Working papers

Wpfg 75 
Reconsidering aspectuality:
interrelations between grammatical and lexical aspect

Pilar Guerrero Medina

University of Córdoba, Spain
Aspectuality in FG
Download this paper in "word" format
Top Page1. Introduction

As pointed out by Comrie (1976: 6 and 11), there is no generally accepted terminology in treatments of aspect. On the one hand, different authors apply different labels to the same concept; on the other hand, the same term is often used to refer to different concepts. In this paper I will try to clarify the terminological confusion concerning aspectual distinctions in the linguistic literature, where the pair of terms "telic"/"atelic", "bounded"/"unbounded" are frequently used as synonyms. More specifically, I will focus on the treatment of lexical aspect in FG, where the telicity parameter is not consistently applied.

In discussions of aspect, it is necessary to make a first distinction between lexically expressed aspectual distinctions ("Aktionsart") and grammatically expressed aspectual distinctions ("Aspect"). The term "Aktionsart" may be thus defined as the lexical equivalent of the grammatical category of "Aspect"1 . Following Dik (1997: 221), the term "aspectuality" will be used to cover both subareas.

Drawing basically upon Dahl (1981), Declerck (1979, 1989) and Depraetere (1995), I will make a further distinction between (a)telicity and (un)boundedness. Telicity will be here considered as an extralinguistic parameter, to a great extent determined by pragmatic factors, while boundedness will be defined as a lexico-grammatical property of the linguistic predication.

Finally, in the second part of this contribution, I will illustrate how aspectuality distinctions manifest themselves in different clause patterns, using specific material from the English language.

    1 Cf. Siewierska (1991: 231; n 2) and Comrie (1976: 6-7; n 2)

Top Page2. Aspectuality in FG

2.1. "Aspect" vs "Aktionsart"
Dik (1997: 106) reserves the term "Aktionsart", regarded as synonymous with "Type of state of affairs", for those distinctions which concern the internal semantics of the predication. Within the framework of FG, the term "state of affairs"(SoA) designates the "conception of something which can be the case in some world" (Dik 1997: 51).2

In what follows I will first discuss the distinctions made in the typology of SoAs in Dik's FG, then focusing on the aspectuality distinctions pertaining to grammatical "Aspect". Table 1 shows the different SoAs distinguished in FG and the major parameters determining them (cf. Dik: 1997: 114).

general term [+control] [-control]
[-dyn] Situation Position State
[+dyn] Event Action Process
[-telic] Event Activity Dynamism
[+telic] Event Accomplishment Change

table 1: Typology of SoAs in FG

The notion of dynamism distinguishes Situations ([-dyn]) from Events ([+dyn)], both of which are then subdivided in terms of the control parameter, resulting in a classification of SoAs into Positions, States, Actions and Processes. Actions and Processes are further subdivided on the basis of telicity, resulting in the following classification: Accomplishment [+tel], Activity [-tel], Change [+tel] and Dynamism [-tel].

As pointed out by Siewierska (1991: 47), there is an evident parallelism between this typology of SoAs and the typology developed by Vendler (1967: 97 ff). Both classifications are compared in (1):
















There is, however, an important difference: Vendler's is a verb classification, while Dik's typology is based on predications. It is interesting to compare these classifications with the one presented by Durst-Andersen and Herslund (1996: 67-68). These authors take their starting point in the perceptual notion of picture, which is regarded as "the mediating link between reality and mind"3. They distinguish between actions and non-actions. The latter are further subdivided into states and activities. While activities and states manifest themselves in reality, actions are constructs with no original in reality. An action manifests itself either as a state situation conceived to be caused by a prior activity, ie as an event, or as an activity situation perceived as intended to cause a future state, ie as a process.

