A COMPUTATIONAL MODEL OF THE SEMANTICS OF TENSE AND ASPECT 
Rebecca J. Passonneau 
Paoli Research Center, Defense Sysltems, UNISYS ~ 
The PUNDIT natural-language system processes references to situations and the intervals over which 
they hold using an algorithm that integrates the analysis of tense and aspect. For each tensed clause, 
PUNDIT processes the main verb and its grammatical categories of tense, perfect, and progressive in 
order to extract three complementary pieces of temporal information. The first is whether a situation has 
actual time associated with it. Secondly, for each situation that is presumed to take place in actual time, 
PUNDIT represents its temporal structure as one of three situation types: a state, process, or transition 
event. The temporal structures of each of these situation types consist of one or more intervals. The 
intervals are characterized by two features: kinesis, which pertains to their internal structure, and 
boundedness, which constrains the manner in which they get located in time. Thirdly, the computation 
of temporal location exploits the three temporal indices proposed in Reichenbach 1947: event time, speech 
time, and reference time. Here, however, event time is formulated as a single component of the full 
temporal structure of a situation in order to provide an integrated treatment of tense and aspect. 
1 INTRODUCTION 
The PUNDIT text-processing system extracts temporal 
information about real-world situations from short mes- 
sage texts. 2 This involves three complementary analy- 
ses. First, PUNDIT determines whether a situation has 
actual time associated with it. A reference to a possible 
or potential situation, for example, would need a differ- 
ent treatment. Second, it determines the temporal struc- 
ture of the predicated situation, or the manner in which 
it evolves through time. Finally, it analyzes the tempo- 
ral location of the actual situations with respect to the 
time of text production or to the times of other situa- 
tions. These three pieces of information are derived 
from the lexical head of a predication (verbal, adjecti- 
val, or nominal), its grammatical inflections (tense, 
progressive, perfect), and finally, temporal adverbs 
such as before, after, and when. Each of these compo- 
nents of temporal meaning is assigned a context-depen- 
dent compositional semantics. A fundamental premise 
of this approach is that the several sentence elements 
contributing temporal information can and should be 
analyzed in tandem (Mourelatos 1981, Dowty 1986) in 
order to determine the times for which predications are 
asserted to hold. This is accomplished by means of a 
model of the semantics of time that incorporates both 
aspect and a Reichenbachian treatment of tense (Rei- 
chenbach 1947). 
The temporal analysis component described here 
was originally designed to handle PUNDIT's first text 
domain, CASREP messages, which are reports describ- 
ing equipment failures on navy ships. 3 This domain was 
a particularly appropriate one for implementing a com- 
ponent to analyze the time information contained ex- 
plicitly within the individual sentences of a text. CAS- 
REPs are diagnostic reports consisting of simple 
declarative sentences. They present a cumulative de- 
scription of the current status of a particular piece of 
equipment rather than narrating a sequence of events. 
Within one sentence, several different situations may be 
mentioned, linked together by explicit temporal connec- 
tives such as before and after. It is thus possible to 
extract a good deal of the important temporal informa- 
tion from these texts without handling intersentential 
temporal relations. However, the implementation of the 
temporal semantic component described here lays the 
necessary groundwork for eventually computing inter- 
sentential relations along lines proposed in Webber 1987 
and this volume. 4 The capacity to process intersenten- 
tial temporal relations is, of course, essential for ade- 
quately handling narrative data. 
2 TEMPORAL INFORMATION 
The premise of the present work is that accurate com- 
putation of the temporal semantics of the verb and its 
grammatical categories of tense, perfect, and progres- 
sive provide a foundation for computing other kinds of 
Copyright 1988 by the Association for Computational Linguistics. Permission to copy without fee all or part of this material is granted provided 
that the copies are not made for direct commercial advantage and the CL reference and this copyright notice are included on the first page. To 
copy otherwise, or to republish, requires a fee and/or specific permission. 
0362-613X/88/010044-60503.00 
44 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
temporal information, including the interpretion of tem- 
poral adverbials, s However, the task of modeling the 
semantic contribution of the verb and its categories is a 
complex one because temporal information is distrib- 
uted across several nonunivocal lexical and grammati- 
cal elements. As the extensive linguistic and philosoph- 
ical literature on tense and aspect demonstrates, the 
precise temporal contribution of any one surface cate- 
gory of the verb is contingent upon co-occurring verbal 
categories, as well as upon the inherent meaning of the 
verb, and even the nature of the verb's arguments 
(Comrie 1976, Dowty 1979, Mourelatos 1981, Vlach 
1981, Vendler 1967). Hence, even a preliminary solution 
to the computational problems of interpreting temporal 
information in natural language requires recognizing the 
relevant semantic interdependencies. This paper pro- 
poses a solution to the computational task of extracting 
temporal information from simple declarative sentences 
based on separating temporal analysis into distinct 
tasks, each of which has access to a selected portion of 
the temporal input. The ultimate goal is to represent 
temporal information as explicitly as possible at each 
stage of analysis in order to provide the appropriate 
information for the next stage. Because the representa- 
tions are constructed incrementally, it is important that 
they should be explicit about what has been derived so 
far, yet sufficiently noncommittal to avoid conflicting 
with subsequent processing. 
The present section of the paper provides the back- 
ground needed for understanding the information that 
the algorithm integrating tense and aspect (presented in 
Section 4) is designed to compute. First, in Section 2.1, 
I explain what is meant by actual time and delimit the 
scope of the phenomena focused on here. Then in 
Section 2.2, I describe the components of temporal 
structure and how they are used to distinguish states, 
transition events, and two ways of referring to pro- 
cesses. Also in this section I review Dowty's (1979) 
aspect calculus and introduce how it is used in deriving 
the representation of temporal structure. 
The remaining sections of the paper focus on the 
implementation. Section 3 describes the input to the 
temporal component. Section 4 presents the algorithm 
for computing the situation representations and their 
temporal location. Part of the computation of temporal 
location involves determining the reference time of a 
predication. Reference time pertains to the interpreta- 
tion of relational temporal adverbials, i.e., adverbials 
that relate the time of a situation to another time (e.g., 
The ship was refueled yesterday, cf. Smith 1981). 6 
Temporal connectives, for example, relate the time of a 
syntactically subordinate predication to a superordinate 
one. A brief discussion of how the reference time 
participates in the interpretation of temporal adverbial 
clauses introduced by connectives such as before, after, 
and when is given in Section 5, whose more general 
topic is the utility of the situation representations for the 
interpretation of a variety of adverbial types. 7 
2.1 ACTUAL TEMPORAL REFERENCE 
Actual situations are those that are asserted to have 
already occurred, or to be occurring at the time when a 
text is produced. This excludes, e.g., situations men- 
tioned in modal, intensional, negated, or frequentative 
contexts. 8 A predication denotes an actual situation 
when two criteria are satisfied. First, at least one of the 
verb's arguments must be interpreted as specific 
(Dowty 1979, Mourelatos 1981, Vlach 1981). For exam- 
ple, the simple past of fly denotes a specific situation in 
Sentence 1 but not in (2), because the subject of the verb 
in (2) is a nonspecific indefinite plural. 
1. John flew TWA to Boston. 
2. Tourists flew TWA to Boston. 
This paper does not address the interaction of the nature 
of a verb's arguments with the specificity of references 
to situations. 
The second criterion is that the situation must be 
asserted to hold in the real world for some specific time. 
Predications in modal contexts (including the future; cf. 
Sentence 3) are excluded because their truth evaluation 
does not involve specific real-world times, but rather, 
hypothetical or potential times. 
3. The oil pressure should/may/will decrease. 
Additionally, frequency adverbials like always may 
force a temporally nonspecific reading, as in (4). 
4. John always flew his own plane to Boston. 
PUNDIT's time component does not currently identify 
modal contexts, frequency adverbials, or nonspecific 
verb arguments. However, it does identify predications 
denoting situation types when the form of the verb itself 
provides this information. 
In evaluating actual time, PUNDIT distinguishes 
between examples like (5) and (6) on the basis of the 
verb and its grammatical categories. An actual use of 
the sentence in (5), for example, would report that a 
particular pump participated in a particular event at a 
specific time. 
5. The lube oil pump seized. 
6. The lube oil pump seizes. 
7. The lube oil pump seized whenever the engine 
jacked over. 
Sentences 6 and 7, on the other hand, report on types of 
recurrent events. In sentence 7, it is the adverb when- 
ever that indicates that the main clause refers to a 
recurrent type of event rather than to a specific event 
token situated at a particular time. In (6), it is the lexical 
aspect of the verb seize in combination with the present 
tense that provides that information. A further differ- 
ence between the two examples is that (7) entails that on 
at least one past occasion the pump actually seized 
when the engine jacked over, while (6) does not entail 
that the lube oil pump ever actually seized. We will see 
in Section 4.1 that (6) would immediately be determined 
not to evoke actual time on the basis of the lexical 
aspect of the verb and its inflectional form. Although 
PUNDIT does not yet handle frequency adverbials, 
Computational Linguistics, Volume 14, Number 2, June 1988 45 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
Section 5 illustrates the procedure by which the main 
clause of (7) would be processed so that its relation to 
the subordinate clause event could be identified later. 
Lexical aspect is the inherent semantic content of a 
lexical item pertaining to the temporal structure of the 
situation it refers to, and thus plays a major role in 
computing temporal information. The aspectual catego- 
ries and their relevance to temporal processing are 
discussed in Section 2.2. 
It should be noted that other semantic and pragmatic 
properties also affect temporal analysis. For example, 
there are conditions under which the present tense of a 
verb referring to an event, as in (6), is associated with an 
actual situation. Under the right conditions, first-person 
performatives (e.g., I warn you not to cross me) accom- 
plish the named event at the moment they are uttered 
(Austin 1977). Even a sentence like (6) can refer to an 
actual event if interpreted as a report of a presently 
unfolding situation, as in a sportscast. Handling tense in 
these types of discourse would require representing 
pragmatic features, such as the speaker/addressee rela- 
tionship, in order to handle the relation of indexicals 
like tense and person to the speech situation (Jakobson 
1957). Section 3 briefly mentions some semantic distinc- 
tions pertaining to the verb in addition to lexical aspect, 
which PUNDIT does handle. Otherwise, however, this 
paper focuses on temporal analysis of third person 
descriptions containing verbs whose arguments refer to 
specific, concrete participants. 
