Printer Friendly

Used to and habitual aspect in English.

Habituality has been much discussed in connection with the semantics of genericity and of quantification. As pointed out by Sali Tagliamonte and Helen Lawrence and by Bengt Altenberg, however, there has been little investigation of such supposed markers of habitual aspect in English as used to (1) and would (2):

(1) Susan used to swim every day.

(2) Susan would swim every day.

In this essay I argue that used to is not, in fact, a marker of past habituality.

Following Bernard Comrie's pioneering work in 1976, the habitual has generally been viewed as a species of imperfective aspect, alongside continuous aspect, represented in English by the progressive tenses. It is certainly true that used to and would translate the imperfective aspect of, say, French (3, 4), when the progressive is inappropriate, as does the simple past tense (5):

(3) Quand j'etais petit, nous allions a la plage chaque semaine. 'When I was young, we used to go to the beach every week.' <>

(4) When I was young, we would go to the beach every week.

(5) When I was young, we went to the beach every week.

The progressive is used when a situation is viewed with regard to its internal temporal constituency and without regard to any temporal bounds, either initial or terminal. (1) Hence it represents an event or process in its course, with no implication as to its completion. The habitual is similar, with the difference that instead of a single event, it concerns a series of recurring events (as in 2), or of bounded states (6).

(2) Susan would swim every day.

(6) I used to be awake all night thinking, "Oh God, Where am I going to go? What's going to become of me?" < 2003/01/01/national/printable534932.shtml>

Under this conception of the habitual, there is clearly something wrong with characterizing used to as a marker of habituality, since, as Comrie observed (28) and unlike would, used to can be used with stative expressions referring to a single, extended state, as in example 7. Example 8 is an acceptable sentence, to be sure, but it refers to a series of bounded states and does not refer, as 7 does, to a single, continuous state. Veerle Van Geenhoven (sec. 3.2), comments that since 7 does not refer to a habit, it is "misleading or wrong to think of the periphrastic form used to as an habitual aspect marker." (I argue here that it is wrong.)

(7) The temple of Diana used to stand at Ephesus.

(8) The temple of Diana would stand at Ephesus (from time to time).

There is something odd about 8. A sentence like this normally concerns a series of events or bounded states (episodes) involving its subject. Thus 9 reports a series of occasions on which Max stood on the corner:

(9) Max would stand on the corner (from time to time).

Where the individual members of the series are states, the verb or adjective is typically a stage-level predicate (Carlson), denoting a relatively temporary, changeable property. A relatively permanent, stable individual-level predicate is unusual:

(10) ?Max would be tall (from time to time).

What renders 8 odd is that mobility is not normally a property of temples; the most normal interpretation of 8 is therefore not that the same temple returned to Ephesus several times but that different temples stood there on various occasions. Admittedly, 8 is not a very normal sentence, even with this interpretation. But 7 is perfectly normal. Somehow, the semantics of used to differs from that of habitual would.

Originally used to was, like habitual would, just the past tense of a verb (use), which was not restricted to the past tense, and could be used also in at least the present (11) anal future (12) tenses, and also as an infinitive (13) (Tagliamonte and Lawrence, citing Bybee, Perkins, and Pagliuca 155-56). (2)

(11) I vse not to kisse men. (Nicholas Udall, Ralph Roister Doister, ca. 1550)

(12) I will use to visit him after dinner, for he dines too late for my head. (Jonathan Swift, Journal to Stella, 1766-68)

(13) you must use to write before he com's. (The Letters of Dorothy Osborne to William Temple, written 1652-54)

Erik Jorgensen shows that even today this verb is not limited to the past tense in British usage, at least, where a pluperfect, had used to (14), is quite acceptable:

(14) Tom had used to pretend ... and he had used to tell wildly funny stories. ... (Patricia Highsmith, The Talented Mr. Ripley, 1968)

An often-cited and important difference between used to and would is that used to implies that the situation represented is no longer the case. Thus, sentence 1 implies that at the present time, Susan no longer swims, while 2 lacks any such implication:

(1) Susan used to swim every day.

