A Semantic and Pragmatic Model of Lexical and Grammatical Aspect

Mari B. Olsen

• Publisher: Routledge
• Year: 2016
• ISBN:
  9781135655136
• Edition: 1
• Topic: Filología
• Subtopic: Lingüística

First Published in 1997. This work studies two related phenomena in human language: the ability of verbs and other lexical items to describe how a situation (event or state) develops or holds in time (LEXICAL ASPECT) and the view some verbal auxiliaries and affixes present of the development or result of a situation at a given time (GRAMMATICAL ASPECT). Through this investigation the author seeks to reveal a formal situation structure represented by aspectual phenomena, a structure to which other linguistic elements make reference, particularly tense. This structure describes the semantics of aspect and provides a principled input to pragmatic aspectual interpretation.

I

Introduction

0.0 OVERVIEW

This work studies two related phenomena in human language: the ability of verbs and other lexical items to describe how a situation (event or state) develops or holds in time (LEXICAL ASPECT) and the view some verbal auxiliaries and affixes present of the development or result of a situation at a given time (GRAMMATICAL ASPECT). Through this investigation I hope to reveal a formal situation structure represented by aspectual phenomena, a structure to which other linguistic elements make reference, particularly tense. This structure describes the semantics of aspect and provides a principled input to pragmatic aspectual interpretation.

Studies of linguistic aspect have diverged on the most basic levels, concerning what aspectual properties need to be accounted for and what constitutes evidence for them. Given the variety of lexical, grammatical, semantic, and pragmatic elements contributing to the interpretation of aspect, it is not surprising that analyses—whether broadly theoretical or examinations of aspect in a particular language—differ widely. The variation between these studies masks the relations among lexical and grammatical aspect phenomena. Furthermore, few (if any) theories provide an explanatory account of the relation between aspect and tense and the tendency for each, in the absence of the other, to acquire default temporal interpretations resembling those generally assigned to the other.

In this study I examine both typological studies of aspect (esp. Comrie 1976 and Smith 1991) and tense (esp. Reichenbach 1947 and Comrie 1985) as well as case studies of individual languages. I propose that a principled perspective on both the literature and the phenomena it describes may by gained in the context of a theoretically sound linguistic framework for describing the generation of aspectual interpretations in individual languages and cross-linguistic variation between aspect systems. This chapter introduces the topic, discussing in a general way the range of aspectual phenomena and how they have been treated in the literature. I propose semantically based distinctions between aspect and tense, and between lexical and grammatical aspect. I then present three assumptions underpinning this investigation: i) that lexical and grammatical aspect represent sets of universal semantic oppositions, ii) that the semantics of aspectual interpretation is compositional, and iii) that a clear distinction may be made between semantic meaning and conversational pragmatic implicature. Upon this foundation, I motivate the primary formal mechanism of this thesis, namely privative features, as drawn from phonological feature theory, and their monotonic composition.

In chapters 2 and 3 I justify the privative feature analysis of lexical and grammatical aspect, respectively, and provide an account of how the features presuppose or fit into a temporally based situation structure. Chapter 4 shows how my analysis of grammatical aspect may be used to maintain the semantic distinction between aspect and tense, and to reduce the representation of tense to a simple relation between a reference time and a deictic center (usually the moment of speaking). Chapters 5 and 6 demonstrate that my model of lexical aspect, grammatical aspect, and tense provides a perspicuous account of English and the Koine Greek of the New Testament, respectively, including analyses for such controversial issues as the representation of the perfect forms in both languages and the temporal reference of the so-called “tense” forms in Koine. Chapter 7 summarizes the main results of the study, focusing on their relationship to current and future linguistic research.

1.0 THE PHENOMENA

In this section I briefly outline my position on the relation of grammatical aspect to phenomena with which it has often been confounded in the literature. I suggest distinguishing grammatical aspect from tense and lexical aspect, respectively, on the basis of semantic criteria.