2.2 TEMPORAL STRUCTURE OF ACTUAL SITUATIONS 
Situations are classified on the basis of their temporal 
structure into three types: states, processes, and tran- 
sition events. Each situation type has a distinct tempo- 
ral structure comprised of one or more intervals. Two 
features are associated with each interval: kinesis and 
boundedness. Both terms will be defined more fully 
below, but briefly, kinesis pertains to the internal struc- 
ture of an interval, or in informal terms, whether 
something is happening within the interval. Bounded- 
ness pertains to the way in which an interval is located 
in time with respect to other intervals, e.g., whether it is 
bounded by another interval. 
This approach to the compositional semantics of 
temporal reference is similar in spirit to interval seman- 
tics in the attempt to account for the semantic effects of 
aspectual class (Dowty 1986, Dowty 1982, Dowty 1979, 
Taylor 1977). However, interval semantics captures the 
distinct temporal properties of situations by specifying a 
truth-conditional relation between a full sentence and a 
unique interval. The goal of PUNDIT's temporal anal- 
ysis is not simply to sort references to situations into 
states, processes, and events, but more specifically to 
represent the differences between the three situation 
types by considering in detail the characteristics of the 
set of temporal intervals that they hold or occur over 
(Allen 1984:132). Thus, instead of specifying a single set 
of entailments for each of the three situation types, the 
temporal semantics outlined here specifies what prop- 
erty of an interval is entailed by what portion of the 
input sentence, and then compositionally constructs a 
detailed representation of a state, ' process, or event 
from the intervals and their associated features. The 
critical difference from interval semantics is that while 
intervals are the fundamental unit from which situation 
representations are constructed, it is proposed here that 
intervals have properties that differentiate them from 
one another. 
2.2.1 SITUATION TYPES AND TEMPORAL STRUCTURE 
The three situation types--states, processes, and tran- 
sition events--are distinguished from one another en- 
tirely on the basis of the grammatically encoded means 
provided by the language for talking about how and 
when they occur. Peopl e certainly can and do concep- 
tualize finer differences among real-world situations and 
can even describe these differences, given sufficient 
time or space. But certain gross distinctions are un- 
avoidably made whenever people mention things hap- 
pening in the world. Here and in the next section we will 
examine the temporal distinctions encoded in the form 
of the verb, often referred to as aspect, which are here 
referred to as temporal structure. Part of the temporal 
structure, that which Talmy (1985) described as the 
pattern of distribution of action through time, is repre- 
sented in the time arguments for the three situation 
types. Another part of the temporal structure, its event 
time, is the component of temporal structure that gets 
located in time by tense and the perfect. All the relevant 
distinctions of temporal structure are represented in 
terms of intervals and moments of time. 
States. Very briefly, a state is a situation that holds 
over some interval of time, which is both stative and 
unbounded. A stative interval is one in which, with 
respect to the relevant predication, there is no change 
across the interval for which the situation holds. Thus 
stative intervals are defined here much as stative pred- 
ications are defined in interval semantics: 
An interval I over which some predication q/ holds is 
stative iff it follows from the truth of 4, over I that q/ 
is true at all subintervals of I (Dowty 1986:42). 
Sentence 8 is an example of a typical stative predication 
whose verb phrase is headed by an adjective. During the 
interval for which the predicate low holds over the 
entity pressure, each subinterval is equivalent to any 
other subinterval with respect to the asserted situation; 
thus its kinesis is stative. 
8. The pressure is low. 
Some of the diagnostic tests for stative predications are 
that they cannot be modified by rate adverbials (*The 
pressure was quickly low), nor referenced with do it 
anaphora (The pressure was very low. *The temperature 
also did it~that.) While inability to occur with the 
progressive suffix has often been cited as another 
diagnostic, it is a less reliable one. Dowty 1979 identifies 
a class of locative stative predications that occur in the 
46 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
progressive (e.g., The socks are lying under the bed.) 
Predicates denoting cognition or behavior have often 
been classified as statives but may occur in the progres- 
sive with reference to a cognitive or behavioral 
process. 9 Although such verbs do not appear in the 
current domain, they would be treated differently from 
pure stative verbs. 
The intervals associated with states are also inher- 
ently unbounded, although a temporal bound could be 
provided by an appropriate temporal adverbial (e.g., 
The pressure was normal until the pump seized). ~° 
When an unbounded interval is located with respect to 
another point in time, it is assumed to extend indefi- 
nitely in both directions around that time, as with the 
punctual adverbial in (9). The moment within the inter- 
val that is explicitly located by tense and the punctual 
adverbial is the situation's event time, depicted as a 
circle in the middle of the interval, with arrows repre- 
senting that the interval extends indefinitely into the 
past and toward the present. 
9. The pressure was low at 0800. 
Situation type: state 
Kinesis: stative 
Boundedness: unbounded 
This sentence would be true if the pressure were low for 
only an instant coincident with 0800, but it is not 
asserted to hold only for that instant; one thus assumes 
that it was low not only at the named time, but also prior 
and subsequent to it. In this sense, the interval is 
unbounded, as represented graphically above. 
Processes. A process is a situation which holds over 
an active interval of time. Active intervals contrast with 
stative intervals in that there is change within the 
interval, a useful distinction for interpreting manner 
adverbials indicating rate of change, e.g., slowly and 
rapidly. Since states denote the absence of change over 
time, they cannot be modified by rate adverbials; pro- 
cesses can be. 
The definition of active intervals is also adapted from 
the characterization of process predications in interval 
semantics: 
An interval I over which some predication qJ holds is 
active iff it follows from the truth of ~b at I that ~b is 
true over all subintervals of I down to a certain limit 
in size (Dowty 1986:42). 
Active intervals can be unbounded or unspecified for 
boundedness, depending on whether the verb is pro- 
gressive. In (10), the active interval associated with the 
alarm sounding is unbounded and bears the same rela- 
tionship to the named clock time as does the stative 
interval in (9) above. 
10. The alarm was sounding at 0800. 
< O 
Situation type: process 
Kinesis: active 
Boundedness: unbounded 
Progressive aspect has often been compared to lexical 
stativity.l~ Here the commonality among sentences like 
(9) and (10) is captured by associating the feature of 
unboundedness both with stative lexical items and with 
progressive aspect. The temporal structures of states 
and unbounded processes are thus identical with re- 
spect to boundedness. However, the distinction be- 
tween the kinesis of (9) and (10) is retained by distin- 
guishing active from stative intervals. 
In (11) the interval associated with the alarm sound- 
ing is unspecified for boundedness, meaning that the 
clock time may occur within the interval for which the 
alarm sounded, or at its onset or termination. 
11. The alarm sounded at 0800. 
Situation type: process 
Kinesis: active 
Boundedness: unspecified 
In (10), where the verb is progressive, the clock time is 
interpreted as falling within the unbounded interval of 
sounding, but in (11), where the verb is not progressive, 
the clock time can be interpreted as falling at the 
inception of the process or as roughly locating the entire 
process. ~2 Nonprogressive forms of process verbs ex- 
hibit a wide variation in the interpretation of what part 
of the temporal structure is located by tense. The 
influencing factors seem to be pragmatic in nature, 
rather than semantic. The solution taken here is to 
characterize the event time of such predications as 
having an unspecified relation to the active interval 
associated with the denoted process, represented graph- 
ically above by the dashed line around the event time. 
Transition Events. A transition event is a complex 
situation consisting of a process which culminates in a 
new state or process. The new state or process comes 
into being as a result of the initial process. Since states 
have no kinesis, they cannot culminate in new situa- 
tions. The temporal structure of a transition event is 
thus an active interval followed by--and bounded by-- 
a new active or stative interval.~3 
That there are these three distinct components of 
transition events can be illustrated by the following 
sentences in which the time adverbials modify one of 
the three temporally distinct parts of the predicated 
event. 
12. It took 5 minutes for the pump to seize. 
13. The pump seized precisely at 14:04:01. 
14. The pump was seized for 2 hours. 14 
The duration 5 minutes in (12) above applies to the 
interval of time during which the pump was in the 
process of seizing. The clock time in (13) corresponds to 
the moment when the pump is said to have made a 
transition to the new state of being seized. Finally, the 
measure phrase in (14) corresponds to the interval 
associated with the new state. 
Following Dowty 1986, Vendler's (1967) two classes 
of achievements and accomplishments are collapsed 
here into the single class of transition events, and for 
Computational Linguistics, Volume 14, Number 2, June 1988 47 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
much the same reasons. That is, achievements differ 
from accomplishments in being typically of shorter 
duration and in not entailing a sequence of subevents, 
but they nevertheless do in fact have some duration 
(Dowty 1986:43). Even so-called punctual events (e.g., 
They arrived at the station; She recognized her long-lost 
friend) can be talked about as if they had duration 
(Talmy 1985, Jackendoff 1987), apparently depending 
on the granularity of time involved. It is my belief that 
handling granularity depends on appropriate interaction 
with a relatively rich model of the world and of the 
current discourse, but would not require new units of 
time; depending on the level of detail required, mo- 
ments could be exploded into intervals, or intervals 
collapsed into moments. For these reasons, punctual 
events are not treated here as a separate class. 
With verbs in Vendler's class of achievements, the 
same participant generally participates in both the initial 
process and the resulting situation, as in (15): 
15. The engine failed at 0800. 
> 
Situation type: transition event 
Kinesis: active 
Boundedness: bounded 
Here, the engine participates in some process (failing), 
which culminates in a new state (e.g., being inopera- 
tive). In each case, however, there are two temporally 
distinct intervals, as shown in the diagram above, one 
bounded by the other. 
Causative verbs typically denote accomplishments 
involving subevents in which the action of one partici- 
pant results in a change in another participant, as in 
(16): 
16. The pump sheared the drive shaft. 
Here, a process in which the pump participated 
(shearing) is asserted to have caused a change in the 
drive shaft (being sheared). The consequence of the 
different argument structures of (15) and (16) on the 
event representation is discussed in the next section. 
