Learning the Semantics of Aspect Gabriele SCHELER Institut fur Informatik Technische Universitat Munchen, D-80290 Munchen
[email protected] Abstract
The main point of this paper is to show how we can extract semantic features, describing aspectual meanings, from a syntactic representation. Aspectual meanings are represented as sets of features in an interlingua. The goal is to translate English to Russian aspectual categories. This is realized by a specialized language processing module, which is based on the concept of vertical modularity. The results of supervised learning of syntactic-semantic correspondences using standard back-propagation show that both learning and generalization to new patterns are successful. Furthermore, the correct generation of Russian aspect from the automatically created semantic representations is demonstrated. The results are relevant to machine translation in a hybrid systems approach and to the study of linguistic category formation.
1 Introduction A common goal of theoretical linguistic work as well as machine translation research is the construction of semantic representations, which can be used as interlingual representations in the context of machine translation. Constructing semantic representations is often controversial and results are dicult to evaluate, because little is known on what constitutes a semantic representation in the human brain and which information it must contain. However, semantic representations are a useful concept, when we have a speci c language processing task. In dierent tasks, we may use the same or similar semantic representations to interface with dierent modules for further processing. In this work, we concentrate on the task of translation, and accordingly interface the semantic representation with a language generation module for another language. To appear D.Jones, New Methods in Language Processing, University College London Press, 1995
1
This task-oriented approach has the important advantage that the speci c contribution of a semantic representation can be assessed and valued. The main point of this paper is to show how we can extract semantic features, describing aspectual meanings, from a specialized syntactic representation. The overall goal is to translate English to Russian aspectual categories, which can not be handled well by rule-based machine translation. The reason is mainly that the constellations of contextual information that determine semantic content and accordingly morphological categories in another language are too numerous and too diverse to be used as triggers for rules. source text tagging
?
specialized syntactic representation encoding
?
binary source text representation set of aspectual meanings
feature extraction
? encoding
-
binary feature vector
? language-speci c morphological category assignment
pattern classi cation
?
target text
Figure 1: Model of the Translation Process of
In Figure 1 a model of the complete translation process is shown. It consists 2
- an input tagger which provides syntactic tags to the continuous input text and thus transforms it to the specialized syntactic representation - the transformation of the syntactic representation into the semantic representation, which is a feature extraction task - a universal (interlingual) set of semantic features describing aspectual meanings - the categorization of the individual aspectual meanings to the languagespeci c grammatical categories, which is a pattern classi cation task. A signi cant trait of the model is that the semantic feature extraction process is realized by a specialized language processing module for dealing with aspectual information. This vertical modularity has been inspired by neurolinguistic research [McCarthy and Warrington, 1988; Marshall, 1988]. Instead of the more usual horizontal modules for morphology, syntax, lexical and sentential semantics, we use vertical modules that can be de ned as areas of processing that combine e.g. temporal morphology, temporal adverbials, temporal semantics and temporal cognition. Thus we have a number of specialized modules dealing with time, aspect, space etc. This means that we do not use a full syntactic representation, with all information that can be extracted by a parsing mechanism, but a specialized syntactic representation, containing in this case quanti cation of noun phrases (subject and direct object), lexical category of the verb with respect to aspectual properties, and temporal and modal sentential adverbs. There is quite a number of work which is related to the present one. Featurebased syntactic representations for pattern association with semantic representations have been used by [McClelland and Kawamoto, 1986] and [St. John and McClelland, 1990]. Their models are used to deal with the association between syntactic cases and semantic roles. The semantic representations achieved were not interfaced with any other module. A combined interpretation and generation approach has been used by [Miikkulainen, 1990] for the analysis of relative clauses. There, the internal representation also consisted of situations and roles. Another vertical module was modeled by [Munro and Tabasko, 1991]. Here locative prepositions were interpreted as spatial relations. However the model lacked generalization in the creation of semantic representations. The paper is structured as follows: In the next section, the use of a featurebased semantic representation is motivated. Then the notion of specialized syntactic representation is further explained, and the grammar that has been used in setting up the representation is presented. The following sections contain results of experiments in interpretation, in generation, and in translation. In the conclusion, results are assessed and further questions are being raised. 3