(2) Susan would swim every day.

Although this implication is often cited as all or part of the meaning of used to, it is not in fact part of its meaning, for as Rick Harrison points out, this implication is defeasible: it can be canceled by a tag like and he still is, as in Harrison's example (15). Hence it is a conventional implicature: an implication dependent on context:

(15) Erik used to be a member of the Volapuk League, and he still is.

The question is why used to has this implicature and the other putative expressions of habituality do not. Compare, for example, the unexceptionable 16 with the odd 17:

(16) Susan used to swim every day, and she still does.

(17) ?Susan would swim every day, and she still does.

It is ironic that used to is acceptable in sentences like 16, while would is not, given that used to has traditionally been supposed to "mean" that the situation in question no longer obtains.

This ironic, and key, distinction between used to and habitual would is not unrelated to the fact that used to no longer has a present tense counterpart, though both the (habitual) simple past tense and habitual would do (18, 19), and to the fact that it, unlike would, at least, has a counterpart in the perfect aspect.

(18) Susan eats lunch in the school cafeteria, never outside of school.

(19) Susan will eat salami from time to time, but never liverwurst.

There is considerable evidence that used to is not, in fact, a past tense expression, as far as its semantics goes; it is a present tense expression, and the pluperfect is really its past tense, as habitual would is the past tense of habitual will, and the habitual simple past is the counterpart of the habitual present tense. On the face of it, that is a strange claim to make, for used to definitely is used to express past situations that no longer occur at the present time.

The key to the mystery of used to lies in the pioneering research of Tagliamonte and Lawrence into the factors playing a role in the choice of expressions of habitual aspect in English. They report large differences in the usage patterns of the supposed markers of habitual aspect, despite their supposed similarity in meaning, though Tagliamonte and Lawrence do not explain these differences. It is often impossible to replace one of these markers with one of the others. We have seen above that used to in Comrie's example (7) cannot really be replaced by would (8), though the simple past is fully acceptable (20):

(7) The temple of Diana used to stand at Ephesus.

(8) The temple of Diana would stand at Ephesus (from time to time).

(20) The temple of Diana stood at Ephesus.

The same is true of Harrison's example (15), which can be recast as 21. On the other hand, would in 22a cannot readily be replaced by used to (22b), though the simple past is perfectly all right (22c). But would in 23 cannot be replaced by either used to (24a) or the simple past tense (24b) without changes in meaning (which is what the "!" means). These facts can be accounted for only by differences in the meaning and/or use of used to, would, and the simple past tense. As I have already suggested, these differences are fundamental and considerable.

(15) Erik used to be a member of the Volapuk League, and he still is.

(21) Erik was a member of the Volapuk League, and he still is.

(22) a. It was his custom of a Sunday, when this meal was over, to sit close by the fire, a volume of some dry divinity on his reading-desk, until the clock of the neighbouring church rang out the hour of twelve, when be would go soberly and gratefully to bed. (R. L. Stevenson, Dr. Jekyll and Mr. Hyde) (3)

(22) b. ? ... until the clock of the neighbouring church rang out the hour of twelve, when he used to go soberly and gratefully to bed.

(22) c. ... until the clock of the neighbouring church rang out the hour of twelve, when he went soberly and gratefully to bed.

(23) When a man got to the mast-head, he would come slowly down again to get something which he had forgotten.... (Richard Henry Dana, Two Years before the Mast)

(24) a. !When a man got to the mast-head, be used to come slowly down again to get something which be had forgotten....

(24) b. !When a man got to the mast-head, be came slowly down again to get something which be had forgotten....

The distributional properties of used to revealed by Tagliamonte and Lawrence are precisely those of the present perfect tense. For example, they show (334-35,342) that used to far more readily occurs in absolute position, and much more often in the opening sentence of a discourse (as in 25, the beginning of a Web page), than its rivals:

(25) I used to torture cats. Now I'm a vegetarian. <>

An account like 25, if it were to begin "I would torture cats. Now I'm a vegetarian," would seem incomplete and removed from context, as would one in the past tense ("I tortured cats"). The present perfect, too, readily occurs in absolute or initial position. There is something oddly "incomplete" about the past tense out of context (as in 26), because it is essentially an anaphoric tense, requiring some time, even the vague "once upon a time," to anchor it within the discourse. But the present perfect is deictic; it links situations directly to the present and so never requires such an antecedent time; compare 26 to 27:

(26) I went home.

