to illustrate the ''broad synonymy between syntagms composed of _peri_''; the same holds true of exx. 4.123 and 124, on p. 159, where Homeric examples of the equivalence of _meta_+dative and +genitive are given. Now B.'s glossing of the _peri_ examples makes clear that there is a difference between the two, but a slightly more nuanced take on the Homeric prepositions (e.g. distinguishing between those prepositions where it still made a significant semantic difference which case was governed and those where it didn't) could have bolstered the overall argument (the latter prepositions, like _meta_ and _peri_, in contrast to _para_, were, it seems, particularly prone to cease construing with the dative already in the Classical period).
A bit more discrimination in assessing the relative prevalence of various prepositional usages would also have been useful in the list (p. 205) of ''proper'' prepositions found in Porphyrogenitus' _De administrando imperio_ (10th century AD), with which B. illustrates the ''shrinking'' of the range of the older prepositions in this period. Of the eighteen in Classical Greek, fifteen, according to B.'s list, are still found, which already seems a rather small drop. What is more, B. curiously omits two prepositions that *are* still found in Greek of this time (_dia_ and _meta_), so the only real ''loss'' is _amphi_, which was already absent from Isocrates, thirteen centuries earlier (see B.'s chart on p. 122). Real evidence for the reduction of the proper-prepositional options open to Porphyrogenitus thus only truly emerges from some of the evidence presented later in the chapter (e.g. the fact that _huper_ was no longer used in a spatial sense, as shown by the charts on p. 229).
There remains one final quibble: while he does provide translations, B. does not transliterate or gloss the vast majority of his Greek examples--or, indeed, much of the Hebrew. Whether or not one should do so is a tricky call and largely depends on the audience one wishes to reach. After all, what is useful in a work aimed at a wider community of linguists comes across as off-putting in one aimed at classicists. But given B.'s interest in the big picture, a presentation more like that of Horrocks 2010 or, even more relevant, Luraghi 2003 would perhaps have been preferable. As it stands, linguists unfamiliar with Greek who wish to use B.'s book are strongly advised to start with Luraghi's work--an admirable introduction to the cases and prepositions of the ancient language--before turning to B. for a splendid survey of what happens to those prepositions in the centuries that follow. They will not be disappointed by the wealth of data that Greek, from antiquity to the present, has to offer.
Bers, V. (1984) _Greek Poetic Syntax in the Classical Age_. New Haven.
George, C. H. (2005) _Expressions of Agency in Ancient Greek_. Cambridge.
George, C. H. (2010) ''Jewish and Christian Greek'', in _A Companion to the Ancient Greek Language_, ed. E. J. Bakker, Malden, MA: 267-80.
Horrocks, G. C. (1981) _Space and Time in Homer: Prepositional and Adverbial Particles in the Greek Epic_. New York.
Horrocks, G. C. (2010) _Greek: A History of the Language and its Speakers_, 2nd edn. Malden, MA.
LSJ = Liddell, H. G. and R. Scott, rev. H. S. Jones (1940) _A Greek-English Lexicon_, 9th edn. Oxford.
Luraghi, S. (2003) _On the Meaning of Prepositions and Cases: The Expression of Semantic Roles in Ancient Greek_. Amsterdam-Philadelphia.
ABOUT THE REVIEWER
ABOUT THE REVIEWER:
Coulter H. George is Assistant Professor of Classics at the University of
Virginia and is the author of _Expressions of Agency in Ancient Greek_
(Cambridge University Press, 2005). His research interests include the
syntax of the Greek verb, particles and prepositions, and contact phenomena
between Greek and the other languages of the ancient Mediterranean.
Currently, his main project is a book on expressions of time in Ancient Greek.