2 Interlingual (universal) semantic representation In this section, we propose a representation for grammatical meanings which are relevant to the morphological forms progressive/simple in English and imperfective/perfective in Russian. It is thus primarily an interlingual representation, i.e. a representation that is not constrained by a single language, but captures semantic properties of several languages. In principle, such a representation can be constructed to have universal properties, i.e. to capture the semantic distinctions that are used in any language, as this is a set that becomes exhaustive with the consideration of suciently diverse languages (cf. [Comrie, 1984] for a discussion of semantic universals). However, the set of features that we use here is only shown to be sucient for two languages. Namely, we can show that from these grammatical meanings the correct aspectual form of an English or Russian sentence can be predicted by presenting a classi er which computes the corresponding transformation function. The representation has the symbolic form of a number of distinct features and values for these features, which are encoded in binary form. Descriptive work, exploring aspectual distinctions in many dierent contexts and collecting dierent uses and conveyed meanings, has sometimes led to a formulation of aspectual meanings in terms of binary feature sets [Bondarko, 1983], [Breu, 1980], [Scheler, 1984]. Being the outcome rather than the starting point of research, these feature sets however were not validated for their capability of producing and predicting the correct morphology for a given sentence. In model-theoretic approaches to aspectual semantics [Hinrichs, 1988], [Dowty, 1979], [Verkuyl, 1993] the main eort consists in constructing logical representations for aspectual meanings. These representations are usually constructed for individual sentences, without paying attention to the functional dependency between overt morphological form and semantic representation and its computability. We use simple feature-value representations instead of model-theoretic constructs for two reasons: - we want the morphological form to be eectively computable from the semantic representation. - model-theoretic constructs, which capture inferences, are hardly necessary to perform an eective and correct translation. They seem to belong to another level of organization, a \deeper" semantic structure. However, feature-based representations can be constructed so as to have a natural logical interpretation, including inferential connections (cf. [Scheler, 1995]). The set of grammatical meanings (features) that have been used in these experiments is given in Appendix A. They consist of fteen features like habit4
uality, event-extension, event-type, degree of completion, duration and between two and ve values for each of these features, such as for event-type: state, atelic event, telic event, cause-state.
Note that the set must contain the features for all semantic distinctions that occur in the training and test sets in order to produce good learning and generalization results. The feature set which has been obtained heuristically (cf. [Scheler, 1984]) can thus be empirically validated. It is also possible to use feature selection techniques for elimination of super uous features/values (cf. [Duda and Hart, 1973]). Elimination of features may be important for generalization to new patterns, to exclude generalizing on irrelevant information. However such a pruning of the feature set has not been performed here. This approach uses certain assumptions about the nature of linguistic categories. It is presumed that in the language production process a certain morphological category is selected by virtue of a certain active feature set of meanings pertaining to this category (features on a semantic map). This feature set comes into existence from the cognitive and intentional states that accompany language production. Similarly, in interpretation, a certain morphological category creates a representation on the feature map.
3 Syntactic features for aspectual meanings The idea in using a specialized syntactic representation is that linguistic input is fed into several specialized modules for analysis, which take the necessary information for their task and retransform it into a suitable form, probably in several steps. As a starting point in this work we use a hypothetical intermediate representation, which can be gained from free text by a fairly simple tagger or parser. Relevant syntactic information has been selected in accordance with the detailed presentation in [Scheler, 1984], cf. [Partee, 1973; Vendler, 1967]. The categorizations presented here are all xed lexical decisions and not contingent on the use of these words in a given sentence. This is in accordance with the principle of a unidirectional parser, and could be changed by the integration of the semantic feature extraction module into a connectionist parser. The syntactic representation contains six slots, i.e. placeholders for functionally dierent terms, and a number of values for each slot. The following description is the \grammar" of the specialized syntactic representation for aspectual meanings:
quanti cation of subject: singular definite, plural definite, singular in-
and plural indefinite, as they are indicated by morphological marking on English noun phrases. temporal adverbials: These are distinguished into classes according to the definite
5
preposition (at, for, on) (or conjunction (when)) used, and the type of noun that forms the head, such as point (\8 o'clock") or period (\three hours"). Words such as now, today, yesterday form a category of their own.
morphological verb markers: past/present, perfect/non-perfect, progressive/simple.
sentential adverbs: These comprise
adverbs always, sometimes, x-times (\ ve times"), usually, adverbs such as suddenly, just, as independent categories, adverbials of manner (\slowly", \with great care", \well"), which have been categorized into those indicating processual nature mannerproc, and those that do not and are neutral in this regard (manner), and sentential negation (negation).
quanti cation of direct object: singular definite, plural definite, singular indefinite and plural indefinite.
lexical verb classes: A list of verb-classes sucient for the chosen examples has
been derived from [Scheler, 1984]: telic-action, achievement, activeperception, atelic-action, passive-perception, gradual state-change, stages-of-actions (\try", \begin", \ nish"), state, motion.