(27) I've gone home.

Another property of used to is that it allows no definite, past-tense temporal adverbials (28), although indefinite ones co-occur with it freely (29): (3)

(28) *I used to live in York in 1914. (4)

(29) I used to live in York sometimes.

(30) In 1914, I used to live in York.

Tagliamonte and Lawrence cite example 30, but it does not show that definite adverbials are possible with the habitual used to. In 30, in 1914 is a frame-setting, sentence adverbial outside the scope of used to. In 28, however, it is a verb-phrase modifier, an adjunct of the verb. Example 28 concerns living-in-York-in-1914, something only a time-traveler could be in the habit of doing, whereas 30 merely says that living-in-York took place in 1914, and that is something that could be true of an ordinary subject.

The English present perfect--at least the resultative perfect, which reports something that has recently happened--is notorious for not allowing definite past-tense temporal adverbials:

(31) I've lived in York in 1914.

Example 31 does have a marginal use as an experiential or existential perfect, which reports whether something has ever happened; 31 could, for example, form part of a recitation of past accomplishments (I've been at the Sphinx at midnight, I've been in Times Square at noon, I've lived in York in 1914), but it cannot report simply an event with present consequences (I've just lived in York in 1914 and now I'm done), unless, again, the speaker is a time-traveler.

Yet a third property of used to is that it is uncommon with negation (Tagliamonte and Lawrence 337ff). While avoidance of used to may well be partly due, as David Denison suggests (323), to speakers being uncomfortable with both usedn't (usen't) and didn't use--as Jorgensen amply documents--there is a semantic reason for the avoidance of negation with used to. The present perfect is also not readily negated, at least in its continuous and resultative readings, though it is more readily negated in its existential use. This is precisely because the continuous reading concerns something true up to the present, while the resultative concerns a present state of affairs that is the consequence of a past event. But a negative reports the nonoccurrence of some contextually presupposed eventuality and hence is generally less salient to the current state of affairs than the occurrence of some eventuality. In the same way, didn't use to or usedn't to presupposes a present state of affairs that it says did not occur in the past. Given that the point of used to is precisely to contrast what happened in the past with what no longer is the case, it is easy to see why it might not readily undergo negation.

Tagliamonte and Lawrence find a marked preference for used to over its rivals with animate and in particular with first-person subjects (334ff). It is because both the present perfect and used to are deictic in that they are associated with animate and especially first-person subjects. Highly salient categories such as animate and human (Wallace 212f) have been associated with the "present-immediate," as opposed to the "nonpresent-remote," and writers who find cognitive bases for the deictic/nondeictic distinction (e.g., Janssen) stress the difference between that which is within the immediate sphere of the speaker (what is focal) and that which is not (the distal). It is understandable that human speakers would find the animate and the human more salient than the inanimate and nonhuman and that tenses that tend to link the temporal location of the speaker with that of the subjects of their sentences, as do deictic tenses like the present perfect and the used to construction, should tend more often to be used with a first-person subject than tenses used in recounting events remote in time and/or space from the speaker.

Finally, though Tagliamonte and Lawerence say that used to is "highly restricted to nonstative verbs" (339), we have seen that it does allow stative predicates, as in 7. But stative expressions are definitely not as common with used to as nonstative ones, in contrast with the simple tenses, which allow statives, both stage-level (32) and individual-level (33), quite as readily as nonstatives, and in contrast with will, which does not allow statives at ali, unless we consider expressions for bounded states (as in 34) to be stative, as opposed to eventive. (5)

(7) The temple of Diana used to stand at Ephesus.

(32) Every time he looked, she was deep in conversation with someone. <>

(33) Pompeii was a resort.