The boundary between the two intervals associated 
with a transition event, the transition bound, is defined 
as a transitional moment between the initial active 
interval and the ensuing active or stative interval asso- 
ciated with the new situation. An important role played 
by the transition bound is that it is the temporal com- 
ponent of transition events that locates them with 
respect to other times. For example, (15) asserts that 
the moment of transition to the new situation coincides 
with 0800. In contrast with examples 9-11, the status of 
the engine prior to 0800 is asserted to be different from 
its status at 0800 and afterwards. The components of 
temporal structureJproposed here are intended to pro- 
vide a basis for deriving what is said about the relative 
ordering of situations and their durations, rather than to 
correspond to physical reality. Thus a transition bound 
is a convenient abstraction for representing how transi- 
tion events are perceived and talked about. Since a 
transition event is one which results in a new situation, 
there is in theory a point in time before which the new 
situation does not exist and subsequent to which the 
new situation does exist. This point, however, is a 
theoretical construct not intended to correspond to an 
empirically determined time. It corresponds exactly to 
the kind of boundary between intervals involved in 
Allen's (1983, 1984) meets relation. 
2.2.2 DOWTY'S ASPECT CALCULUS 
The intervals for which situations hold are closely 
linked with the semantic decompositions of the lexical 
items used in referring to them. This allows PUNDIT to 
represent precisely what kinds of situations entities 
participate in and when. The decompositions include 
not only N-ary relational predicates among the verb's 
arguments (Passonneau 1986), but also the aspectual 
operators for processes and events proposed in Dowty 
1979. The main clauses for examples 9, 10, 15, and 16 
are given below as examples 17-20. 
17. The pressure was low. 
Decomposition: low(patient(\[pressure 1 \])) 
18. The alarm was sounding. 
Decomposition: do(sound(actor(\[alarm 1 \]))) 
19. The engine failed. 
Decomposition: 
become(inoperative(patient(\[engine 1 \])) 
20. The pump sheared the drive shaft. 
Decomposition: 
cause(agent(\[pump 1 \]),become(sheared(patient 
(\[shaftl\])))) 
In (17), the semantic predicate low is associated with the 
predication be low, and is predicated over the entity 
referred to by the subject noun phrase, the pressure. 15 
The time component recognizes this structure as a 
stative predication because it contains no aspectual 
operators. 
The decomposition for (18) consists of a basic seman- 
tic predicate, sound, its single argument, and the aspect- 
ual operator do, indicating that its argument is in the 
class of process predicates; the actor role designates the 
active participant. 
The decompositions of transition-event verbs contain 
the aspectual operator become, whose argument is a 
predicate indicating the type of situation resulting from 
the event. With inceptive verbs, as in (19), the actor of 
the initial process is also the patient or theme of the 
resulting situation, although this dual role is not repre- 
sented explicitly in the decomposition. If a distinct actor 
causes the new situation, the verb falls into the class of 
causatives and the actor of the initial process is conven- 
tionally called an agent, as in (20). Other decompositional 
analyses (Dowty 1979, Foley 1984) conventionally rep- 
resent the initial process of transition-event verbs by 
associating an activity predicate (e.g., do) with the actor 
or agent of the initial process (e.g., cause(do(agent()), 
48 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
become(inoperative(patient())))). The decompositions 
in (19) and (20) can be considered abbreviated versions 
of these more explicit predicate/argument structures. 
The become operator of transition-event verbs thus 
provides a crucial piece of information used when 
deriving representations of transition events. Given a 
reference to a specific transition event that has already 
taken place, the temporal component deduces the ex- 
istence of the new situation that has come into being by 
looking at the predicate embedded beneath the become 
operator. This is described more fully in Section 4.2.3. 
As will be shown in Section 4, PUNDIT represents 
actual situations as predicates identifying the situation 
type as a state, process, or event. In order to familiarize 
the reader with the representation schema without 
needless repetition of detail, a single example of a 
situation representation is given below for (17). 
17. The pressure was low. 
state(\[lowl\], 
low(patient(\[pressurel\])), 
period(\[lowl\])) 
Each situation representation has three arguments: a 
unique identifier of the situation, its semantic decom- 
position, and its time argument, in this case, the interval 
(or period) over which the predicate holds. The same 
pointer (e.g., \[lowl\]) is used to identify both a specific 
situation and its time argument because the actual time 
for which a situation holds is what uniquely identifies it. 
The participants in a situation help distinguish it from 
other similar situations, but while the same entities can 
participate in other situations, time never recurs. 
Having introduced the distinct situation types and 
the temporal structures that distinguish them, the next 
steps are to show how they are computed and how they 
permit a simple computation of temporal location. This 
will be done in Section 4. Since the preceding discus- 
sions also introduced the representation of lexical as- 
pect and the relevance of the verbal categories, it is now 
possible to clearly summarize the input which the 
temporal analysis component receives. 
3 INPUT TO THE TEMPORAL COMPONENT 
PUNDIT's time component performs its analysis after 
the sentence has been parsed and recursively after the 
semantic decomposition of each predicating element in 
the sentence has been created (Palmer 1986). Although 
this paper focuses on the temporal analysis of certain 
kinds of tensed verbs, the basic algorithm described 
here has been extended to handle other cases as well. 
Describing the full input to the temporal component 
provides an opportunity to mention some of them. 
The input to the time component for each tensed 
clause includes not only the surface verb and its tense 
and aspect markings, but also the decomposition pro- 
duced by analyzing the verb and its arguments (cf. 
Section 2.2.2.). The input to the time component is thus 
a list of the following form: 
\[\[Tense, Perfect, Progressive\], 
Verb, Decomposition, {Context}\] 
Each element of the list will be described in turn. 
3.1 VERBAL CATEGORIES 
The first element in the input list is itself a list indicating 
the form of the verb, i.e., its grammatical inflection. 
\[\[Tense, Perfect, Progressive\], 
Verb, Decomposition, {Context}\] 
The tense parameter is either past or present.16 If the 
verb is in the progressive or perfect, the corresponding 
parameter appears while absence of either in the input 
sentence is reflected in its absence from the list. 
3.2 THREE ORDERS OF VERBS 
The next two elements in the input to the time compo- 
nent are the surface verb and its decomposition. Lexical 
aspect is encoded in the decomposition as described in 
Section 2.2.2 for the cases where it is relevant. How- 
ever, it is a more fundamental classification pertaining 
to the verb which helps determine the cases where 
aspect is relevant. 
\[\[Tense, Perfect, Progressive\], 
Verb, Decomposition, {Context}\] 
Since this information is only for treating more complex 
cases than are described in this paper, the following 
discussion is intended only to indicate that the model 
has been extended to cover verbs whose semantic 
structure contains temporal information of a different 
order than the inherent temporal structure of an actual 
situation. After a brief description of three temporal 
orders of verbs, the discussion will return to explication 
of the input required for implementing the basic model. 
In addition to the aspectual distinction among state, 
process, and transition-event verbs, there are other 
distinctions related to temporal semantics. A particu- 
larly significant one is among what I call first-, second-, 
and third-order verbs, by analogy with the distinction 
among first-, second-, and third-order logics. A first- 
order verb is one whose arguments are concrete enti- 
ties, e.g., humans, machines, and other physical ob- 
jects. A second-order verb takes as its arguments states, 
processes, and events, but does not in and of itself refer 
to a situation. Rather, its semantic content is primarily 
temporal or aspectual (e.g., occur, follow). Third-order 
verbs refer to complex situations (e.g., result, cause) 
whose participants are themselves situations. The 
aspectual distinctions among verbs referring to states, 
processes, and transition events are only relevant to 
first-order verbs. 
Second-order verbs can be identified by the impos- 
sibility of temporal modification of a situation referred 
to by the verb, independent of the situation(s) referred 
to by the verb's argument(s) (Newmeyer 1975), as can 
Computational Linguistics, Volume 14, Number 2, June 1988 49 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
Order Examples Definition 
First "fail", "operate" verbs that denote situations and 
whose arguments are not propositional' 
Second "occur", "follow" verbs that provide temporal 
information about their propositional arguments 
Third "result", "cause" verbs that denote situations but which also provide 
temporal information about their propositional arguments 
Table 1. Three orders of verbs. 
be seen 
(24). 
21. 
22. 
23. 
24. 
by contrasting examples 21 and 22 with (23) and 
The failure occurred on Tuesday. 
The failure was discovered on Tuesday. 
*The failure that happened on Monday occurred 
on Tuesday. 
The failure that happened on Monday was dis- 
covered on Tuesday. 
Example 22 mentions two distinct situations, a discov- 
ery and a failure. In (21), however, the subject of the 
sentence, the failure, denotes an event, but the verb 
occur does not denote a separate situation. It provides 
tense and aspect information for interpreting its argu- 
ment. In other words, the temporal information in (21) is 
very similar to that contained in (25): 
25. Something failed on Tuesday. 
A pragmatic difference between the two sentences is 
that in (21) it is not necessary to mention what failed 
whereas in (25), the verb fail must have a subject. Other 
verbs in this class are foUow, precede, continue, happen, 
and so on. Because these verbs contribute primarily 
temporal information, they are conventionally referred 
to as aspectual verbs (Freed 1979, Lamiroy 1987, New- 
meyer 1975). 
It is easy to see that the analysis of aspectual verbs 
must be implemented somewhat differently from verbs 
like fail, which directly denote situations. In a sentence 
like (25), the relevant temporal information is contained 
in the verb and its tense and aspect marking alone. In 
contrast, the temporal information in (24) pertaining to 
the fail event is distributed not only in the verb and its 
tense and aspect markers, but also in its subject. 
Temporal analysis of sentences like (24) must be per- 
formed not only at the main clause level, but also at the 
level of embedded propositions. In essence, analysis of 
aspectual verbs is of a different order. Consequently, 
verbs like fail are classified here as first-order verbs 
while the so-called aspectual verbs are classified as 
second order. 
PUNDIT's temporal component also handles a third 
class of verbs, classified as third order. A third-order 
verb denotes a real-world situation, but its arguments 
are other situations. Consequently, the verb may con- 
tribute temporal information about the arguments as 
well as about the situation it denotes. The verb result 
illustrates this type. Sentence 26 asserts the existence of 
a result situation; the result relationship holds between 
an instigating situation mentioned in the noun phrase 
loss of air pressure, and a resulting situation mentioned 
in the noun phrase failure. 