This syntactic representation is transformed into a semantic feature representation with pattern association techniques. The semantic representation is the basis for the selection of an aspectual category in another language. Initially, 49 patterns were created. (The examples are given in Appendix B.) From a standard English grammar [Thompson and Martinet, 1969] a number of sentences were selected, exemplifying the dierent uses of English tenses and aspects. The results especially for generalization could be much enhanced, when several syntactic variants for the most salient examples were created. I.e. examples were created to allow the classi er to determine the in uence of minor syntactic changes on the semantic representations. Five variations on the sentence \He is always doing homework" are shown in Table 1, together with their syntactic and proposed semantic representation. This is a method which is well known in theoretical linguistic work (cf. e.g. [Verkuyl, 1993] on aspectology), and which is also used much in second-language learning, i.e. the attempt to implement a speci c linguistic system onto an existing one. Possibly it is being used in rst language acquisition as well. 6
He has not done his homework.
negation * sing-def telic-action simple perfect present sing-def
past * * relational-to-present holistic * * * action existential telic negated * single nonhabitual. He was doing homework from 5 to 7. * from-point-to-point sing-def telic-action prog non-perfect past pluindef
past * * non-relational processual * occurs-at-period-in * action referential atelic * * single non-habitual. Sometimes he is doing his homework with much care. manner-proc,sometimes * sing-def telic-action prog non-perfect present sing-def
* * non-relational processual * * * action existential atelic * long-duration single habitual. Right now all students are doing their homework. * now plu-def telic-action prog non-perfect present sing-def
present * * non-relational processual extends-around * * action referential atelic halfway-through * single non-habitual. Yesterday, Tom did his homework twice. x-times yesterday sing-def telic-action simple non-perfect past sing-def
past hesternal * non-relational holistic * occurs-at-period-in * action referential telic completed * repeated non-habitual.
Table 1: Training Patterns with Syntactic Variation
4 Results of experiments
4.1 Categorization of grammatical meanings (Generation) The rst experiment concerns classi cation of grammatical meanings, i.e. learning and generalization of grammatical category assignment. A total of 63 semantic representations were used. Thereof 20 were created from English aspectual forms, 20 were created from Russian aspectual forms, and 23 were additional test cases. The following encoding technique for the symbolic features was used: For each feature, it was determined how many bits are necessary to code all possible values for this feature. A neutral value ('') has been added for each feature. This results in 34 bits for the semantic representation. The results are presented in Table 2. \Learning" concerns the correct assignment of categories by a neural net, which has been trained with complete input-output pairs, \generalization" concerns the correct assignment of a category to an unlabeled input by the trained neural net. Generalization to test cases and training sentences from the other language have been indicated separately. 7
The method used was \Standard Backpropagation" from SNNS (cf. [Zell and others, 1993]), presentation of patterns in topological order, 5 hidden units, 2 or 3 output units, input units as indicated, learning rate = 0:2. language learning correctness (abs=20) English 100% Russian 100%
test sentences (abs=23) 21/91% 22/96%
translation (abs=20) 17/85% 19/95%
generalization total (abs=43) 38/88% 41/95%
Table 2: Learning Grammatical Aspect The results can be summarized as follows: - Learning of the training set was easy, fully possible and quick for both English and Russian representations. - Generalization to new examples was present and can be considered successful. - There is not much dierence of generalization for additional test sentences and translations. This shows that category formation proceeds from arbitrary input patterns, as long as these are cognitively adequate and describe a possible state of aairs. The feature set is interlingual, no special language-speci c feature sets are required.