As shown above, the "Aktionsart" subarea, a matter of intrinsic lexical coding, is captured in Dik's typology of SoAs (Dik 1997: 221). And it is precisely the classification of states of affairs which determines the general approach to aspect and tense followed in FG, specifying the nature of the interaction between lexical and grammatical aspect (cf. Siewierska 1991: 44).
In FG the category of grammatical aspect is divided into four subareas, as shown in (2) (cf. Dik 1997: 221):

(2) i. Perfectivity vs Imperfectivity:  
  Perfective: The SoA is presented from an outside point of view
  Imperfective: The SoA is presented from an inside point of view, as being incomplete or in progress.
  ii. Phasal aspect: Ingressive, Progressive, Continuous and Egressive Aspect
  iii. Quantificational aspect: Iterative, Habitual, and Frequentative
  iv. Perspectival aspect: Prospective, Immediate Prospective, Recent Perfect, and Perfect Aspect.

The two first subgroups (ie Perfectivity/Imperfectivity and Phasal Aspects) are called "internal aspects", as they primarily concern the internal dynamics of the SoAs. Quantificational Aspects, which concern different types of quantification over SoAs, and Perspectival Aspects, which concern the relevance of the SoA to an external temporal reference point, are labelled "external aspects" (Dik 1997: 225).4

In what follows, I will focus on the opposition perfective/imperfective, studying the interaction between imperfectivity and telicity (see section 2.2.2). It should be noted that the English Progressive, though expressing one dimension of what may be covered by the Imperfective of other languages, cannot be equated with the Imperfective Aspect as such (cf. Dik 1997: 224). However, as pointed out by Comrie (1976: 7), the opposition progressive vs. non-progressive, grammatically expressed by means of a verbal periphrasis in English, is to a certain extent comparable to the perfective/imperfective distinction, which has not been grammaticalized in this language.5

2.2. Telicity: a controversial parameter
2.2.1. Telicity vs momentaneousness
As shown above, telicity is one of the parameters applied in the FG typology of SoAs. In FG a telic SoA is defined as one that is fully achieved only if it reaches its natural terminal point (cf. Dik 1997: 108). On the other hand, atelic SoAs, which lack an inherent terminal point, are completely realized whenever they obtain. Compare with Comrie's definition of a telic situation:

(3) (...) "a telic situation is one that involves a process that leads up to a well-defined terminal point, beyond which the process cannot continue". (Comrie 1976: 45)

In Comrie's (1976: 47) terms, the linguistic expression John reached the summit is not telic, "since one cannot speak of the process leading up to John's reaching the summit by saying John is reaching the summit." In FG, however, this predication is classified as momentaneous (and therefore telic) (cf Dik 1997: 11). The treatment of momentaneous SoAs (which, by definition, have no internal structure) as telic is, as indicated by Siewierska (1991: 51), a controversial aspect of the FG approach to telicity. The distinction between [+momentaneous] (ie "punctual") and [-momentaneous] (ie "durative") within telic SoAs is highly implausible since a telic SoA always entails duration.

On the other hand, only dynamic SoAs are classified as "telic" or "atelic" in FG. This is another inconsistency in Dik's approach, since non-dynamic SoAs, which lack an inherent terminal point and do not involve change, should be classified as atelic (cf. Siewierska 1991: 54).

2.2.2. (A)Telicity vs (Un)Boundedness
Following Depraetere (1995), I contend that it is necessary to draw a distinction between the notions (a)telicity and (un)boundedness. As stated by the author, telicity relates to the "potential actualization" of a situation, whereas the boundedness parameter measures the "actual realization" of the situation:

(4) (A)telicity has to do with whether or not a situation is described as having an inherent or intended endpoint; (un)boundedness relates to whether or not a situation is described as having reached a temporal boundary. (Depraetere 1995: 2-3)

The telicity parameter is thus based on the attainment of a terminal point, which may be "inherent" or "intended". It seems necessary to clarify this conceptual distinction. In my view, the notion "inherent endpoint" appears to be related to the notion of change, affecting either the first or second argument of the predication. Along these lines, Rijksbaron (1989: 4) states that only telic SoAs (kineseis) involve change in the strict sense. He argues (1989: 33) that the change denoted by telic SoAs may manifest itself in various ways. The second argument may refer to an "effected entity", like a house in build a house, or to an affected entity, like a bicycle in repair a bicycle. Affectedness is also involved in SoAs where the change is undergone by the referent of the first argument. It would be the case, for example, of the referent of the noun filling the first argument position in John recovered. On the other hand, the notion "intended endpoint" is a pragmatic notion connected with the presence of a conscious agent deliberately instigating an SoA (cf. Declerck 1979: 766).