(34) From time to time she would be back in Toronto for treatment and we would have the opportunity to meet with her. <http://>

While the present perfect does allow stative expressions, it restricts them to nonresultative uses. Unbounded states (35) are interpreted as the continuous perfect, something still true of the present; bounded states are interpreted as either the continuous (36) or the existential perfect (37), but neither is normally interpretable as the resultative perfect:

(35)--Why have you been away from the office these last two weeks?

--I have been ill.

(36)--Why do you look so tired?

--I'm exhausted. I've gone without sleep for a whole night.

(37)--You don't know what it is to be an insomniac.

--Sure I do. I've gone without sleep for a whole night.

The reason for this is that the continuous perfect represents a continuing and hence unbounded state of affairs, while the existential allows a bounded state of affairs, something which happened wholly in the past. The resultative perfect concerns a present state of affairs resulting from something that happened wholly in the past. Unbounded states lack the necessary terminal bound, and bounded states normally lack the consequence relationship between their termination and the present state of affairs. In other words, in an example like 38, the present situation, that the vase is broken, is a direct consequence of the breaking of the vase.

(38) The vase has broken.

While the termination of a state sometimes functions like the inherent termination of an event, it generally does not.

For the same reason, used to does not readily occur with states in its "habitual" reading. States are inherently stable and are not as susceptible to change as actions or processes, but the whole point of used to is that there is a change from one state of affairs to another. Moreover, the present state of affairs must contrast with the former state of affairs. It may involve a simple antithesis, as in 39, but it need not, and it is possible to say something like 40. But not just any contrast is allowed, as shown by the rather peculiar 41.

(39) Newfoundland used to be an independent polity, but now it isn't.

(40) Newfoundland used to be an independent polity, but now it's part of Canada.

(41) ?Newfoundland used to be an independent polity, but now it's a big island.

(42) Newfoundland used to be an independent polity, but now it's just a big island.

Sentence 40 is acceptable because Newfoundland's lack of independence is directly relatable to the fact that it's now part of Canada. What has being a big island to do with independence? It is not enough, however, to say that no longer being independent is the same thing as being part of Canada, first because of course it isn't (what about Texas?) but also because the properties in question must not merely happen to coexist but be consequentially related. Compare the perfectly normal 42 to the strange 41; being an independent polity "means" not just being a big island. If the present perfect and the used to construction both have problems with stative expressions, it's because states normally do not, in and of themselves, give rise to consequent states of affairs.

That the present perfect and used to share so many properties is no accident. Both make implicit reference to the present time, and as we have already observed, unlike an anaphoric tense such as the past, neither requires a reference time to anchor to, since they already are anchored by the present. For example, at the beginning of Lawrence Durrell's novel The Dark Labyrinth, the past tense in the initial sentence presupposes a fictitious point in time, which serves to anchor subsequent anaphoric tenses in the text:

(43) During the early part of June, 1947, a small party of sight-seers found itself trapped in what was then the newly discovered labyrinth of Cefalu, in the island of Crete. They had penetrated the network of caves and corridors with a guide from a tourist agency....

But neither used to nor the present perfect works that way. What they do instead is to relate the event or state recounted directly to the present time, and subsequent statements cannot anchor anaphorically on them. The past tense was in the second sentence of 44 does not mean that the book was by Jane Austen at the time that I read it, nor does were in 45 refer to the time at which I read the novels:

(44) I have just read a great book. It was by Jane Austen.

(45) I used to read novels by Dostoyevsky. They were surprisingly funny.

Contrast the first two sentences of Durrell's novel (43). Here the reference point for had penetrated in the second sentence is precisely the past time at which the fictional state of affairs recounted in the first sentence (finding themselves trapped in the labyrinth) occurs.

Emile Benveniste and Harald Weinrich distinguish two classes of tenses. The present perfect and the used to construction stand alongside the present and future tenses as what Benveniste calls tenses of "discourse" and Weinrich calls tenses belonging to the "world of commentary." These are all deictic tenses, tenses that occur most naturally in everyday discourse and in nonnarrative gentes such as description and commentary. They serve to identify the world and the time of the narrator or speaker with the world and the temporal continuum of the discourse or text, whereas tenses such as the past, which belongs to what Benveniste calls tenses of histoire (story or history) and Weinrich calls tenses belonging to the "world of narration," create an encapsulated world with its own time-line, distinct from the real context, and the real time, of the narrator.