4.2 Semantic feature extraction (Interpretation)
The syntactic representations, consisting of six slots with between two and thirteen dierent values, have been translated into binary feature vectors, using the same technique as for semantic representations. The resulting coding has a length of 25 bits. The architecture was again that of a feedforward neural network, trained with standard backpropagation. A separate encoding level for the syntactic representation was included, i.e. a chance to recode the slot-value representation in an opaque way, which proved an asset in generalization (cf. [Hinton, 1986] for a discussion of recoding). The number of units in the dierent layers were 25-15-12-34. The training technique used was cross-validation by leaving-one-out (cf. [Weiss and Kulikowski, 1991], pp. 26{39) which is a preferential method for small samples of less than 100 patterns. Because semantic representations for the training set have to be created by hand, it is dicult to obtain large samples. However, as we shall see, the network learns well with small samples and can then generate 8
additional semantic representations, which can be checked for their suitability by interfacing them with another task. The network was trained with 48 patterns (2500 cycles) and then the remaining pattern was tested for generalization. A typical result for learning is shown in Table 3. This process was repeated 49 times, until all patterns had been tested. The results of generalization are also shown in Table 3. The third row (marked by y) indicates the results, when we exclude \outliers", i.e. patterns with more than 10 errors. learning phase total No. of patterns correct (%) No. patterns w/ n 5 errors (%) No. patterns w/ n > 5 errors (%) Avg. No. errors/pattern generalization phase total No. of patterns correct (%) No. patterns w/ n 5 errors (%) No. patterns w/ n > 5 errors (%) Avg. No. errors/pattern generalization phasey total No. of patterns correct (%) No. patterns w/ n 5 errors (%) No. patterns w/ n > 5 errors (%) Avg. No. errors/pattern
48 42 (87.5%) 48 (100%) 0 (0%) 0.18 49 5 (10.2%) 35 (71.4%) 14 (28.6%) 3.81 46 5 (10.8%) 35 (76.1%) 11 (23.9%) 3.23
Table 3: Learning and Generalizing Syntactic-to-Semantic Pattern Association The results show that learning, i.e. implementing the functional relationship is no problem, as expected with a small training set. The gures on generalization show that some implicit rules on how to set semantic values given syntactic input have been abstracted. The task was not an easy one, and an average of three to four errors per pattern remained. The results can be discussed as follows: Most information that is needed in order to determine semantic features for aspectual meanings is indeed local syntactic and has been captured by the specialized syntactic representation. However, for most patterns certain features could not be uniquely determined. There are several strategies possible for a remedy: One possibility would be to include more syntactic information. However, a 9
careful analysis has shown that further information is usually drawn from the same set of aspectually relevant syntactic features [Scheler, 1984]. To give an example: the present perfect of recency, (proximity: recent) is usually indicated by the adverb just. When \just" is missing, it is possible that a pattern nonetheless is somehow \similar" to other typical present perfects of recency. However, one would expect this similarity to be expressed by the same lexical category of the verb (achievement or telic-action) and/or by the absence of a direct object, typical present perfects of recency being of the type \He has (just) gone", \I have (just) eaten" etc. Therefore one would not expect to gain much by including more local syntactic information. Another question is whether features of lexical content and situational context would improve performance. It is dicult to obtain reliable features describing situational context (but cf. [Gallant, 1991] for an attempt). Lexical content features would have to be imported from other analyzing modules, a problem that has not been satisfactorily solved (cf. [McClelland et al., 1989], [Scheler, 1989] for a proposal on lexical-syntactic interaction during processing). An analysis of the example sentences that were not well generalized would lead to expect some improvement from the inclusion of a wider semantic context. The third idea is to have the system modify the semantic representation from knowledge of the likelihood of semantic representations. Such knowledge could be incorporated into weights in lateral connections among the semantic featurevalues. To make a distinction between such knowledge from experience and the learning related to the present task requires a dierent learning method. In this way, performance would be probably signi cantly improved.
4.3 Translation of aspectual categories
The nal experiment concerns the translation of aspectual categories, i.e. generation from the semantic representations that the system has found using untrained sentences. The training set was rst manually translated from English to Russian1 to provide a measure of success for the automatic translation. The generalized representations of the syntax-semantics mapping experiment (Table 3) were given to the Russian aspect classi er that had been trained on semantic representations before. The results are given in Table 4. They show that the errors were negligible in determining Russian aspect. The explanation for this successful performance is straightforward: The average error for generalization was less than 4 bits out of 34 bits. In addition, when neighboring binary features are wrong, this is often only one error on the symbolic level. Accordingly of the 15 features to determine a binary classi cation, only 1 or 2 were set dierently than in the learning task. 1
I wish to thank Dr. Terterjan for carefully checking all translations.