As pointed out by Depraetere (1995: 6-7), Dahl (1981) is one of the linguists who has argued in favour of a distinction between telicity and boundedness. Dahl speaks of the T and P properties. However, this distinction is not 100% equivalent with Depraetere's. According to Dahl, the P property, which could be equated with the boundedness property in Depraetere's approach, entails the T property. If the P property corresponded to the boundedness property and the T property to telicity, all bounded sentences would be classified as telic, and this is not the case, as illustrated in John lived in London for a year, which Depraetere (1995: 5-7) classifies as bounded and atelic.

Telicity could be defined as a pragmatic notion, only applicable to situations, conceived of as non-linguistic entities7. On the other hand, boundedness should be characterized as a property of the clause, or more specifically, of the situation as it is represented in a particular clause. I agree with Declerck (1979: 764) that situations are not inherently bounded or unbounded:

(5) a. John drank whisky [-bounded]
  b. John drank six glasses of whisky [+bounded]

Both (5a) and (5b) can refer to the same situation; however, in the first case the situation is represented as unbounded, while in the second it is represented as bounded. In Declerck's (1979: 782) terms, the NP six glasses of whisky functions as a quantifying complement that measures the situation.

And there are cases of linguistic expressions which are not inherently bounded or unbounded. Declerck (1979: 767) gives the following examples:

(6) a. The endless procession walked by the church.
  b. The insect crawled through the tube.
  c. John filled the tank with water.

The examples in (6) are linguistic representations of situations that are neither inherently bounded nor unbounded. They are therefore compatible with temporal expressions of the form in an hour or for an hour entailing either interpretation.

Comrie (1976: 46) points out that the semantic range of telic verbs is restricted considerably when combined with the imperfective. I would reformulate this statement as follows:

(7 ) A telic SoA may be presented as unbounded, when viewed as non-complete or in progress.

Compare, for instance, (8a) and (8b):

(8) a. I made a chair [+telic] [+bounded]
  b. I was making a chair [+telic] [-bounded]

Both linguistic expressions can be said to describe the same situation8; however, (8a) is telic and bounded, since it entails the attainment of the terminal point: "the chair was completed"; on the other hand, (8b) is telic but unbounded: "the chair was not completed at the time of speech". Telic events are those that possess an inherent terminal point, but they are not necessarily presented as bounded, as shown in (8b). As pointed out by Depraetere (1995: 4), "the (a)telic character of a sentence, unlike (un)boundedness, is not affected by the progressive". The examples in (8) are telic, irrespective of whether a progressive verb form is used or not.

This calls for a revision of the telicity parameter, as applied by Dik in his typology of SoAs. The Activity category, defined by Dik as [+control] and [-telic] includes examples such as John was reading a book, which, in my view, is an unbounded predication codifying a telic SoA9. The linguistic predication John was reading a book is unbounded, because it does not represent the situation as terminating. It does not imply, however, that John did not actually finish the book (cf. Declerck 1979: 767).