Because both the present perfect and used to are deictic, neither relates the time of a situation directly to any time given in the discourse. For this reason, used to never forms part of the foreground, the principal line, of narrative, and it never advances narrative time. It cannot form even subsidiary narrative lines: sentence 46 does not present a temporal sequence but rather a series of events independent of one another.

(46) He used to get up, he used to shower, be used to get dressed, and (*then) be used to eat breakfast.

But the present perfect does not form sequences either:

(47) He has gotten up, be has showered, be has gotten dressed, and (* then) be has eaten breakfast.

Consequently used to, like the present perfect, is at home in genres such as reportage that do not narrate but instead relate facts directly to the time of utterance, the deictic center. Interestingly enough, both the present perfect and used to appear to form narrative sequences when the auxiliary verb is not repeated, as in 48 and 49 respectively:

(48) He used to get up, shower, get dressed, and then eat breakfast.

(49) He has gotten up, showered, gotten dressed, and then eaten breakfast.

In both cases, the entire sequence is within the scope of the auxiliary. The sense of 48 is 'it used to be the case that ...' and that of 49, 'it has been the case that ...,' in both instances the "case" being simply that the sequence expressed by the phrase following the auxiliary verb occurred.

It should not be surprising, given all these similarities, that used to shares with the present perfect its connotation of "present relevance." The difference in acceptability between the sentences in 50 has to do with the difficulty of seeing present relevance in a figure from the relatively distant past. The same is true of the used to sentences in 51.

(50) a. Bill Clinton has been President of the U.S.

b. Abe Lincoln has been President of the U.S.

(51) a. Bill Clinton used to be President of the U.S.

b. Abe Lincoln used to be President of the U.S.

It has been observed that the present perfect presupposes that an event be repeatable (McCawley; Leech 32ff). Thus 52 is odd unless the Gauguin exhibit is still continuing. Compare with 53, which presupposes that the 1939 Fair is somehow available for visiting.

(52) Have you visited the Gauguin exhibition? (Leech 33)

(53) Have you visited the 1939 World's Fair?

Used to likewise discriminates between the repeatable and the nonrepeatable. What renders 54a so odd is that directing Annie Hall is the kind of thing you can do only once; a series of films (54b) or a play, with its repeated performances (54c), is fine; 54d implies that people are still making the Godfather movies, even if Coppola is not:

(54) a. ?Woody Allen used to direct Annie Hall.

b. Woody Allen used to direct critically acclaimed films.

c. He used to direct Hamlet at the Stratford Festival.

d. Francis Coppola used to direct the Godfather movies.

So there is considerable evidence that the used to construction is akin to the present perfect tense. Like the present perfect, it is an essentially present tense that nonetheless concerns past events and states. It is not surprising then that, just as the pluperfect is the past counterpart of the present perfect, it should serve a similar function where used to is concerned. But the present perfect is not the same thing as used to. What then is the used to construction?

It is an anti-present-perfect. The present perfect situates an event or state that began in the past, and may also have ended in the past, on an interval of time called the extended now, which includes the present. The perfect can be used in 55, for example, because the second speaker's mother has not yet returned from the store; if she had, the event of her going would be separated from the present by this event of returning, which would thereby serve to undo the present consequences of her going to the store, namely that she is still away.

(55)--Is your mother in?

--No, she's gone to the store.

The present perfect thus includes the present in what is essentially a period of the past. The used to construction, on the other hand, precisely excludes the present from a past period. This present period is pragmatically determined, and in precisely the same way as the "extended now" of the perfect. That is, the era presupposed by the used to construction immediately precedes the era--a sort of extended now--containing the present. Both the extended now of the perfect and the current era presupposed by the used to construction begin immediately after the termination of a past situation, an event in the case of the perfect, a state of affairs or a series of events in the case of used to.