26. Loss of air pressure resulted in failure. 
Additionally, the meaning of result includes the tempo- 
ral information that the instigating situation (the loss) 
precedes the resulting situation (the failure). A full 
temporal analysis of sentences like (26) requires two 
steps. The first is to analyze the temporal structure of 
the situation denoted by the verb. The second is to draw 
the correct temporal inferences about the verb's prop- 
ositional arguments. Such verbs combine some of the 
properties of both first- and second-order verbs and 
thus constitute a third order of analysis. Classifying a 
verb as a third-order verb drives the search for temporal 
inferences associated with its arguments. 
The classification of these three orders of verbs, 
summarized in Table 1, is recorded independently of the 
lexical decompositions used by both the temporal- 
analysis component and the semantic interpreter. At 
present, verb-order information is used only by the 
temporal-analysis component. It essentially selects for 
the appropriate flow of control through the temporal- 
processing procedures. Although PUNDIT recognizes 
the distinction between first-, second-, and third-order 
verbs, and processes the relevant temporal information 
in each case, the remainder of the paper will deal only 
with the analysis of first-order verbs. 
3.3 LEXICAL ASPECT 
The third element in the input list is the decomposition 
structure produced by the semantic analysis of the verb 
and its arguments. 
\[\[Tense, Perfect, Progressive\], 
Verb, Decomposition, {Context}\] 
The important aspectual features of the decomposi- 
tions, discussed in Section 2.2.2, can be summarized as 
follows. If the decomposition of a first-order verb 
contains a become operator, the verb is in the transition- 
event class; otherwise, if it contains a do operator, the 
verb is in the process class; else, the verb (or other 
predicate) is stative. 
50 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
3.4 DISCOURSE CONTEXT 
The final element in the input to the temporal compo- 
nent is a data structure representing the current dis- 
course context. 
\[\[Tense, Perfect, Progressive\], Verb, Decomposition, 
{Context}\] 
The first element of this data structure is a list of 
unanalyzed syntactic constituents. At this stage of 
processing, PUNDIT has produced a full syntactic 
analysis of a surface sentence (or sentence fragment), 
and a semantic decomposition of some predication 
within the sentence. After the semantic analysis of a 
clause, the constituent list contains all those syntactic 
constituents that do not serve as arguments of the verb, 
e.g., adverbial modifiers of the verb phrase and sen- 
tence adjuncts. After the analysis of the main clause of 
Sentence 27, for example, the constituent list would 
contain two unanalyzed constituents; the prepositional 
phrase introduced by during, and the subordinate clause 
introduced by when. 
27. The pump failed during engine start, when oil 
pressure dropped below 60 psig. 
This list of constituents is processed after the temporal 
content of a predication is analyzed in the search for 
temporal adverbials that modify the predication (cf. 
Section 5 below). The data structure representing the 
current discourse context contains temporally relevant 
information, such as the tense and voice of the main 
clause. The main-clause tense is used for the analysis of 
situations mentioned in embedded tenseless constitu- 
ents, while voice is used in analyzing adjectival pas- 
sives. 
The next section describes an algorithm for interpret- 
ing the four pieces of information relevant to actual 
references to states, processes, and events. It demon- 
strates how the temporal structure and temporal loca- 
tion are generated from the verb's grammatical catego- 
ries of tense, perfect, and progressive, and from its 
lexical aspect. 
4 ALGORITHM FOR THE TEMPORAL ANALYSIS OF 
INFLECTED VERBS 
The introductory and discussion sections have undoubt- 
edly reinforced the view that semantic processing of 
temporal information is a complicated problem, even 
when the scope of the problem is constrained to the 
simple cases addressed here. Relevant information is 
distributed within and across distinct constituents, and 
their contribution to temporal information can depend 
upon co-occurring elements. Yet these are in no way 
insurmountable problems. The fundamental design prin- 
ciples behind my approach to temporal processing have 
been to carefully separate the analysis into distinct 
subtasks, to pare down to a minimum the information 
available to each task, and to provide a simple compo- 
sitional semantics for each kind of temporal input. In 
this section, I outline the basic algorithm for the tem- 
poral analysis of inflected verbs. This algorithm ana- 
lyzes the four components of the inflected verb de- 
scribed in the preceding section (lexical aspect, 
progressive, perfect, tense). The output that is gener- 
ated can than serve as input for further temporal proc- 
essing. Section 5 illustrates the integration of this basic 
algorithm into a more global procedure that succes- 
sively interprets the main and subordinate clauses of 
complex sentences where the subordinating conjunction 
is a temporal adverbial. 
The basic algorithm for the temporal analysis of 
inflected verbs has a simple tripartite control structure 
designed to answer three distinct questions: 
1. Does the predication denote a specific situation with 
actual time reference? 
2. If so, what is the temporal structure of the situation, 
i.e., how does it evolve through time and how does it 
get situated in time? 
3. Finally, what is the temporal location of the situation 
with respect to the time of text production, and what 
is the temporal vantage point from which the situa- 
tion is described? 
Figure 1 illustrates the algorithm's global control struc- 
ture, with the modules corresponding to each question 
as well as the relevant input for each module. The first 
module examines all four temporal parameters de- 
scribed in Sections 3.1 and 3.5 in order to reject certain 
cases. The second module requires only the two param- 
eters pertaining to the computation of temporal struc- 
ture. It sends a component of the temporal structure, 
the event time, to the third module, which locates the 
event time by analyzing the remaining two temporal 
parameters, tense and perfect. 
4.1 MODULE 1: ACTUAL TIME 
The first task performed by PUNDIT's temporal com- 
ponent is to identify references to specific situation 
tokens; that is, instances of situations which have 
actually occurred. The input is the lexical verb and its 
grammatical categories. In certain cases, the form of the 
verb itself can indicate that the predication refers to a 
type of situation, rather than to a specific token. Thus 
the screening step described here rejects these cases 
and otherwise assumes that the predication denotes a 
specific situation. As pointed out in Section 2.1, the 
verb itself provides insufficient information in two kinds 
of cases: those where explicit disconfirming information 
occurs elsewhere in the sentence (e.g., arguments of the 
verb, modals, frequency adverbials; cf. examples 2 and 
7, repeated below): 
2. Tourists flew TWA to Boston. 
7. The lube oil pump seized whenever the engine 
jacked over. 
and those where pragmatic features of the discourse 
context affect the interpretation of semantic input (as in 
a sportscast). While Module 1 currently serves only as 
a filter, it could be made to generate informative output 
Computational Linguistics, Volume 14, Number 2, June 1988 51 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
ASPECT PROGRESSIVE PERFECT 
4, 
L 
ACTUAL 
TIME 
I I 
ASPECT z PROGRESSIVE 
2. 
TEMPORAL 
STRUCTURE 
TENSE 
ET 
i PERFECT TENSE 
TEMPORAL 
LOCATION 
STATE (unbounded) ET is RT = ST 
PROCESS (unbounded) ET < RT = ST 
or 
(unspecified) ET is RT < ST 
TRANSITION 
EVENT (bounded) ET < RT < ST 
Figure 1. Algorithm for temporal analysis of inflected verb. 
for subsequent processing of semantically and pragmat- 
ically more complex phenomena. 
In Section 2.1 it was shown that two classes of 
inflected verbs generally denote situation types, rather 
than actual tokens. These are process verbs and transi- 
tion-event verbs in the simple present tense (i.e., non- 
progressive and nonperfect), as exemplified in (28) and 
(29). 
28. Number 2 air compressor operates at reduced 
capacity. (operate is a process verb.) 
29. They replace the air compressor every three 
years. (replace is a transition event verb.) 
For the compound tenses, present tense interacts with 
the progressive and perfect verbal categories. The pro- 
gressive alters the aspectual properties of nonstative 
verbs so that they refer to unbounded situations, and 
unbounded situations--unlike the other temporal struc- 
tures-can be located in the actual present (cf. Section 
4.2.2). With the perfect forms, the situation being 
referred to is always located in the past, and tense 
pertains to the situation's reference time rather than its 
event time (cf. Section 4.2.3). Thus, as shown in Figure 
1, all four elements in the temporal data structure are 
inspected in order to identify the two cases exemplified 
in (28) and (29). 
Table 2 summarizes the relation between the in- 
flected verb and actual temporal reference. 
In the current implementation of PUNDIT, predica- 
tions that meet the first condition do not receive further 
temporal analysis. 
4.2 MODULE 2: COMPUTE TEMPORAL STRUCTURE 
Module 2 computes the first type of specific temporal 
information associated with reference to an actual situ- 
ation. It generates an explicit representation of the 
situation's temporal structure. This structure includes 
one or more time arguments associated with the seman- 
tic predicates in the decomposition, and the situation's 
event time. Each situation type--state, process, transi- 
tion event--receives an appropriate situation label, time 
argument(s), and event time. The temporal structure 
evoked by an inflected verb can be computed entirely 
on the basis of the values of the two aspectual elements 
in its input (Lexical Aspect, Progressive), as shown in 
Figure 1. The algorithm for Module 2, summarized in 
Table 3, will be described in the following three sections 
corresponding to the three situation types. 
Though not shown in the figure or in Table 3, Module 
2 also receives another input data structure: the seman- 
tic decomposition. The decomposition is analyzed dur- 
ing the processing of transition-event situations in order 
to associate distinct time arguments with distinct se- 
mantic predicates in the decomposition. This procedure 
is explained in the appropriate section below. 
4.2.1 STATES 
As shown in Table 3, if the lexical aspect of the 
predicate is stative (Aspect = stative), then the progres- 
sive parameter is irrelevant for computing temporal 
structure. Lexical stativity is sufficient to identify the 
LEXICAL ASPECT PROGRESSIVE PERFECT TENSE ACTION 
Nonstative no no present reject 
accept 
Table 2. Module 1: Actual Time. 
LEXICAL PROGRESSIVE LABEL TIME EVENT 
ASPECT ARGUMENT TIME (ET) 
stative Yes/No State unbounded stative interval includes ET 
process or transition event Yes Process unbounded active interval includes ET 
process No Process unspecified active interval has ET 
transition event No Event transition bound unifies with ET 
52 
Table 3. Module 2: Temporal Structure. 
Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
situation as a state whose time argument is an un- 
bounded stative interval. 