    2 An SoA can also be created in a "mental world" of Speaker and Addressee, and then be described by a predication (cf. Dik 1997: 51)
    3 This position seems to be in accordance with the FG approach, where SoAs do not represent real-world situations, but "the codified view of reality built into the grammar of a language" (Siewierska 1991: 43-44).
    4 Comrie (1976: 3-5) defines aspect as a way of viewing the internal temporal constituency of a situation. From this definition, formulated essentially in semantic terms, it is evident that aspect is connected with time. However, as stated by the author, aspect and tense are two distinct categories, concerned with time in very different ways: "Aspect is not concerned with relating the time of the situation to another time-point, but rather with the internal temporal constituency of the one situation; one could state the difference as one between situation-internal time (aspect) and situation-external time (tense)".
    5 Progressiveness, one of the semantic subdivisions of Imperfectivity cf. Dik 1997: 223) is defined by Comrie (1976: 33) as "imperfectivity that is not occasioned by habituality".
    6 As Siewierska (1991: 232; n 12) indicates, telicity is used in FG in accordance to what Dahl (1981: 81) calls the "Western" as opposed to the "Eastern" tradition.
    7 It is important to differentiate between the properties of extralinguistic SoAs and the properties of the linguistic predications which codify these SoAs. However, it is not always easy to classify an SoA as telic or atelic, since the speaker can refer to a particular situation in different ways (cf. Depraetere 1995: 4).
    8 The term "situation" is here used to refer both to dynamic and non-dynamic SoAs (cf. Comrie 1976: 13).
    9 It should be noted that the criteria proposed by Dik (1997: 109-110) to determine the telicity of an SoA rather assess the (un)boundedness of the predication.

Top Page3. Illustration

We have seen how the imperfective progressive verb form may represent a situation as unbounded. But there are other factors which may determine the bounded or unbounded character of the predication. In what follows I will try to illustrate how the bounded/unbounded aspectual distinction manifests itself in different English clause patterns.

3.1. Adverbial modification (Level-1 satellites)
As shown in (9b)10 , boundedness may be determined by a Direction satellite (to the station), since the adverbial imposes temporal boundaries to the situation (Depraetere 1995: 4).

(9) a. John walked in the park. [-bounded]
  b. John walked to the station [+bounded]

In the FG Typology of SoAs, (9a) is classified as an Activity ([-telic]), and (9b) as an Accomplishment ([+telic]). However, as Siewierska (1991: 56) points out, if SoAs are defined on nuclear predications, we would not expect satellites, which do not belong to the nuclear predication, to affect the nature of SoAs.

3.2. Argument terms in transitive constructions
Compare the examples in (10) and (11)11 :

(10) a. John painted a portrait [+bounded]
  b. John painted portraits [-bounded]

(11) a. Demonstrators passed the station [-bounded]
  b. The demonstrators passed the station [+bounded]

In (10a) the nature of the second argument gives a bounded character to the predication. In my view, this bounded predication ("telic" in Dik's terminology) does not represent a situation that may go on indefinitely, as Dik (1997: 108) seems to suggest, since it is obvious that there is no real-world situation which can be continued indefinitely. The pragmatic implication following from this linguistic predication is either that John was in the habit of painting portraits or that John painted an indefinite number of portraits. In both (10a) and (10b), the situation is pragmatically telic, since it has an intended terminal point.

In (11a), it is the first argument of the predication which determines the bounded character of the predication. Since (11b) presupposes a specified quantity of demonstrators, it is clear that when the Action finished, all the demonstrators passed the station. The situation, however, is not inherently bounded or unbounded. As indicated by Depraetere (1995: 10), we should bear in mind that "the changes effected by a particular feature often result in the sentence referring to a different extra-linguistic reality". Although changing a singular NP into a plural NP may turn a clause into a bounded predication, it may also result in the clause representing a different SoA.

3.3. Adjectival predicates in resultative constructions
The examples in (12), (13), (14) and (15)12 show how resultative constructions, which in Goldberg's (1995: 81) terms, involve a metaphorical interpretation of the result phrase as a metaphorical type of Goal , can turn unbounded predications into bounded predications.