10
No. patterns Phase total correct(%) learning 48 47 (97.9%) generalization 49 45 (91.8%) Table 4: Translation of Aspectual Categories using System-Created Semantic Representations Two of the misgenerated patterns were also \outliers" in the interpretation task. This approach, which uses knowledge of meaning components and syntactic indicators is also contrasted with a \pure" neuronal or statistic approach, where English input is matched directly with Russian morphological categories, and a functional dependence is sought without further explicit representations. In this experiment, syntactic representations for 49 English sentences were created, encoded in binary form, and labeled according to Russian morphology. The results of learning and generalization are shown in Table 5. No. patterns Phase total correct(%) learning 48 48 (100%) generalization 49 38 (77.5%) Table 5: Translation of Aspectual Categories without Explicit Semantic Representation Accuracy of assigning Russian aspect in generalization was better than chance (50%), but not good enough to be practically useful for translation or checking of grammatical correctness. It was signi cantly lower than the results on using an explicit semantic representation, where 91.8% of generalized semantic representations, containing errors, were nonetheless correctly assigned to the Russian aspectual category. We have seen that the semantic representations are redundant with respect to category assignment, in the sense that there are fteen features involved for a binary choice, and several features usually concur. The fault-tolerance of the classi er allows to handle slightly \corrupt" representations and still assigns the correct category. Using explicit semantic representations as target structures for syntactic representations thus proves an asset in complex tasks like translation.
11
5 Conclusion It has been shown that extracting semantic features for a speci c area of meaning and generation of morphological categories is a task that is well suited for neural network models. Standard back-propagation is a useful method to determine whether the problem speci cation is correct, i.e. whether there are generalizable patterns involved and whether the problem has been put in such a way, as to make a functional implementation feasible. For a closer look, we may want to consider techniques of coding and corresponding learning mechanisms more fully (cf. [Scheler, 1994b]). In this paper, however, we have been concerned with opening up an area of investigation, namely learning the assignment of overt morphological categories to their semantic speci cation. It could be shown that the proposed function is learnable and generalizable by a standard mechanism. This approach is a considerable improvement over rule-based methods. The in uence of diverse syntactic information onto the semantic representation has not to be manually analyzed in detail in order to set up a great number of individual rules with dierent trigger patterns. Instead, the various aspects of the information in the syntactic representation are extracted and a semantic representaion is set up automatically. Generation is modeled as a pattern classi cation process, where a morphological selection is made on the basis of a semantic feature representation. This task could be shown to be easily solvable by a standard neural network. Accuracy in translation was tested with an intermediate, explicitly constructed semantic representation. It was notably better than a purely statistic approach, where English and Russian output were coupled directly. The advantage of a fault-tolerant mechanism was evidenced by the fact that translation results are highly successful even from semantic representations containing errors. We have presented a feature representation for aspectual grammatical meanings, a learned classi er for English and Russian morphological aspect and a transformer from a syntactic to a semantic representation. The trained system can be incorporated into an existing machine translation system as a specialized module which improves rule coded translation of grammatical aspect and reduces coding eort for new languages considerably (only tagging input and back-propagation learning is required). An important goal for further research is the translation of aspectual forms from continuous texts. For this goal, a syntactic tagger has to be added to the system. A similar system can then be used as a grammar checker for the correctness of aspectual forms.