Example 30 gives a simple stative sentence, the 
relevant input to Module 2, and the final situation 
representation. Note that (30) illustrates the use of the 
progressive with a verb in the locative class of statives 
noted in Dowty 1979, and mentioned in Section 2.2.1.17 
30. Metallic particles are clogging the strainer. 
Lexical Aspect = stative 
Situation Representation: 
state(\[clogl\], 
clog(instrument(\[materiall\]),theme (\[strainer2\])), 
period(\[clogl\])) 
As soon as the lexical aspect is recognized to be stative, 
Module 2 generates the state label and period time 
argument used in creating the representation depicted 
above. A period time argument in the context of a state 
representation denotes a stative interval. The situation 
representation in (30) indicates that a specific state, 
clogl, holds over the stative interval, period(\[clogl\]); 
the decomposition in the representation indicates the 
participants and the relation between them that holds 
over this interval. By definition, this interval also has an 
event time associated with it, whose relation to the 
interval we can determine by its boundedness feature. 
Stative intervals are assumed to be unbounded unless 
an endpoint is provided by further processing (e.g., 
through adverbial modification, inference). For un- 
bounded intervals, the event time is always an arbitrary 
moment included within the interval. This is represented 
as a binary predicate of the following form, where the 
moment time argument is the event time: 
Event Time = moment(\[clogl\]) 
such that includes(period(\[clogl\]), moment(\[clogl\])) 
This predicate and the state representation given above 
exemplify the output of Module 2 for state situations. 
The event time generated here is then passed to Module 
3 in order to determine its temporal location. We will 
return to this same example in the discussion of tempo- 
ral location in Section 4.3. 
4.2.2 PROCESSES 
There are three surface forms that denote process 
situations: nonprogressive process verbs, progressive 
process verbs, and progressive transition-event verbs. 
The nonprogressive and progressive cases have distinct 
temporal structures, due to differences in the relation of 
the event time to the active interval over which the 
process holds. Since this is the only difference among 
the three cases, the similarities in temporal structure 
will be presented before the event time is discussed. 
A nonstative predication that either has a process 
verb or is in the progressive (i.e., the three combina- 
tions of nonprogressive process, progressive process, 
and progressive transition-event) evokes a process rep- 
resentation. Thus the following three example sen- 
tences would each be represented with a process label 
and a period time argument, representing the active 
interval over which the process holds. Examples 31 and 
32 illustrate the two forms of process verbs that evoke 
process situations; since they receive the same repre- 
sentation, it is shown only once. Example 33 shows the 
third type of reference to a process, with a progressive 
transition-event verb. 
31. The diesel operated. 
Lexical Aspect = process 
Progressive = no 
32. The diesel was operating. 
Lexical Aspect = process 
Progressive = yes 
31-32. Situation Representation: 
process(\[operate 1 \], 
do(operate(actor(\[diesel\]))) 
period(\[operate 1 \])) 
33. The pump is failing. 
Lexical Aspect = transition event 
Progressive = yes 
Situation Representation: 
process(\[faill\], 
become(inoperative(patient(\[fail 1\]))), 
period(\[faill\])) 
The process representation for (33) contains the full 
decomposition for the verb fail with its aspectual oper- 
ator become. In this context, the become operator does 
not denote a transition to a new situation, but rather, 
indicates a process of becoming, which might or might 
not culminate in such a transition. 
Referring again to Table 3, we note that the active 
intervals for both (32) and (33) will be unbounded, in 
contrast to (31), where the active interval is unspecified 
for boundedness. The consequence of this difference on 
the representation of the event time is outlined in the 
following paragraphs. 
Unbounded processes. The predicate specifying the 
relation between the event time of an unbounded proc- 
ess and the period over which the process holds is 
identical to that for states. That is, the period time 
argument includes an arbitrary moment, which serves as 
the situation's event time, as shown below. 
32. The diesel was operating. 
Event Time = moment(\[operatel\]) 
such that includes(period(\[operatel\]), moment 
(\[operatel\])) 
33. The pump is failing. 
Event Time = moment(\[faill\]) 
such that includes(period(\[faill\]), moment(\[faill\])) 
The progressive always implies unboundedness, and in 
this respect resembles lexical statives. Again, it is 
important to remember that an unbounded interval can 
acquire endpoints through further processing (e.g., of 
temporal adverbials, as in The diesel was operating until 
the pump failed.). 
Unspecified processes. For nonprogressive process 
verbs, the period associated with the predication is 
unspecified for boundedness (cf. discussion of Example 
Computational Linguistics, Volume 14, Number 2, June 1988 53 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
LEXICAL PROGRESSIVE LABEL TIME EVENT 
ASPECT ARGUMENT TIME (ET) 
transition event No Event transition bound unifies with ET 
(From) Table 3. 
11 in Section 2.2.1). This gives rise to an indeterminate 
relationship between the event time and the period time 
argument over which the process holds; i.e., the event 
time may start, end, or be included within the period. 
This unspecified relationship is represented by means of 
a binary has predicate, as shown in example 31. 
31. The diesel operated. 
Event Time = moment(\[operatel\]) 
such that has(period(\[operatel\]), moment 
(\[operatel\])) 
Both event-time predicates given so far (i.e., includes, 
has) indicate a relation between an arbitrary moment 
and a single interval over which a state or process holds. 
There is otherwise nothing distinctive about the mo- 
ment selected to be the event time of a process or state 
situation. In contrast, as Table 3 indicates, and as 
discussed in Section 2.2.1, the event time of a transition 
event is equated with a distinctive component of its 
temporal structure, viz., the transition bound between a 
process that initiates the event and the new situation 
reached at the culmination of the process. 
4.2.3 TRANSITION EVENTS 
Table 3 shows only one component of the temporal 
structure of a transition event (the relevant line of the 
table is repeated below). 
As noted in Section 2.2.1, a transition event has three 
temporal components: an initial active interval leading 
up to a transition, the moment of transition, and the 
interval associated with the new, resulting situation. In 
theory, then, one could represent the full temporal 
structure of a transition-event predication (e.g., The 
pump failed) as three contiguous states of affairs: an 
initial process (e.g., failing) leading up to a transitional 
moment (e.g., becoming inoperative) followed by a new 
state of affairs (e.g., inoperative). At present, PUNDIT 
explicitly represents only the latter two components of 
transition-event predications: the moment (transition 
bound) associated with an event of becoming, and the 
period associated with the resulting situation. This 
representation has been found to be adequate for the 
current applications. Thus transition events are actually 
assigned two situation representations: an event repre- 
sentation with a moment time argument, represented 
with the input decomposition, and a resulting state or 
process situation with a period time argument, for 
which a new decomposition is derived from the input 
decomposition. Example 34 illustrates a typical transi- 
tion-event sentence, the relevant input for computing 
temporal structure, and the two situation representa- 
tions. 
34. The pump failed. 
Lexical Aspect = transition event 
Progressive = no 
Situation Representation: 
event(\[faill\], 
become(inoperative(patient(\[pump 1\]))), 
moment(\[faill\]) 
Situation Representation: 
state(\[fail2\], 
inoperative(patient(\[pump 1 \])), 
period(\[fail2\])) 
The first situation representation corresponds to the 
transition event itself. Module 2 generates the event 
label and moment time argument used in creating the 
type of event representation shown above for nonpro- 
gressive transition event verbs. The moment argument 
of a transition event is the transition bound implying the 
onset of a new situation. When Module 2 creates an 
event with a moment argument, it also creates a repre- 
sentation for the implied situation. In Example 34, the 
new situation is a state. When creating the representa- 
tion for the situation resulting from a transition event, it 
is necessary to determine the appropriate situation 
label, time argument, and semantic decomposition for 
the new situation. This is where the semantic decom- 
position for transition events plays a role, as will be 
described below. 
All transition-event verbs contain a state or process 
predicate embedded beneath an instance of the aspect- 
ual operator become. The full decomposition represents 
the type of situation associated with the moment of 
transition. The portion embedded beneath become is the 
situation type associated with the new situation. For 
example, the decomposition passed to the time compo- 
nent for Sentence 34 would be: 
become(inoperative(patient(\[pump 1 \]))). 
As shown in (34), this decomposition appears in the 
representation of the transition event itself. The argu- 
ment to the become operator is then extracted for use in 
the new situation representation: 
inoperative(patient(\[pump 1 \])) 
The extracted decomposition is inspected to determine 
its aspectual class, completely analogously to the pro- 
cedure for determining the aspectual class of the input 
predicate (cf. Section 3). In this case, the embedded 
predicate decomposition is stative because it contains 
no aspectual operators. If it contained the do operator, 
the new situation would have been a process. 18 In this 
fashion, the decomposition guides the selection of the 
54 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
situation label and time argument for the situation 
inferred to result from the transition event. 
The final piece of temporal structure derived for a 
transition event is the temporal relation between the 
moment associated with the transition event (e.g., 
moment(\[faill\])) and the period associated with the 
resulting situation (e.g., period(\[fail2\])). The event mo- 
ment is the onset of the period. Following Allen 1983, 
this is called a start relationship. By definition, then, 
every transition bound starts some period. In the case of 
Example 34, the moment of failure starts the period for 
which the pump is in an inoperative state. 
start(moment(\[fail 1\]), period(\[fail2\])) 
The event time of a transitional event is always identi- 
fied with the transition bound. Thus for examples like 
(34), the moment time argument serves as the event 
time of the transition event. This identity relation is not 
represented as a predicate, but rather, is handled via 
unification, as indicated in Table 3. 
4.3 MODULE 3: COMPUTE TEMPORAL LOCATION 
PUNDIT's temporal component employs a Reichen- 
bachian analysis of tense whereby situations are located 
in time in terms of three temporal indices: the event 
time, speech time, and reference time. ~9 It diverges 
from Reichenbach primarily by distinguishing between 
the event time and the temporal structure of a situation. 
While Reichenbach acknowledged that the progressive, 
for example, pertains to temporal duration, he did not 
discuss the differences in temporal structure associated 
with distinct situation types and their interaction with 
tense. Here, the event time is only a single component 
of the full temporal structure of a situation. In this 
section, we will see how this method of defining the 
event time makes it possible to compute temporal 
location independently of lexical or grammatical aspect 
while preserving the distinctive temporal information 
they contribute to references to actual situations. 