(12) a. Fred cried. [-bounded]
  b. Fred cried the handkerchief wet. [+bounded]

(13) a. He cried. [-bounded]
  b. He cried himself asleep. [+bounded]

(14) a. Sam talked for an hour. [-bounded]
  b. Sam talked himself hoarse. [+bounded]

(15) a. Terry pushed the door (for an hour) [-bounded]
  b. He pushed the door shut [+bounded]

Following Goldberg (1995: 180), I assume that the necessary constraint on the appearance of resultatives can be stated in semantic terms: "the resultative can only apply to arguments that potentially undergo a change of state". As shown in the examples above, the combination of the postverbal NP and secondary predicate expresses a state which constitutes the endpoint of the activity (cf. Hoekstra 1988).13

    10 The examples in (9) have been taken from Dik (1997: 109).
    11 The examples in (10) and (11) have been adapted from Dik (1997: 108-109).
    12 The examples in (12) have been taken from Vanden Wyngaerd (1999: 78). The examples in (13), (14) and (15) come from Goldberg (1995: 184, 194 and 196).
    13 According to Hoekstra (1988), the combination NP + adjective forms a "small clause".

Top Page4. Conclusions

In this paper I have dealt with the telic/atelic and bounded/unbounded distinctions, which have had a long tradition in the linguistic literature. I have pointed out the necessity of distinguishing between telicity, viewed as a pragmatic parameter, and boundedness, considered as a lexico-grammatical parameter which may depend on the verb and on the semantic properties of the nominal constituents of the predication.

I have tried to demonstrate that telicity is a vague parameter, being determined, to a great extent, by pragmatic factors. A shown in (3), it is only in a given clause pattern that grammatically aspectual distinctions manifest themselves, originating the bounded/unbounded distinction.

The issue of aspectuality distinctions undoubtedly deserves a more detailed study in a cross-linguistic perspective. I hope, however, that this contribution has helped to cast a new light on one of the most confusing areas of the linguistic literature.

COMRIE, BERNARD (1976) Aspect. Cambridge: Cambridge University Press.
DAHL, ÖSTEN (1981) "On the definition of the Telic-Atelic (Bounded-Nonbounded) distinction". In Philip Tedeschi and Annie Zaenen (eds), 79-90.
DECLERCK, RENAAT (1979) "Aspect and the bounded/unbounded (telic/atelic) distinction". Linguistics 17, 761-794.
DECLERCK, RENAAT (1989) "Boundedness and the Structure of Situations", Leuvense Bijdragen 78, 275-308.
DEPRAETERE, ILSE (1995) "On the necessity of distinguishing between (un)boundedness and (a)telicity". Linguistics and Philosophy 18, 1-19.
DIK, S.C. (1997) The Theory of Functional Grammar. Part 1: The Structure of the Clause. (ed. K. Hengeveld). Berlin/New York: Mouton de Gruyter.
DURST-ANDERSEN, PER & MICHAEL HERSLUND (1996) "The syntax of Danish verbs. Lexical and syntactic transitivity". In Elisabeth Engberg-Pedersen et al. (eds), 65-102.
ENGBERG-PEDERSEN, ELISABETH et al. (eds) (1996) Content, Expression and Structure. Studies in Danish Functional Grammar. Studies in Language Companion Series. Amsterdam: John Benjamins.
GOLDBERG, ADELE E. (1995) Constructions. A Construction Grammar Approach to Argument Structure. Chicago/London: The University of Chicago Press.
HOEKSTRA, TEUN (1988) "Small clause results". Lingua 74, 101-139.
SIEWIERSKA, ANNA (1991) Functional Grammar. London: Routledge.
TAYLOR, BARRY (1977) "Tense and continuity". Linguistics and Philosophy 1: 199-220.
TEDESCHI, PHILIP & ANNIE ZAENEN (eds) (1981) Syntax and Semantics. Tense and Aspect. New York: Academic Press.
VANDEN WYNGAERD, GUIDO (1999) "Aspects of (Un)boundedness". In Svetlana Vogeleer et al. (eds), 77-102.
VOGELEER, SVETLANA et al. (1999) Belgian Journal of Linguistics. Tense and Aspect. Amsterdam: John Benjamins.

Print PageTop Page