1.1 Grammatical Aspect and Tense

Since grammatical aspect concerns time—both how situations develop over time and how speakers view situations at a specific time—it has often been confounded with tense. In fact, grammatical aspect has sometimes been called secondary or relative tense. I suggest that the temporal notions encoded by grammatical aspect (also known as verbal or viewpoint aspect) should be semantically separated from those encoded by tense, even though both concepts involve time, interact with each other and may be expressed by the same morphemes. (For the technical definition of semantic meaning as used in this dissertation, see § 2.3.) I distinguish aspect from tense in order to make a consistent analysis of aspect possible; however, the distinction also results in a more parsimonious analysis of tense (see chapter 4). The separation of tense from aspect is supported by the fact that some languages encode one but not the other. In fact, surprisingly few languages lack aspectual distinctions, although many lack tense, including Classical Arabic and Mandarin (Herbig 1896 (quoted in Thelin 1978:2); also see Dahl 1985:1; Lyons 1977:705).

In some languages the distinction between aspect and tense is morphological, as well as semantic, because grammatical aspect and tense are marked by separate affixes, as in Russian (1).¹

(1) On po-rabota-1.

He PFV-work-PAST

‘He worked a bit.’ (Smith 1991:321)

The separation of grammatical aspect and tense morphology in such languages supports the idea that aspect is also semantically orthogonal to tense, a position taken in much (if not most) of the literature (Hockett 1958 (quoted in Thelin 1978:13); Thelin 1978; Freed 1979; Chung and Timberlake 1985; Abusch 1985; Tenny 1987; Smith 1991; Verkuyl 1993). The term “aspect,” first discussed by the Greeks (Binnick 1991:135), was adapted by the 19th century Slavicists, to label morphological categories distinct from tense, such as the perfective in (1) above. (The English term translates the Russian vid, from the root meaning ‘to see’ (Lyons 1977:705; OED 1989).)

Other languages appear to have semantic categories similar to Slavic grammatical aspect; however in some of these languages aspect is carried by morphemes which also mark tense. For example, in French (and other Romance languages) aspectual contrasts are carried by the same morphemes that mark past tense, as shown in (2).

(2) Jeanpeign-ait un portrait.

Jeanpaint-PAST.IMPFV a portrait

‘Jean was painting a portrait.’ (Smith 1991:290)

Some have claimed that the English simple past tense form also marks perfective aspect as well as past tense (Brinton 1933; Smith 1991).² Thus, it is not possible, from a cross-linguistic perspective, to define grammatical aspect on purely morphological grounds,³ since it is often not morphologically separate from tense.

In this dissertation, I adopt Comrie’s (1976:2–3) semantic distinction between tense and aspect. It is based on the semantic distinctions found in languages where the two are morphologically separate, but may also be applied to languages where they are not, such as French, Spanish, and English. Comrie’s distinction centers on a property which he calls DEIXIS (1976:5). Tense is DEICTIC; it “relates the time of the situation referred to some other time,” that is, to a DEICTIC CENTER, usually the time of speaking (Comrie 1976:1–2, my emphasis; cf. references therein, Reichenbach 1947, and Cann 1993:23, chapter 8). Past tense (3) locates situations anterior to the deictic center, as (3) shows. The present tense in (4) indicates contemporaneity of the situation with the center, and the future tense in (5) a situation posterior to it.

(3) I lived in an apartment in Puerto Rico.

(4) I live in a house in Maryland.

(5) I will live overseas someday.

Aspect is NON-DEICTIC: it “is not concerned with relating the time of the situation to any other time-point, but rather with the internal temporal constituency of the one situation” (Comrie 1976:5; cf. Smith 1991:135). It represents how an event takes place over time, without linking it to a specific time. Thus, the Russian sentences in (6) and (7) both refer to the same past event: Tolstoy’s writing of War and Peace. However they differ in terms of “a temporal perspective which focuses all or part of the situation” (Smith 1991:xv). According to Smith (1991), the verb form in (6) presents the writing as IMPERFECTIVE, that is, as (roughly) ongoing in the past. The verb in (7), with a PERFECTIVE prefix, presents the writing as (roughly) complete in the past (examples based on Smith 1991:306, 301).