The tense and perfect parameters specify the se- 
quencing relations among the event time, reference 
time, and speech time, with each of the four configura- 
tions of tense and perfect specifying a distinct ordering, 
as shown in Figure 1 and repeated below: 
ET is RT = ST 
ET < RT = ST 
ET is RT < ST 
ET < RT < ST 
simple present 
present perfect 
simple past 
past perfect 
The speech time, or time of text production, is given. It 
serves as the temporal fulcrum with respect to which 
the other temporal indices are located. As shown in 
Table 4, the presence or absence of the perfect indicates 
whether the event time and reference time are distinct, 
in which case the event time precedes the reference 
time, or whether they are identical. Tense is taken to 
indicate the relation between the reference time and the 
speech time, following Reichenbach's suggestion: the 
position of R\[T\] relative to S\[T\] is indicated by the 
words "past", "present", and "future" (Reichenbach 
1947). 
PARAMETER VALUE RULES 
Perfect Yes precedes(ET,RT) 
No ET is RT 
Tense Past precedes(RT,ST) 
Present coincide(RT,ST) 
Table 4. Module 3: Temporal Location. 
Since we are dealing here with actual time, rather than 
potential or hypothetical time, there is only past or 
present. That is, the reference time either precedes or 
coincides with the speech time. 
The reference time and the event time are identical to 
one another for the simple tenses (ET is RT), which has 
the effect that tense applies to the event time. Thus, for 
the simple present, the event time and the speech time 
coincide. Note that a distinction is made here between 
identity and coincidence of distinct indices. For any 
speech act or text containing a description of a situa- 
tion, the speech situation and the described situation are 
always conceptually and observationally distinct, thus 
also their respective temporal indices. These indices are 
therefore represented as distinct times, which, in the 
present tense, happen to coincide. However, with the 
simple tenses, there is no reason to create a distinct 
reference time and a relation saying that it coincides 
with the event time. Rather, there are two different 
functions, which, in the case of the simple tenses, are 
filled by the same temporal index. The function of the 
reference time is explained more fully below. 
Webber (this volume) reviews and expands upon the 
role reference time plays in intersentential temporal 
reference. Reference time also plays a role in interpret- 
ing relational adverbials like now, yesterday, when, and 
so on. Adverbs like now and yesterday relate the refer- 
ence time of a predication to an implicit time, viz., the 
speech time. Relational adverbs like before, after, and 
when relate the time of the predication they modify to an 
explicitly mentioned time, i.e., the reference time asso- 
ciated with their syntactic complements. In the absence 
of the perfect, the reference time is identical with the 
event time, as in (35) and (36). 
35. The pressure is normal now. 
36. The pressure was low yesterday. 
In the perfect tenses, the reference time and event time 
are distinct. The event time of both the present and past 
perfect predications in (37) and (38) is past, i.e., the 
moment of failure is in the past. 
37. The pump has now failed, z° 
Computational Linguistics, Volume 14, Number 2, June 1988 55 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
38. The pump had failed when the gear began to 
turn. 
With the present perfect, it is the reference time that is 
present, as shown in (37) by the admissibility of the 
adverb now, which also refers to the present. On one 
reading of (38), the event time, or moment of failure, 
precedes the reference time, i.e., the time specified by 
the when clause. The perfect tenses can also be used 
simply to affirm truth or falsehood,~ thus (38) has 
another reading in which the perfect does not contribute 
a distinct reference time, but merely asserts that it is in 
fact the case that the pump failed when the gear began 
to turn. 
4.3.1 SIMPLE TENSES 
The distinct relations of event time to temporal struc- 
ture corresponding to the three categories of bounded- 
ness--unbounded, unspecified, and bounded--corre- 
late with distinctive behavior of the present tense. If the 
temporal structure associated with a predication is an 
unbounded interval, the simple present locates some 
time within the interval coincident with the speech time. 
Examples 35-38 illustrate the simple present in the 
context of the four types of predications that hold over 
unbounded intervals. 
35. The pressure is low. 
Lexical aspect: stative 
Progressive: no 
36. Metal particles are clogging the strainer. 
Lexical aspect: stative 
Progressive: yes 
37. The pump is operating 
Lexical aspect: process 
Progressive: yes 
38. The pump is failing. 
Lexical aspect: transition event 
Progressive: yes 
In these examples, the predicate is asserted to hold for 
some interval of unknown duration, which includes the 
speech time. Since this interval corresponds by defini- 
tion to actual time, it cannot be known to continue 
beyond the speech time into the future. However, that 
it can extend indefinitely into the past is illustrated by 
(39), where the situations referred to in the first and 
second conjuncts are assumed to be the same. 
39. The pressure is low and has been low. 
Predications involving process or transition-event verbs 
in the simple present have already been eliminated by 
Module 1 on the assumption that sentences like (40) and 
(41) do not refer to actual time. 
40. The pump operates. 
Lexical aspect: process 
Progressive: no 
41. The pump fails. 
Lexical aspect: transition event 
Progressive: no 
If a predication is not explicitly unbounded, i.e., if it has 
or may have an endpoint, then the present tense cannot 
be interpreted as locating the event time in the actual 
present. An event time located within an unbounded 
interval corresponds to persistence of the same situa- 
tion, whereas an event time that may also be an end- 
point corresponds to a transition. The way in which 
example:s like (40) and (41) are interpreted can be 
explained by considering that we cannot announce 
changes in the world at the exact moment that we 
perceive them, although in the guise of reportage or 
sportscasting, we act as though we can. 
In contrast to the simple present, the simple past can 
locate the event time of any temporal structure prior to 
the speech time. What is distinctive about the past tense 
in the context of the different temporal structures 
pertains to the temporal structure surrounding the event 
time. If the temporal structure is an unbounded interval, 
then the event time is some moment prior to the speech 
time within a persisting interval, and the same situation 
extends unchanged forward towards the present and 
back into the past. Example 42 illustrates the lack of 
contradiction in asserting the continuation up to the 
present of the past, unbounded situation mentioned in 
the first clause. 
42. The pump was failing and is still failing. 
The temporal structure associated with the situation 
mentioned in the first clause of (43), in the simple past, 
is an unspecified interval. Here it is unclear whether the 
two conjuncts refer to the same situation. Since the 
event time of the first conjunct is represented noncom- 
mittally, i.e., it may or may not be an endpoint of the 
interval, both interpretations are provided for by the 
representations generated here. 
43. The pump operated and is still operating. 
Finally, the simple past of a predication denoting a 
transition event definitely locates an endpoint. The 
event time of (44) is the transitional moment between an 
initial process of failing and a resulting state of being 
inoperative. 
44. The pump failed and is still failing. 
The first clause of (44) is represented by PUNDIT to 
assert the following temporal information: there was a 
moment of transition at which the pump failed, viz., its 
event time (moment(\[faill\])); this moment started a 
period in which the pump was inoperative (start 
(moment(\[faill\], period(\[fail2\]))); and finally, it pre- 
ceded the speech time (precedes(moment(\[faill\]), Speech 
Time)). The second clause cannot refer to the same 
transition event because a unique transition bound 
cannot both precede and coincide with the speech time, 
nor can it both be an endpoint of, and contained within, 
an interval. Rather, the second clause refers to a distinct 
situation, either a process that the speaker presumes 
will eventually result in a new failure, or an iteration of 
successive failure events. Of these two possibilities for 
the second clause, PUNDIT currently generates only 
the former. 
56 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
4.3.2 PERFECT TENSES 
The perfect tenses have a more complex semantics and 
pragmatics than the simple tenses. The semantic inter- 
pretation given here accounts for the temporal interpre- 
tations assigned to the perfect tenses in which the event 
time and reference time are distinct from one another. 
There are uses of the perfect that do not have these 
temporal effects, as pointed out in McCawley 1971, i.e., 
cases where the event time and reference time would 
not be distinct. Here we consider only the temporally 
relevant uses of the perfect, where each perfect tense 
specifies two temporal relations: in both cases, the 
event time precedes the reference time; and tense 
indicates whether the reference time coincides with or 
precedes the speech time. 
The following examples illustrate the present and 
past perfect with a variety of temporal structures. The 
only difference between these examples and the simple 
present tenses examined in the preceding section is the 
relation between the reference time and event time. The 
relation between temporal structure, event time, and 
speech time is the same as for the simple past. 
45. The engine has been operating. (unbounded 
process) 
46. The engine has operated. (unspecified process) 
47. The pump has failed. (transition event) 
48. The pressure had been low. (state) 
49. The pump had failed. (transition event) 
5 INTERPRETING TEMPORAL ADVERBIALS 
It is assumed that temporal adverbials can be analyzed 
in terms of the same components of temporal structure 
and temporal sequencing constraints that apply to situ- 
ations. The situation representations developed here 
provide a foundation for interpreting three distinct types 
of adverbial modification corresponding to the three 
features represented in temporal structure, i.e., kinesis, 
intervals, and moments. Rate adverbs like slowly and 
rapidly, which modify the manner in which situations 
evolve through time, modify active intervals and not 
stative intervals. For an example like (50), no explicit 
active interval would be represented, thus one would 
have to be coerced in order to interpret the adverb. 
50. The pressure was rapidly low. 
Examples like (51), on the other hand, provide a moti- 
vation for representing the initial active interval of a 
transition event (cf. Section 4.2.3), since the adverb 
essentially selects for such an interval. 
51. The engine quickly failed. 
Durational adverbials like for X, where X is a temporal 
measure phrase, modify any interval, but not their 
endpoints. Finally, relational adverbs, which specify 
temporal sequence, modify the reference time of situa- 
tions. 
Adverbials can combine relational and durational 
elements. In X, where X is a temporal measure phrase, 
not only specifies a duration, but also relates the 
endpoint of this duration to some other time, e.g., the 
time at which the utterance is produced, as in (52). 
52. The lights will go off in 10 minutes (e.g., from 
now). 
Temporal connectives like before and after can combine 
with temporal measure phrases to yield complex adver- 
bials specifying both a duration and a relation, as in (53). 
53. The engine seized five minutes before the alarm 
sounded. 