12
A Semantic Representation: Features and Values 1. event-time
past - present - future
2. proximity (past)
recent - hodiernal - hesternal - less-than-a-year - more-thana-year
3. proximity (for futuric action) immediately - not-immediate
4. relational
relational-to-past - relational-to-present - non-relational
5. event-extension
punctual (instantaneous) - extended (processual, durative) holistic
6. reference point in time (event:) occurs-at - starts-at - extends-around 7. reference period (event:) occurs-at-point-in - occurs-at-period-in - occurs-atstarting-point - occurs-at-end-point
8. reference times/periods (event occurs at:) all-of-them (all/most) - some-of-them 9. action-status action - non-action
10. reference-type existential (inde nite) - referential (de nite) - general (mass noun reference) 11. event-type state - atelic-event - telic-event - cause-state
12. degree of completion
attempt - halfway-through - completed - negated
13. duration
long-duration - limited-duration (short duration)
13
14. number of occurrences single - repeated
15. habituality
habitual - non-habitual
B Example sentences 1. At six o'clock I am usually bathing the baby. usually at-point sing-def telic-action prog non-perfect present sing-def * * non-relational processual extends-around * * action existential telic halfwaythrough * single habitual. 2. Tom is always going away for the weekends. always for-period sing-def achievement prog non-perfect present * present * * non-relational punctual * occurs-at-starting-point * action existential cause-state completed * single habitual. 3. He is always doing homework. always * sing-def telic-action prog non-perfect present mass present * * non-relational processual * * * action existential telic halfway-through long-duration single habitual. 4. I am always tripping over this suitcase. always * sing-def achievement prog non-perfect present sing-def present * * non-relational punctual * * * action referential cause-state halfwaythrough * repeated non-habitual. 5. I taste salt in my porridge. * sing-def state simple non-perfect present mass present * * non-relational holistic * * * non-action referential state * * * nonhabitual. 6. I hear you well. manner * sing-def state simple non-perfect present sing-def present * * non-relational holistic * * * non-action referential state * * * nonhabitual. 7. Their children are really very quiet. * plu-def state simple non-perfect present attr * * non-relational processual * * * non-action general state * * * non-habitual. 8. I can't open the door, I am having a bath. now sing-def atelic-action prog non-perfect present sing-indef present * * non-relational processual extends-around * * action referential telic halfway-through * single non-habitual. 9. Are you liking this excursion? No I'm hating it. now sing-def state prog non-perfect present sing-def present * * non-relational processual extends-around * * non-action referential 14
state halfway-through long-duration single non-habitual. 10. I don't expect much of him. negation * sing-def state simple non-perfect present mass present * * non-relational processual * * * non-action general state * * * nonhabitual. 11. When John entered, I was bathing the baby. when-point sing-def telic-action prog non-perfect past sing-def past * * non-relational processual extends-around * * action existential telic halfway-through * single non-habitual. 12. I bathe the baby every day. every-unit-of-time sing-def telic-action simple non-perfect present sing-def * * non-relational holistic * * all-periods action existential telic completed * single habitual. 13. On Sundays I bathe the baby. on-periods sing-def telic-action simple non-perfect present sing-def * * non-relational holistic * * all-periods action existential telic completed * single habitual. 14. Most people bathe their baby every day. every-unit-of-time plu-def telic-action simple non-perfect present sing-def * * non-relational holistic * * all-periods action general telic completed * single habitual. 15. Tom is always going away for the weekends. always for-period sing-def achievement prog non-perfect present * present * * non-relational punctual * occurs-at-starting-point * action existential cause-state completed * single habitual. 16. Today Tom is going away to visit his brother. today sing-def achievement prog non-perfect present to-do future * immediately non-relational punctual * occurs-at-point-in * action referential cause-state completed * single non-habitual. 17. When Mary entered, Tom went away. when-point sing-def achievement simple non-perfect past * past * * non-relational punctual occurs-at * * action referential cause-state completed * single non-habitual. 18. Tom has just gone away to visit his brother. just * sing-def achievement simple perfect present to-do past recent * relational-to-present punctual * * * action existential cause-state completed * single non-habitual. 19. Tom has just gone away . just * sing-def achievement simple perfect present * past recent * relational-to-present punctual * * * action existential cause-state completed * single non-habitual. 20. He always does his homework. 15
