Chapter One Introduction 1.1. Problems of Aspect Study Aspect is a natural language category which has long since attracted the intellectual energy of linguists, philosophers and logicians, but is nevertheless as open as ever to yet another effort and another look. Being a complex category, interacting with a number of subsystems of the language, Aspect has been studied from many different angles, and this has given birth to a large number of theories. While Y. Maslov 1 sees Aspect as a binary opposition, crowning a Telicityrelated hierarchy of Action Modes, K. Karolak 2 views it as the elementary building material defining a small set of situation types. Other authors directly relate Aspect to lexical groups of verbs. J.-P. Desclés 3 presents the category in the same topological framework as Tense, Correlation and Voice. A unity in the semantics of Aspect and some Tense forms is maintained by J. Lindstedt. 4 H. Verkuyl 5 proposes a compositional model of Aspect, based on the principle of syntactic asymmetry, and postulates an important relation between Aspect and NP-quantification. There have even been attempts to view Aspect as a Slavonic equivalent of the article! 6 In Chapters Four through Six of this book, we take up some of these hypotheses and test them against Bulgarian data. The challenge presented by the category of Aspect and its description is not simply due to its inner complexity or the complexity 1 Cf. Maslov, 1984. 2 Karolak, 2001. 3 Desclés, 1990. 4 Lindstedt, 1985. 5 Verkuyl, 1993. 6 Kabakchiev, 2000. 15
of its interrelations with other categories. It is largely defined by the localisation of its grammatical expression in only part of the Indo- European language family, not the best studied one, and in the scantiness of sufficiently full presentations of Slavonic data in other languages, including English. A considerable part of the controversies in aspectual theory may thus well be due to the incomplete presentation, even misrepresentation, of the primary data, and thus to the formulation of hypotheses on partial or erroneous material. It is my belief that a comprehensive, sufficiently detailed and faithful presentation of Aspect in a language where it finds grammatical expression independently and in its interaction with syntactic and grammatical context, can form a considerable step towards the creation of a generally accepted, working formal model of this category. One of the ambitions of the present work is, therefore, before anything else, to offer a sufficiently full and objective presentation of Bulgarian Aspect and Aspect-related phenomena. Chapters Two and Three include data, drawn from a non-selective multilingual corpus of non-constructed original texts and their published translations. 7 Chapters Four and Five offer a re-view of the Bulgarian verb classes, set out by the Bulgarian linguist Kalina Ivanova 8 on the basis of an analysis of the complete verbal stock of Bulgarian (as presented in a three-volume academic dictionary of the language). The English translations of the Bulgarian verbs in Chapter Four and Chapter Five are drawn from entries in a comprehensive Bulgarian-English dictionary. 9 7 Cf. for descriptions of the Trilingual Electronic Corpus of the Electronic Archive of the Bulgarian Language, Stambolieva and Aleksandrova 1995, Stambolieva 1996 a., 1996 b. 8 Ivanova, 1974. 9 Cf. Bulgarian-English Dictionary, 1975. 16
1.2. Universalist Approaches Not all linguists consider Slavonic data to be vitally important in formulating hypotheses on the semantic content of Aspect and its status in the system of language. Such a position is legitimised by the universalist approach to language phenomena. The universalist hypothesis is founded on the (quite reasonable, in fact) assumption that language content is translatable. If sentences are translatable, there must be a semantic representation, common for an open set of language pairs. The representation would thus acquire universal character: Sentence LA > semantic content > Sentence LB, LC, LD etc., Figure 1 where the semantic content is assumed to remain unchanged under translation from a source language A to languages B, C, etc. Even if a grammatical category of language A is not present in another language, say B, it is hypothesised that B can still express its semantic content, by means of functional equivalents of different types. In the context of this hypothesis, we can speak of Aspect in a broad sense. Thus, on the basis of the grammatical status of the verb in a Bulgarian sentence, we can assign the sentence Perfective or Imperfective Aspect. This information can be presented as forming part of the underlying semantic frame of the sentence. If this sentence is translated in, say, English, the expression of Perfectivity or Imperfectivity will be taken up by the means available in the system of the English language. The universalist hypothesis, though broadly speaking correct, should not, however, be given too strong a reading. Translatability and translation adequacy do not necessarily mean one-to-one correspondence. Information is inevitably lost when a target language possesses a less fine-grained system for the encoding of a certain semantic content than its language source. 17
The universalist drive does not stop at the idea of translatability. The argument proceeds as follows: Knowing that every Slavonic sentence inevitably expresses Aspect; having further established that every sentence from a source Slavonic language can be given an adequate translation in any other language, it is concluded that Aspect is a universal category. What is meant by universal category is: a category which can be expressed in every individual language (including, of course, the native language of the linguist). Since many linguists adopting this approach are not familiar, or vaguely familiar, with languages where Aspect is grammaticalised, the focus of their work is inevitably shifted from attempts to establish the functional equivalents of Aspect in their respective languages to rather abstract speculations on what Aspect could actually be. The idea of presenting a description of Universal Aspect, starting with the world of free ideas before moving on to the specific data of each language is