In this section, we will look briefly at the two types of 
durational phrases compared in Vendler 1967 in order to 
demonstrate the advantages of the representations de- 
veloped here for interpreting them. Then we will look 
briefly at the algorithm for interpreting complex sen- 
tences with subordinate adverbial clauses. 
5.1 DURATIONAL ADVERBIALS 
Unbounded situations. Predications denoting states and 
processes have duration, as shown by the interpretation 
of durational adverbial phrases of the formforX, where 
X is a time measure, as in (54) and (55): 
54. The pressure was low for 10 minutes. (state) 
55. The gear was turning for 10 minutes. (unbounded 
process) 
However, as noted in preceding discussions, the past 
tense in reference to states and unbounded processes 
does not apply to the whole duration. It applies to the 
moment within the interval designated as the situation's 
event time. Since in (54) and (55) the event time is past 
and the speech time is present, the two temporal indices 
create an explicit temporal extent within which to locate 
the durational phrases. The for adverbial phrase also 
evokes an unbounded duration, meaning that the mea- 
sure phrase does not necessarily encompass the entire 
duration, as shown by the lack of contradiction in 
asserting the continuation of the interval up to the 
present, as in (56) and (57). 
56. The pressure was low for 10 minutes and is still 
low. 
57. The gear was turning for 10 minutes and is still 
turning. 
The present perfect would allow one to assert some- 
thing semantically very similar to (56) and (57), but 
more laconically (e.g., The pressure has been low for 10 
minutes.). However, a context in which (56) would be 
more correct than the corresponding perfect is perfectly 
possible; it would have.to be a context where the 
pressure is now low, was low over some interval of 10 
minutes' duration, but where this interval is more than 
10 minutes prior to the present, and where the pressure 
has continued to be low up to the present (e.g., A: The 
alarm should go off if the pressure is low for 10 minutes. 
B: Well, the pressure was low for 10 minutes and it's 
still low, but the alarm still hasn't gone off.). 
The past tense with an unbounded interval evokes a 
span of time between the past event time and the 
present speech time within which to situate the measure 
of time given by a for adverbial. However, there is no 
Computational Linguistics, Volume 14, Number 2, June 1988 57 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
such span of time associated with the present tense of 
an unbounded interval, hence the impossibility of a for 
measure phrase in examples (58) and (59). 22 
58. ? The pressure is low for 10 minutes. 
59. ? The gear is turning for 10 minutes. 
Note that the present perfect, like the simple past, does 
provide a temporal point prior to the present, thereby 
creating a span of time for the durational phrase to apply 
to, as in (60) and (61): 
60. The pressure has been low for three hours. 
61. The gear has been turning for five minutes. 
The temporal structures generated for examples like 
(56)-(59) make it possible to correctly interpret the 
adverbial phrases they contain. The measure phrases in 
(56) and (57) can be interpreted not simply because the 
mentioned situations have duration, but more impor- 
tantly because of the distinctness of the two temporal 
indices, event time and speech time. In (58) and (59), 
where event time and speech time coincide, there is no 
explicit span of time within which to situate the measure 
phrase. Cases where there is no explicit component of 
temporal structure in the situation representation to 
match up with the temporal structure evoked by a 
temporal adverbial are probably candidates for the kind 
of coercion discussed in Moens and Steedman (this 
volume). 
The durational adverbial phrases in (62)-(64) not only 
specify a duration, but also an endpoint (Vendler ~967). 
Since progressive process predications are unbounded, 
there is no actual endpoint to be mapped to, hence, 
under one reading, (62) cannot be interpreted as a 
situation with an actual time; rather, it seems to refer to 
an activity that was supposed to take place five minutes 
from some time previously specified in the discourse 
context (e.g., paraphrasable as It was to be the case that 
the gear would turn five minutes from the present). 
There is another possible reading, paraphrasable as It 
turned out to be the case that the gear turned five 
minutes after some previously specified time, as in the 
context I applied some lubricant to the gear and it was 
turning in five minutes, which, like (58) and (59) above, 
may be examples requiring coercion. 23 In contrast, 
examples 63 and 64 can be interpreted as actual situa- 
tions whose endpoints coincide with the endpoints of 
the five-minute duration. 
62. The gear was turning in five minutes. 
63. The gear turned in five minutes. 
64. The engine was repaired in five minutes. 
The two types of durational adverbials behave differ- 
ently when modifying the different types of temporal 
structures in ways that tend to confirm the representa- 
tions proposed here. 
5.2 COMPLEX SENTENCES 
The temporal adverbials encountered in the CASREPs 
domain consisted predominantly of phrases introduced 
by temporal connectives, e.g., when, before, and after 
(Smith 1981). The general problem in analyzing the 
strictly temporal information associated with such con- 
nectives is to associate some time evoked by the matrix 
clause with some time evoked by the complement 
phrase. In general, connectives are represented as as- 
sociating the reference time of the matrix clause with 
the reference time of the compl'ement. The procedure 
involved in analyzing the temporal relations specified 
by a beJbre adverb (or other temporal connective) has 
the six steps illustrated in (65) below. 
(65) The compressor failed before the pump seized. 
Step 1: ,Analyze semantics of the main clause 
Step 2: Find reference time of main clause (RT1) 
Step 3: Recognize temporal adverb 
Step 4: Analyze semantics of subordinate clause 
Step 5: Find reference time of subord, clause (RT2) 
Step 6: Look up semantic structure of connective 
The compressor failed 
moment(\[faill\]) 
before 
the pump seized 
moment(\[seize 1 \]) 
precede(RTl, RT2) 
Result: precedes(moment(\[faill\]), moment(\[seizel\])) 
First, the temporal semantics of the main clause is 
analyzed. One of the outputs of this analysis is the 
reference time of the main clause, which in this case 
would be represented as moment(\[faill\]). Then the time 
component finds the adverbial phrase before the pump 
seized in the constituent list, which it recognizes as 
consisting of a temporal connective (before) and a 
complement. The complement clause is sent to the 
semantic interpreter (Palmer 1985) and is returned to the 
time component for temporal analysis. The fourth step, 
the temporal analysis of the subordinate clause, yields 
the information that the reference time of the subordi- 
nate clause is moment(\[seizel\]). Finally, the time com- 
ponent looks up the predicate structure representing the 
semantics of the temporal connective. Before is repre- 
sented as a binary predicatewpreeedes---whose first 
argument is the reference time of the main clause and 
whose second argument is the reference time of the 
complement clause. 
Currently, relational adverbs like before, after, and 
when are represented as predicates relating the refer- 
ence times of the modified and modifying situations. 
The procedure for handling temporal connectives as- 
sumes a priori that the reference times of the syntacti- 
cally superordinate and subordinate constituents are the 
required :input. In future work, these and other adverbs 
will be treated more explicitly as semantic predicates 
with selectional constraints that guide the search for the 
appropriate components of temporal structure associ- 
ated with the referents of the relevant constituents. 
6 CONCLUSION 
The situation representations presented here model the 
temporal meaning of inflected verbs by assigning a 
semantic value to each of four components; the inherent 
lexical aspect, the tense, and the presence or absence of 
58 Computational Linguistics, Volume 14, Number 2, June 1988 
Rebecca J. Passonneau A Computational Model of the Semantics of Tense and Aspect 
the perfect and progressive. Two significant advantages 
to the overall proposal are the simplicity of the algo- 
rithm that computes the representations, and the gen- 
erality of the building blocks used in constructing them. 
The algorithm accounts for the context dependencies 
among the four semantic components through a single 
mechanism, i.e., an appropriate characterization of the 
event time and its relation to the full temporal structure 
of a state, process, or transition event. These temporal 
structures are composed of intervals that may be active 
or stative, and that may be bounded, unbounded, or 
unspecified for boundedness. 
The situation representations have certain advan- 
tages in and of themselves. For example, the linkage 
between the components of temporal structure and 
Dowty's aspect calculus, and the incorporation of a 
Reichenbachian treatment of tense, make it possible to 
represent very precisely what predicates hold when. 
Further, the dual possibility of associating the become 
operator either with an unbounded interval or a transi- 
tion bound between intervals circumvents the so-called 
imperfective paradox. An additional advantage is the 
utility of these representations for further processing. 
The preceding section illustrated how the three building 
blocks of the representations (i.e., the notion of persis- 
tence of some situation through an interval, kinesis of 
the situation, and boundedness of the interval) make it 
possible to interpret accurately three corresponding 
kinds of temporal adverbials, and to identify those cases 
where coercion is required. Finally, explicit represen- 
tation of the reference times and event times within 
distinct types of temporal structures should make it 
possible to account for the differential contribution of 
situations to narratives and other types of discourse. 
ACKNOWLEDGMENTS 
I was fortunate in having the opportunity to consider the 
problems of temporal analysis in the context of a 
congenial work environment with stimulating col- 
leagues and a large, relatively comprehensive text proc- 
essing system. The members of my group provided 
much useful criticism and commentary, especially Ly- 
nette Hirschman, Deborah Dahl, Martha Palmer and 
Carl Weir. Bonnie Webber was extremely generous in 
her encouragement, and offered invaluable suggestions. 
I also profited from discussions with Mark Steedman, 
and his careful reading of earlier versions of this paper. 
This work was supported by DARPA under contract 
N00014-85-C-0012, administered by the Office of Naval 
Research. Approved for public release, distribution 
unlimited. 
NOTES 
1. Formerly Paoli Research Center, SDC--A Burroughs Company. 
2. Prolog UNDerstanding of Integrated Text: it is a modular 
system, implemented in Quintus Prolog, with distinct syntactic, 
semantic and pragmatic components (Dahl 1987a, Dahl 1986, 
Dowding 1987, Palmer 1986). 
3. PUNDIT has now been adapted to four domains. 
4. Webber, in work carried out in part at the Paoli Research Center, 
proposes a focusing algorithm for computing intersentential 
temporal relations which is analogous to Sidner's focusing 
mechanism for definite anaphoric expressions. Future work by 
Webber and Passonneau will integrate the two dimensions of 
inter- and intrasentential temporal analysis. 
5. Various types of tenseless predications are processed by PUN- 
DIT's temporal component, including nominalizations, certain 
clausal modifiers of noun phrases (e.g., pressure decreasing 
below 60 psig caused the pump to fail), and sentence fragments 
(Linebarger 1988). However, this paper focuses on the simpler 
case of tensed clauses. 