always * sing-def telic-action simple non-perfect present sing-def present * * non-relational processual * * * action existential telic completed * single habitual. 21. He has not done his homework. negation * sing-def telic-action simple perfect present sing-def past * * relational-to-present holistic * * * action existential telic negated * single non-habitual. 22. He was doing homework from 5 to 7. from-point-to-point sing-def telic-action prog non-perfect past mass past * * non-relational processual * occurs-at-period-in * action referential telic * * single non-habitual. 23. Sometimes he is doing his homework with much care. manner-long,sometimes * sing-def telic-action prog non-perfect present sing-def * * non-relational processual * * * action existential telic * long-duration single habitual. 24. Right now all students are doing their homework. now plu-def telic-action prog non-perfect present sing-def present * * non-relational processual extends-around * * action referential telic halfway-through * single non-habitual. 25. Yesterday, Tom did his homework twice. x-times yesterday sing-def telic-action simple non-perfect past sing-def past yesterday * non-relational holistic * occurs-at-period-in * action referential telic completed * repeated non-habitual. 26. In Montreal at the airport I tripped over my suitcase and sprained an ankle. in-place sing-def achievement simple non-perfect past sing-def past longer * non-relational punctual * occurs-at-point-in * action referential cause-state completed * single non-habitual. 27. Suddenly they tripped over the suitcase and fell. suddenly * plu-def achievement simple non-perfect past sing-def past * * non-relational punctual occurs-at * * action referential cause-state completed * repeated non-habitual. 28. I am glad I haven't tripped over it. negation * sing-def achievement simple perfect present sing-def past recent * relational-to-present punctual * * * action existential cause-state negated * single non-habitual. 29. Unfortunately I am tripping over suitcases everywhere. every-unit-of-time sing-def achievement prog non-perfect present mass * * non-relational punctual * occurs-at-point-in all-periods action general causestate completed * single habitual. 30. For months I was tasting salt in my porridge, before I got to know, why. for-period sing-def state prog non-perfect past mass past longer * non-relational punctual * occurs-at-point-in * non-action existential 16
state * * * habitual. 31. Today I tasted salt in my porridge. today sing-def state simple non-perfect past mass past hodiernal * non-relational punctual * occurs-at-point-in * non-action referential state completed * single non-habitual. 32. I was tasting the porridge, when it exploded with a bang. when-point sing-def active-perception prog non-perfect past sing-def past * * non-relational processual extends-around * * action referential telic halfway-through * single non-habitual. 33. I have tasted the porridge thoroughly, it tastes good. manner-long * sing-def active-perception simple perfect present sing-def past * * relational-to-present holistic * * * action existential telic completed longduration single non-habitual. 34. I've been hearing all about this accident from him. * sing-def state prog perfect present all-of-x past recent * relational-to-present processual * * * non-action existential state completed long-duration single non-habitual. 35. Suddenly I heard an explosion. suddenly * sing-def state simple non-perfect past sing-indef past * * non-relational punctual * * * non-action referential atelic completed * single non-habitual. 36. For a long time we were hearing little explosions. for-period plu-def state prog non-perfect past mass past * * non-relational processual * occurs-at-period-in * non-action referential atelic * long-duration repeated non-habitual. 37. In winter birds hear a lot better than in summer. manner in-period plu-indef state simple non-perfect present * * * non-relational holistic * occurs-at-period-in some-periods non-action general state * * * habitual. 38. The children are being very quiet. now plu-def active-perception prog non-perfect present attr present * * non-relational processual * * * non-action referential state halfwaythrough long-duration single non-habitual. 39. On Sundays, the children are usually very quiet. usually on-periods plu-def active-perception simple non-perfect present attr * * non-relational holistic * occurs-at-period-in * non-action existential state * * * habitual. 40. On Sundays, the children were usually very quiet. usually on-periods plu-def active-perception simple non-perfect past attr past * * non-relational holistic * occurs-at-period-in * non-action existential state * * * habitual. 41. On Sundays, children are usually very quiet. 17
usually on-periods plu-indef active-perception simple non-perfect present attr * * non-relational holistic * occurs-at-period-in * non-action general state * * * habitual. 42. At 8 he was having breakfast. at-point sing-def atelic-action prog non-perfect past * past * * non-relational processual extends-around * * action referential telic halfway-through * single non-habitual. 43. He was always having breakfast at 8 in the morning. always at-point sing-def atelic-action prog non-perfect past * past * * non-relational processual extends-around * * action existential telic halfway-through * single habitual. 44. Every morning, from 8 to 8.30 I am having a bath. every-unit-of-time,from-point-to-point sing-def atelic-action prog non-perfect present * * * non-relational processual * occurs-at-period-in * action existential telic halfwaythrough * single habitual. 45. When the door rang, we were having breakfast in the kitchen. when-point plu-def atelic-action prog non-perfect past in-place past * * non-relational processual extends-around * * action referential telic halfway-through * single non-habitual. 46. I have just had a bath. just * sing-def atelic-action simple perfect present * past recent * relational-to-present holistic * * * action existential telic completed * single non-habitual. 47. I have had a bath twice today. x-times today sing-def atelic-action simple perfect present * past recent * relational-to-present holistic * * * action existential telic completed * repeated non-habitual. 48. I hate excursions. * sing-def active-perception simple non-perfect present mass * * non-relational processual * * * non-action general state * * * non-habitual. 49. I am expecting a letter today. today sing-def active-perception prog non-perfect present sing-indef present * * non-relational processual * occurs-at-period-in * non-action referential state halfway-through * single non-habitual.
References
[Bondarko, 1983] A.V. Bondarko. Principy funkcional'noj grammatiki i voprosy aspektologii. Nauka, 1983.