(6) Lev Tolstoj pisa-l Vojnu i mir.

Lev Tolstoy write-PAST War and Peace

‘Lev Tolstoy was writing War and Peace.’

(7) Lev Tolstoj na-pisa-1 Vojnu i mir.

Lev Tolstoy PFV-write-PAST War and Peace

‘Lev Tolstoy wrote War and Peace.’

Thus, tense and aspect are concerned with time from complementary perspectives. Tense indicates whether situations are located at a reference time, which may be anterior to, contemporaneous with, or posterior to a deictic center. Selection of past, present, or future tense therefore depends on where the reference time of the situation is located with respect to this center. Aspect looks at the internal time of situations, independent of where the situation is temporally located. Selection of imperfective or perfective aspect depends on whether the speaker chooses to present the situation as ongoing or as completed. The distinction between tense as deictic and aspect as non-deictic is adopted by many researchers (Heger 1963 (cited in Gross 1974:25); Anderson 1973:39–40; Thelin 1978:14; Freed 1979; Johnson 1981:152–3; Abusch 1985:1; Chung and Timberlake 1985; Tenny 1987; Smith 1991, Verkuyl 1993). Others equate the distinction with objectivity vs. subjectivity: tense is characterized as objective time and aspect as a subjective presentation thereof (Gross 1974:19; Thurgood 1990:302, fn. 3; Porter 1989:88).

The semantic distinction between aspect and tense must be distinguished from the traditional labels grammarians assign to forms in languages: not all “present tense” forms locate situations contemporaneously with the deictic center. The English “present tense” forms in (8), for example, may be used to describe past situations, as indicated by the adverb yesterday.

(8) Yesterday I get home and find my bird Chomsky loose.

Similarly there is evidence to suggest that Russian “imperfective” forms do not uniformly present situations as imperfective and therefore should not be analyzed as aspect (see Forsyth 1970; Comrie 1976; Gawronska 1993). The diagnostic for distinguishing forms with true aspect and tense semantics from forms generating similar pragmatic implicatures is presented in § 2.3.

The separation of tense and grammatical aspect with respect to the deictic property has two consequences in this study. The first (and intended) result is a simpler and more elegant analysis of aspect, as a view of a situation at a given time, independent of where it is located on a temporal axis. As a further consequence, the representation of tense may be limited to a simple relation between a reference time and the deictic center, often construed as the speech time, independent of how aspect presents a situation. The interpretation of “complex tenses,” such as Reichenbach’s (1947) extended and perfect tenses (also see Comrie 1985), follows from an interaction of simple tense and aspect features (see chapter 4).

1.2 Lexical and Grammatical Aspect

As suggested in the previous section, grammatical aspect focuses on the “internal temporal constituency” of situations. The nature of that constituency has also been treated as aspect in the literature. I label the internal temporal constituency lexical aspect; it has also been known as situation aspect, inherent aspect, Aktionsart (German for ‘type of action’), actionality, aspectual class, verb class, and predicate class. Lexical aspect, together with the imperfective and perfective oppositions known as grammatical aspect, comprise the aspectual phenomena treated in this dissertation.

The distinction between lexical and grammatical aspect is quite widely adopted in the literature (Verwandowsky 1976, quoted in Lys 1988:7).⁴ It may, perhaps, be attributed to Garey’s analysis of French (1957:105, drawing on a suggestion by Sten 1952).⁵ Others, for example in Slavic studies, restrict aspect to the grammatical aspect category. The distinction between lexical and grammatical aspect, or, more specifically, the question of whether lexical aspect should be considered aspect at all is probably the most controversial and important issue in the literature.

As mentioned above, the linguistic concept of aspect was...