by Robert S. D. Crellin (University of Cambridge)
The semantics of the Ancient Greek perfect active, morphologically directly inherited from the Proto-Indo-European form of the same name, has long been a matter of scholarly contention. What makes this verb form semantically so interesting is that for most of the duration of Ancient Greek it seems to combine several tense-aspect and diathetical features, namely anteriority, resultativity, stativity, and detransitivisation in causative predicates, which are in many languages conveyed by separate morphological means. To put this in concrete terms, in Ancient Greek you would use a perfect active form to render both English ‘I have made a chair’ and ‘I stand’. This has led to a variety of approaches to analysing the perfect’s semantics, to varying degrees either positing lexicalisation of apparently aberrant readings, or generalising a particular aspect of the semantic behaviour of the perfect, but beyond what the data as a whole can support.
My recent monograph, The Syntax and Semantics of the Perfect Active in Literary Koine Greek, seeks to address the question of the semantic description of the Greek perfect head-on, on the basis of the plentiful but understudied corpus of literary post-Classical Greek (c. 300 BCE – 300 CE). After some theoretical preliminaries, the study first assesses the full range of behaviours of the perfect in this corpus according to verb and predicate type, before building up a semantic description of the perfect from which the perfect’s diverse interactions with different predicate-types can be derived. This is achieved by adopting insights inter alia from Tenny’s Aspectual Interface Hypothesis (Tenny 1994) as well as the generative tradition, to provide an account of the perfect’s semantics which thoroughly integrates its argument structure relations with its tense-aspect denotations. It is demonstrated that it is possible to formulate a semantic description by which one may predict, with a predicate of known semantic properties, how the perfect will be read. Specifically, the perfect derives a homogeneous atelic eventuality from the predicate which it heads.
The simplest case is that in predicates which are themselves homogeneous and atelic, such as states, the perfect may simply derive another state from the predicate:
|‘He turned down a quiet narrow passage…’ (Jos. AJ 19.104, monograph p. 50)|
However, since the negative of a state may also be said to have the same event structural properties, a cancelled state is also a valid reading of the perfect, e.g.:
|‘Phregellai… a city… which previously had the majority of the places just mentioned as dependent towns.’ (Strabo 5.3.10, monograph p. 236)|
Indeed, it may be said more generally that the negative of any eventuality is homogeneous and atelic (see e.g. de Swart 1996: 229 who, following Verkuyl 1993, takes such predicates as states). This allows the perfect to derive such an eventuality from predicates where the subject is not presented as entering into a state, by instantiating the predicate and locating the subject in the homogeneous and atelic post-situation e.g.:
|‘… which is exactly what I have now done.’ (Jos. AJ 12.213, monograph p. 17)|
The relationship between argument structure and tense-aspect denotation becomes clear in labile change-of-state predicates. Here the perfect active may have resultative or anterior semantics according to its syntax. Read as detransitivising, the perfect is read as resultative, whereas when the predicate is transitive, it reads as an anterior. In each case, the perfect is seen to derive a homogeneous atelic eventuality and predicates this of the subject:
|‘[Claudius] had stood… taking cover in the darkness there.’ (Jos. AJ 19.216, monograph p. 190)|
|‘But Philip… had sheltered the Illyrians behind a hill…’ (Plb. 5.13.5, monograph p. 190)|
The monograph is a development of my PhD thesis, completed with AHRC funding at the Faculty of Classics, Cambridge, in 2012, which looked at the same question. The latter, however, is more empirically focused, using large datasets to generalise about the behaviour of the perfect from the macro-perspective. The present work presents a counterpoint to that project, by taking the micro-perspective, and seeking to establish, by delving into the underlying structure of the Greek language in this period, the semantic basis on which this form’s rather idiosyncratic behaviour might be described.
Crellin, Robert. 2012. The Greek Perfect Active System: 200 BC – AD 150. PhD Dissertation. University of Cambridge.
Swart, Henriette D. E. de. 1996. Meaning and Use of not … until. Journal of Semantics (13). 221–263.
Tenny, Carol. 1994. Aspectual Roles and the Syntax-Semantics Interface. Dordrecht: Kluwer Academic Publishers.
Verkuyl, Henk. 1993. A Theory of Aspectuality: the Interaction between Temporal and Atemporal Structure. Cambridge: CUP.
Robert Crellin’s book, The Syntax and Semantics of the Perfect Active in Literary Koine Greek, is freely accessible to members of the Philological Society via the Wiley Online Library and their membership number. Members are asked to contact one of the Society’s secretaries with any questions in this regard.