The present era is indeed one which contrasts with the preceding era expressed by used to. But the implicature that the situation no longer holds in the present is cancelable. How could this be if used to presupposes that the situation no longer holds at present but holds in the era immediately preceding? The answer is that while normally the present era contrasts with the immediately preceding era precisely in regard to the stated situation (that is, the situation in question defines the preceding era and hence its contrast with the present), in those cases in which the implicature is cancelable, the era is not defined by the stated situation itself but by something else. Thus, in 15, the difference between the past and present eras cannot be Erik's being or not being a member of the Volapuk League, since in both eras he is a member of it.

(15) Erik used to be a member of the Volapuk League, and he still is.

Rather, something else distinguishes the two periods. Erik's membership in the Volapuk League is presented as characteristic of the previous era, but because this normally brings with it the implicature that the present era contrasts with the previous one in this regard, a tag is used to cancel that implicature, since he is also a member in the present era.

In general, the used to construction presupposes a past era, an extended interval of time in which some state of affairs holds, and the defining characteristic of the era in question may implicitly be the situation presented in the sentence, as in 51a:

(51) a. Bill Clinton used to be President of the U.S.

Where it is not, the defining characteristic is often rendered explicit, as in examples 56-59. (I have italicized the expression defining the era in question in each example.) Notice how now may be used, as in 56 and 57, to explicitly contrast the present era with that past era.

(56) Well, when I was working I used to love not going to work. Now I love having all the kids around. <>

(57) We used to barbeque a lot on holidays, when I was a kid and when I was married. Now it's hard to get everyone in one place at one time, so we catch as catch can. In recent years it has become a tradition to go to the Taste of Colorado. (Ibid.)

(58) When I was a child I used to believe that when children's teeth fell out they had to put it under the pillow and "the little mouse" would take it and leave a coin instead. <>

(59) When at Cambridge I used to practise throwing up my gun to my shoulder before a looking-glass to see that I threw it up straight. (The Autobiography of Charles Darwin)

In these examples, it is presupposed that these eras ended before the present. The writer of 56 is no longer working, that of 58 is no longer a child, and Darwin (59) was presumably writing when he was no longer at Cambridge. This change in the underlying situation defines the eras, and so Darwin could well have canceled the implicature that he no longer practiced throwing up his rifle, if he had so chosen. In examples such as 60-62, where the era is implicit, and moreover, defined by the situation stated in the sentence, no such cancellation is possible.

(25) I used to torture cats. Now I'm a vegetarian. <>

(41) ?Newfoundland used to be an independent polity, but now it's a big island.

(60) But such criticism is now less frequent than it used to be. (Nathaniel W. Stephenson, Abraham Lincoln and the Union: A Chronicle of the Embattled North)

(61) "That's just what they used to say," he said again. (H. Rider Haggard, Allan Quatermain)

(62) I cannot see that gunpowder, telegraphs, steam, daily newspapers, universal suffrage, etc., etc., have made mankind one whit the happier than they used to be, and I am certain that they have brought many evils in their train. (H. Rider Haggard, Allan Quatermain)

(63) ?I used to treat cats well. Now I'm a vegetarian.

The past era in example 60 is that in which criticism was more frequent than at present; in 61, that in which they used to say whatever it is they said; and 62 that in which gunpowder and these other "improvements" had not yet been invented or discovered. Example 25 is fine, but 63 is odd in the way that 41 is, since it is hard to discern the contrast between treating cats well and being a vegetarian.

The whole point of the used to construction is not to report a habit in the past but rather to contrast a past era with the present. The requirement for contrast always trumps the implicature that the situation no longer holds, which is why 64 is not as odd as 65:

(64) When George Bush pere was President, the U.S. used to have trouble with the Iraqis.

(65) ?The U.S. used to have trouble with the Iraqis.

(The eras in 64 contrast not in whether the U.S. has trouble with the Iraqis but rather in whether George Bush pere is President.) In short, used to is not a marker of past tense. It's a tense of "discourse" and, like the present perfect, essentially a present tense. Nor is it a marker of habituality. Like the present perfect, it is about a state of affairs, not a series of occurrences.