6. Reference time also plays a role in intersentential temporal 
reference (cf. Hinrichs, Moens and Steedman, Nakhimovsky, 
Webber, this volume). 
7. For the sake of brevity, the treatment of temporal adverbs with 
nominal complements is not described in this paper, but cf. Dahl 
1987b. 
8. These are not currently handled in the PUNDIT system. Predi- 
cations embedded in any one of these contexts do not directly 
denote specific situations but rather denote types of situations 
which, e.g., might occur, have not occurred, or tend to occur. 
Treatment of these contexts awaits the development of a repre- 
sentation which distinguishes between specific situations which 
hold for some real time and types of situations which hold for 
some potential time. One such proposal appears in Roberts 1985, 
which allows for the creation of temporary contexts. 
9. Cf. discussion of examples like I am thinking good thoughts, and 
My daughter is being very naughty, in Smith 1986. 
10. In general, temporal adverbials can modify an existing compo- 
nenl of temporal structure or add components of temporal 
structure. 
11. For comparisons of stativity and the progressive, cf. Vlach 1981, 
where the two are equated, Smith's counterargument (1986), and 
the interesting proposal in Mufwene 1984. 
12. Nakhimovsky (this volume) makes essentially the same argu- 
ment, namely that English lacks overt perfective grammatical 
aspect. In other words, the indeterminacy associated with the 
simple past of a process verb is evidence for the argument that 
the perfective or culminated reading associated with simple past 
transition event verbs, which are discussed in the next subsec- 
tion, is a consequence of the interaction between tense and 
lexical aspect, rather than of the simple past tense itself. 
13. This treatment of transition events closely resembles the event 
structure which Moens and Steedman refer to as a nucleus. They 
define a nucleus as a structure comprising a culmination, an 
associated preparatory process, and a consequent state (Moens 
1987). 
14. The durational adverbial in (14) forces a stative reading for a 
predicate which in isolation would be ambiguous between a 
passive and the adjectival passive (Levin 1986), in which the past 
participle is interpreted statively or adjectivally. 
15. The atom \[pressurel\] is an identifier of the entity referred to in 
the noun phrase and is created by PUNDIT's reference resolu- 
tion component (Dahl 1986). 
16. For sentence fragments such as erosion of blade tip evident, the 
tense parameter is actually untensed. The time component 
assigns present or past tense readings to fragments, depending 
on the aspectual class of the fragment (Linebarger 1988). 
17. As noted elsewhere, the aspectual classification of verbs is not 
completely determinate (Talmy 1985). Clog may very well be a 
verb that can refer either to a process or a state, and it might be 
possible to decide dynamically the lexical aspect of a specific 
instance through interaction with a sophisticated model, such as 
one which incorporates the notion of resource use, as suggested 
by Nakhimovsky (this volume). However, in PUNDIT the 
aspectual classification of verbs is domain dependent. 
18. The embedded decomposition never contains the become oper- 
ator; a decomposition with two become operators (e.g., 
become(become(inoperative(patient(\[pumpl\]))))) would be inco- 
herent. 
19. Reichenbach's (1947) treatment of tense and other token reflex- 
ive (indexical) elements is similar to Jakobson's (1957). 
20. Dowty and others have pointed out that the situation mentioned 
in a present perfect often persists up to the speech time (1982). 
However, this is generally not the case with reference to 
unbounded processes (e.g., The pump has operated), and seems 
to depend on a variety of pragmatic factors for the other situation 
types. 
21. Cf. McCawley's (1971) discussion of the ambiguities of the 
perfect, especially the assertorial perfect. 
22. Since isolated sentences can generally be given a variety of 
readings, it is often necessary to add qualifications regarding the 
intended reading of linguistic examples. Sentences like (56) and 
(57), for example, can be interpreted as the pressure is to be low 
for 10 minutes. Such interpretations are outside the scope of this 
paper, for they pertain to hypothetical rather than actual times. 
23. Thanks to Bonnie Webber and Mark Steedman for pointing out 
the second reading mentioned here, 

REFERENCES 
Allen, James F. 1983 Maintaining Knowledge about Temporal Inter- 
vals. Communications of the ACM 26(11): 832--43. 
Alien, James F. 1984 Towards a General Theory of Action and Time. 
Artificial Intelligence 23(2): 123-60. 
Austin, J. L. 1977 How to Do Things with Words. Harvard University 
Press, Cambridge, MA. 
Comrie, Bernard. 1976 Aspect: An Introduction to the Study of 
Verbal Aspect and Related Problems. Cambridge University 
Press, Cambridge, England. 
Dahl, Deborah A. 1986 Focusing and Reference Resolution in PUN- 
DIT. Presented at AAAI, Philadelphia, PA. 
Dahl, Deborah A.; Dowding, John; Hirschman, Lynette; Lang, 
Francois; Linebarger, Marcia; Palmer, Martha; Passonneau, Re- 
becca; and Riley, Leslie. 1987a Integrating Syntax, Semantics, 
and Discourse: DARPA Natural Language Understanding Pro- 
gram. R&D Status Report. Unisys Defense Systems, Paoli Re- 
search Center. Paoli, PA. 
Dahl, Deborah A.; Palmer, Martha S.; and Passonneau, Rebecca J. 
1987b Nominalizations in PUNDIT. Proceedings of the 25th 
Annual Meeting of the ACL. Stanford, CA. 
Dowding, John; and Hirschman, Lynette. 1987 Dynamic Translation 
for Rule Pruning in Restriction Grammar. Second International 
Workshop on Natural-Language Understanding and Logic Pro- 
gramming. Vancouver, B.C., Canada. 
Dowty, David. 1979 Word Meaning and Montague Grammar. D. 
Reidel, Dordrecht, Netherlands. 
Dowty, David. 1982 Tenses, Time Adverbials, and Compositional 
Semantic Theory. Linguistics and Philosophy 5: 23-55. 
Dowty, David. 1986 The Effects of Aspectual Class on the Temporal 
Structure of Discourse: Semantics or Pragmatics. Linguistics and 
Philosophy 9: 37--61. 
Foley, William; and Van Valin, R. 1984 Functional Syntax and 
Universal Grammar. Cambridge University Press, Cambridge, 
England. 
Freed, Alice. 1979 The Semantics of English Aspectual Complemen- 
tation. Reidel, Dordrecht, Netherlands. 
Jackendoff, Ray. 1987 The Status of Thematic Relations in Linguistic 
Theory. Linguistic Inquiry 18(3): 369-411. 
Jakobson, Roman. 1957 Shifters, Verbal Categories, and the Russian 
Verb. In Selected Writings. Mouton, The Hague. 
Lamiroy, Beatrice. 1987 The Complementation of Aspectual Verbs in 
French. Language 63(2): 278-98. 
Levin, Beth; and Rappaport, Malka. 1986 The Formation of Adjecti- 
val Passives. Linguistic Inquiry 17: 623-62. 
Linebarger, Marcia; Dahl, Deborah A.; Hirschman, Lynette; and 
Passonneau, Rebecca J. 1988 Sentence Fragments Regular Struc- 
tures. Proceedings of the 26th Annual Meeting of the ACL. 
Buffalo, NY. 
McCawley, James. 1971 Tense and Time Reference in English. In 
Studies in Linguistic Semantics. Fillmore, Charles J., and Langen- 
doen, D. Terence (eds.). Holt, Rinehart and Winston, New York, 
NY; 97-114. 
Moens, Mark; and Steedman, Mark J. 1987 Temporal Ontology in 
Natural Language. Proceedings of the 25th Annual Meeting of the 
ACL. Stanford, CA. 
Mourelatos, Alexander P.D. 1981 Events, Processes and States. In 
Tense and Aspect. Tedeschi, P.J., and Zaenen, A. (eds.). Aca- 
demic Press, New York, NY; 191-212. 
Mufwene, Salikoko S. 1984 Stativity and the Progressive. Indiana 
University Linguistics Club, Indianapolis, IN. 
Newmeyer, Frederick. 1975 English Aspectual Verbs. Mouton, The 
Hague. 
Palmer, Martha S. 1985 Driving Semantics for a Limited Domain. 
Ph,D. diss. University of Edinburgh, Edinburgh, Scotland. 
Palmer, Martha S.; Dahl, Deborah A.; Schiffman (Passonneau), 
Rebecca J.; Hirschman, Lynette; Linebarger, Marcia; and Dowd- 
ing, John. 1986 Recovering Implicit Information. Proceedings of 
the 24th Annual Meeting of the ACL. Columbia University, New 
York, NY. 
Passonneau, Rebecca J. 1986 Designing Lexical Entries for a Limited 
Domain. Logic-Based Systems Technical Memo No. 42. Paoli 
Research Center, System Development Corporation. Paoli, PA. 
Reichenbach, Hans. 1947 Elements of Symbolic Logic. The Free 
Press, New York, NY. 
Roberts, Craige. 1985 Modal Subordination and Anaphora in Dis- 
course. Paper presented at the 60th LSA. Seattle, WA. 
Smith, Carlota S. 1981 Semantic and Syntactic Constraints on Tem- 
poral Interpretation. In Tense and Aspect. Tedeschi, l?.J., and 
Zaenen, A. (eds.). Academic Press, New York, NY; 213-37. 
Smith, Carlota S. 1986 A Speaker-based Approach to Aspect. Lin- 
guistics and Philosophy 9:97-115. 
Talmy, Leonard. 1985 Lexicalization Patterns: Semantic Structure in 
Lexical Forms. In Language Typology and Syntactic Description: 
Grammatical Categories and the Lexicon. Cambridge University 
Press, Cambridge, England; 63-163. 
Taylor, Barry. 1977 Tense and Continuity. Linguistics and Philoso- 
phy 1: 99-220. 
Vendler, Zeno. 1967 Verbs and Times. In Linguistics in Philosophy. 
Cornell University Press, Ithaca, New York, 97-121. 
Vlach, Frank. 1981 The Semantics of the Progressive. In Tense and 
Aspect. Tedeschi, P.J., and Zaenen, A. (eds.). Academic Press, 
New York, NY; 271-292. 
Webber, Bonnie. 1987 The Interpretation of Tense in Discourse. 
Proceedings of the 25th Annual Meeting of the ACL. Stanford, 
CA. 