[Breu, 1980] Walter Breu. Semantische Untersuchungen zum Verbalaspekt im Russischen. Slavistische Beitrage 137. Otto Sagner, 1980. [Comrie, 1984] Bernard Comrie. Universals:form and function. In Brian Butterworth, Bernard Comrie, and "Osten Dahl, editors, Explanations for Language Universals, pages 87{103. Mouton, 1984. [Dowty, 1979] David Dowty. Word Meaning and Montague Grammar. Synthese Language Library 7. Reidel, 1979. [Duda and Hart, 1973] Richard O. Duda and Peter E. Hart. Pattern Classi cation and Scene Analysis. John Wiley, 1973. [Gallant, 1991] S.I. Gallant. A practical approach for representing context and for performing word sense disambiguation using neural networks. Neural Computation, 3:293{309, 1991. [Hinrichs, 1988] Erhard W. Hinrichs. Tense, quanti ers, and contexts. In Special Issue on Tense and Aspect. Computational Linguistics, 1988. [Hinton, 1986] G.E. Hinton. Learning distributed representations of concepts. In Proceedings of the Eighth Annual Conference of the Cognitive Science Society, 1986. [Marshall, 1988] John C. Marshall. Sensation and semantics. Nature, 334:378, August 1988. [McCarthy and Warrington, 1988] R. McCarthy and E. Warrington. Evidence for modality-speci c meaning systems in the brain. Nature 334, pages 428{ 430, 1988. [McClelland and Kawamoto, 1986] J. L. McClelland and A. Kawamoto. Mechanisms of sentence processing: Assigning roles to constituents. In D. E. Rumelhart and J. L. McClelland, editors, Parallel distributed processing: Explorations in the microstructure of cognition, pages 77{109. Cambridge, MA: MIT Press, 1986. [McClelland et al., 1989] J. L. McClelland, Mark St. John, and Roman Taraban. Sentence comprehension: A parallel distributed processing approach. Language and Cognitive Processes, 4(3/4):287{335, 1989. [Miikkulainen, 1990] Risto Miikkulainen. A PDP architecture for processing sentences with relative clauses. In Proceedings of Coling'90, Helsinki, pages 201{ 206, 1990. [Munro and Tabasko, 1991] P. Munro and M. Tabasko. Translating locative prepositions. In Proceedings of NIPS-91, volume 3, pages 598{604, 1991.
[Partee, 1973] Barbara Partee. Some Analogies between Pronouns and Temporal Expressions in English. Journal of Philosophy, 70:601{609, 1973. [Scheler, 1984] Gabriele Scheler. Zur Semantik von Tempus und Aspekt, insbesondere des Russischen. Magisterarbeit, LMU, Munchen, April 1984. [Scheler, 1989] Gabriele Scheler. Remarks concerning the interaction of grammar and semantics. Technical report, Computerlinguistik, Universitat Heidelberg, 1989. [Scheler, 1994a] Gabriele Scheler. Multilingual generation of grammatical categories. Technical Report FKI-190-94, Technische Universitat Munchen, Institut fur Informatik, April 1994. [Scheler, 1994b] Gabriele Scheler. Pattern classi cation with adaptive distance measures. Technical Report FKI-188-94, Technische Universitat Munchen, Institut fur Informatik, January 1994. [Scheler, 1995] Gabriele Scheler. Logical interpretation for semantic features. in preparation, 1995. [St. John and McClelland, 1990] M. St. John and J. L. McClelland. Learning and applying contextual constraints in sentence comprehension. Arti cial Intelligence, 46(1-2):217{257, 1990. [Thompson and Martinet, 1969] A.J. Thompson and A.V. Martinet. A Practical English Grammar. Oxford University Press, 1969. [Vendler, 1967] Zeno Vendler. Linguistics in Philosophy. Cornell University Press, 1967. [Verkuyl, 1993] Henk J. Verkuyl. A theory of aspectuality. The interaction between temporal and atemporal structure. Cambridge Studies in Linguistics 64. Cambridge University Press, 1993. [Weiss and Kulikowski, 1991] Sholom M. Weiss and Casimir A. Kulikowski. Computer Systems That Learn. Morgan Kaufmann, 1991. [Zell and others, 1993] Andreas Zell et al. Snns User Manual v. 3.1. Universitat Stuttgart: Institute for parallel and distributed high-performance systems, 1993.