Works Cited

Altenberg, Bengt. "Expressing Past Habit in English and Swedish." Grammatik i Fokus. Lund. Feb. 2003.

Benveniste, Emile. "Les relations de temps dans le verbo francais." Bulletin de la societe de linguistique de Paris 54 (1959): 69-82.

Bybee, Joan L., Revere D. Perkins, and William Pagliuca. The Evolution of Grammar: Tense, Aspect, and Modality in the Languages of the World. Chicago: U of Chicago P, 1994.

Carlson, Gregory. "Reference to Kinds in English." Diss. U of Massachusetts, Amherst, 1977.

Comrie, Bernard. Aspect: An Introduction to the Study of Verbal Aspect and Related Problems. Cambridge: Cambridge UP, 1976.

Denison, David. English Historical Syntax: Verbal Constructions. London: Longman, 1993.

Harrison, Rick. "Verb Aspect." 2002. <>.

Janssen, Theo A. J. M. "Deixis from a Cognitive Point of View." Meaning as Explanation: Advances in Linguistic Sign Theory. Ed. Ellen Contini-Morava and Barbara Sussman Goldberg. Berlin: Mouton, 1995. 245-70.

Jespersen, Otto. A Modern English Grammar on Historical Principles. Part 4: Syntax. Vol. 3. London: Allen, 1931.

Jorgensen, Erik. "Used To (+ Infinitive)." English Studies 69 (1988): 348-54. Leech, Geoffrey N. Meaning and the English Verb. London: Longman, 1971.

McCawley, James. "Tense and Time Reference in English." Studies in Linguistic Semantics. Ed. Charles Fillmore and D. Terence Langendoen. New York: Holt, 1971. 97-113.

Smith, Carlota S. The Parameter of Aspect. Dordrecht: Kluwer, 1991.

Tagliamonte, Sali, and Helen Lawrence. "I Used to Dance, but I Don't Dance Now: The Habitual Past in English." Journal of English Linguistics 28 (2000): 32453.

Van Geenhoven, Veerle. "Aspect, Pluractionality, and Adverbial Quantification." Perspectives on Aspect. Utrecht. Dec. 2001.

Wallace, Stephen. "Figure and Ground: The Interrelationships of Linguistic Categories." Tense-Aspect: Between Semantics and Pragmatics. Ed. Paul J. Hopper. Philadelphia: Benjamins, 1982.

Weinrich, Harald. Tempus: Besprochene und Erzahlte Welt. 1964. 2nd ed. Stuttgart: Kohlhammer, 1971.


(1) The first part of this characterization is from Comrie (3), and the second is from Smith, who contrasts aspects in terms of boundedness (120).

(2) Literary examples cited without page numbers are from Gutenberg e-text versions.

(3) Examples 28-31 are from Jespersen (14).

(4) Actually, this sentence would probably strike many speakers as acceptable when uttered by a time-traveler, but other speakers would simply find it unacceptable.

(5) An eventive predicate is one inherently referring to an event, such as run across the street, while a stative predicate is one which inherently refers to a state, such as tired. That does not mean that an eventive sentence cannot be built on a stative predicate, nor a stative sentence on an eventive predicate. For example, Sue was asleep for an hour represents an event, though asleep is a stative predicate. Events differ from states in containing a transition from one state to another; the termination of a state necessarily involves a change of state, so an expression for a bounded state such as asleep for an hour generally renders the sentence as a whole eventive.

Robert I. Binnick

University of Toronto
COPYRIGHT 2006 Northern Illinois University
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2006 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Binnick, Robert I.
Geographic Code:1CANA
Date:Mar 22, 2006
Previous Article:On the conceptual-procedural distinction.
Next Article:Disjunctive numerals of estimation.

Related Articles
Court's hands are tied.
Hands of English courts are tied.
Understanding Sharks.

Terms of use | Privacy policy | Copyright © 2018 Farlex, Inc. | Feedback | For webmasters