extremely attractive, yet it seems to me somewhat unrealistic, because of the nature of the relation between language and thought. This relation is not simple and one-directional: e.g. from concepts and judgements to words and sentences. Language is not simply the materialisation of thought, it is not only a means of communication, needed in the absence of direct mechanisms of thought-accessing. Language is often referred to as the direct reality of thought because it has a leading role in thought-formation itself. The creation of concepts, the formulation of judgements may well be impossible without the material form of language of a concrete, specific natural language. Hence the difficulties of foreign language acquisition and, in the context of this book and of my own professional experience, the often unsurmountable difficulties that foreign learners (particularly those whose mother tongue is not Slavonic) usually have not only in the correct generation, but even in understanding some Bulgarian Aspect-Tense forms or complex Action Mode combinations. Without of course denying the possibility to carry out contrastive investigations on the basis of preliminary semantic classifications, I still maintain that particularly in the case of Aspect it is rewarding to precede such investigations by a careful study of the manifestations 18
of the category in languages where it takes material form. Such considerations largely motivated the data-oriented approach of the present book. 1.3. Aspectual Studies and the Bulgarian Language Bulgarian is not, simply, one of a number of languages of similar structure where Aspect is grammatically expressed. In fact, it is not even a typical representative of the Slavonic language group. It has, however, long attracted the interest of aspectologists because of the interplay of two factors: First, Bulgarian not only possesses a full-blown system of Action Modes, but it also manifests the Perfective / (secondary) Imperfective Aspect opposition to the highest degree of regularity and grammaticalisation within the Slavonic language group. Speaking of the opposition Perfective vs. derived (secondary) Imperfective forms and of the regularity of the Bulgarian morphological mechanism of the secondary PA / IA opposition, the eminent Russian aspectologist Yu. Maslov notes: It should not be thought that the principle of the positive suffixal expression of the Imperfective Aspect and the negative, null expression of the Perfective Aspect forms an exclusive feature of the Bulgarian language area [...] However, it is precisely in the Bulgarian language area that this principle has found its fullest and most consistent development. The specifics of the Bulgarian system in this respect (unlike several other respects, noted above) is not in the deviation from the Slavonic language type, but in the fullest expression of the development tendencies, built in the Slavonic grammatical system [...] 10 Generally speaking, the specific development of Bulgarian with respect to its verbal system consists in a tendency towards the preservation and specialisation of categories, the generalisation of semantic content and the regularisation of form. 10 Maslov, op. cit., p. 97 (translation mine). 19
Second, due to the very specific conditions of its development, Bulgarian has preserved all the categories, often presented as the functional equivalents of Slavonic Aspect in non-slavonic languages. Bulgarian has preserved the Indo-European opposition, in the Preterit, between the Aorist and the Imperfect; it has preserved the category of Correlation, acting in the domains of the Present, Past and Future; it has a rich system of Moods. Further, in the nominal system, Bulgarian has developed an article. Because of the presence of these categories, Bulgarian Aspect does not take up the functions of Tense or Correlation; it cannot be said to be a compensating mechanism for the expression if definiteness, either. In Bulgarian, Aspect can best be seen to function as a category per se, without interference from other categories, but in simultaneous functioning with them. 1.4. Structure of the Book Apart from this first, introductory chapter, the book contains five central chapters and a concluding chapter. Chapter Two is not directly related to the problem of Aspect. Its first section offers a brief introduction into the ways, in which Bulgarian differs from the other languages of the Slavonic group and the factors which led to this specific development. The concise presentation of Bulgarian phrase structure in the second section of the chapter is intended to facilitate the understanding of the Bulgarian examples and to provide a background for the assessment of the phenomena which are in the focus of the book. Chapter Three presents an overall view of Aspect as a semantic category, of its relation to lexical, syntactic and grammatical context and of its specific expression in Bulgarian, in contrast to English. The regularity of the opposition between the Perfective and Imperfective Aspect within one lexical group of verbs defines Bulgarian as an excellent source of data for the study of Aspect in the context of verb classes. Chapter Four summarises some influential 20
views on the relation between Aspect and Verb Class and proposes a new look on the definition of and relation between Aspect, Situation Structure and Action Mode. In Chapter Five, I re-view situation types and lexical verb groups from the point of view of the eventual syntactic changes accompanying the processes of Perfectivation or Imperfectivation. The corpus of English translations is checked for syntactic patterns related to the expression of the semantics of different Situation Structure or Action Mode types. Many grammatical categories have been claimed to contribute aspectual information: Taxis, Tense (and, particularly, the Preterit opposition between the Aorist and the Imperfect), Voice, Correlation, Mood (and, particularly, the Non-Evidential Mood). Chapter Six is a presentation of the grammatical categories of the Bulgarian verb as possible or impossible contexts for the insertion of Perfective or Imperfective verbs. Finally, Chapter Seven is a concluding chapter, containing a summary of the discussions. 21