This book "supplies a vocabulary of English words and idiomatic phrases 'arranged … according to the ideas which they express'. The thesaurus, continually expanded and updated, has always remained in print, but this reissued first edition shows the impressive breadth of Roget's own knowledge and interests."
Coming soon: The Brill Dictionary of Ancient Greek by Franco Montanari is the most comprehensive dictionary for Ancient Greek to English for the 21st Century. Order your copy now!
Date: Thu, 12 May 2005 14:09:49 -0300 From: Raquel Freitag <rkofreitag@terra.com.br> Subject: Crosslinguistic Views on Tense, Aspect and Modality
EDITORS: Hollebrandse, Bart; van Hout, Angeliek; Vet, Co SERIES TITLE: Cahiers Chronos 13 TITLE: Crosslinguistic Views on Tense, Aspect and Modality PUBLISHER: Rodopi YEAR: 2005
Raquel Freitag, Federal University of Santa Catarina
This book is a new volume in the collection "Cahiers Chronos". It is composed by papers presented at 5th Chronos Conference (University of Groningen, The Netherlands, 2002). A short introduction containing an overview of the volume content and structure is followed by seventeen papers which are categorized into three sections: tense, aspect and modality. The editors' introduction alerts the reader to the fact that the ordering is "somewhat arbitrary given that some of papers across these rigid boundaries, as they discuss the interplay of tense and aspect or tense and modality" (p. i).
TENSE
In "Description of past events in Germany" Ten Cate discusses the phenomenon so-called 'Präeteritumschwund', the decline in the use of the preterit tense (simple past) to describe states-of-affairs, which took place before the time of speech. This form is token over by perfect tense. The phenomenon is not restrict to the present and preterit, but also affects the pluperfect. It is possible because there are contexts in which the pluperfect functions as a preterit as well. Following Reichenbach's theory (1947) (S = speech time; R = reference time; E = event time) Thieroff (1992) observes that the pluperfect tense is ambiguous because the temporal adverbs relate the E as well as R. Ten Cate examines the hypothesis that the R has disappeared in the German temporal system.
(1) a. Er war gestern abgereist 'He had departed yesterday'
Following Comrie (1985), in the reading in which the temporal adverb is related to E, a perfect or pluperfect would be adequate, as well.
(1) b. Er is gestern abgereist 'He has departed yesterday' (1) c. Er reiste gestern ab 'He departed yesterday'
In the alternative reading in which R is fixed by the temporal adverb, the use of pluperfect is obligatory.
(1) d. Eine Stunde bevor ich ihn gestern suchte, war er abgereist 'One hour before I looked for him yesterday he had departed'
Ten Cate rejects the hypothesis because if R is absent the ambiguity between preterit and pluperfect would be impossible. He concludes that one of the start factors for 'Präeteritumschwund' is the historical tendency to prefer analytic instead of synthetic verb forms.
In "The absolute and the relative present tense with future time reference in English and German", Beheydt purposes show the different uses of the present tense in two cognate Germanic languages. Her study is based in a corpus of English novels and their translations in Dutch. The present tense with absolute future time reference, as in (2) -- present simple -- and (3) -- present progressive -- locates a situation at some time later than the temporal zero-point.
(2) a. Either we 'do' business with Skolling or we 'go' bankrupt b. Of we 'doen' zaken met Solling of we 'gaan' failliet
(3) a. Actually he's publishing his memoirs. The Peverell Press 'are bringing' them out next spring. b. Hij zijn memoires publiceren. Peverell Press 'brengt' ze uit in het voojaar.
But, the present tense forms may also have relative future reference time, as in (4). The present tense expresses a simultaneity relation with another future time point.
(4) a. The book is in production [...], and Stilgoe won't want to break the contract if it 'means' publicly explaining why. b. Het boek is al in productie [...] en Stilgoezal geen contractbreuk willen plegen als hij vervolgens in het openbaar 'moet' uitleggen waarom.
Beheydt argues these two presents are distinguished. They are different temporal structures. The absolute present tense is motivated by the wish to represent a situation as a present fact, whereas the relative present tense represents a situation in a post-present temporal domain. Finally they have a different distribution -- the relative present tense never occurs in independent clauses, the absolute present tense mostly does.
De Cuyper' paper "Noun phrases and temporal information in Dutch" treats the relation between the temporal interpretation of nominal phrases and verbal phrases. She examines the importance of situation time of the verb to determine the situation time of the noun.
The author shows that Musan's proposal (1997) -- the quantification over individual triggers temporal independence and quantification over stages triggers temporal dependence -- do not predicts the right. Not just any stage of a individual can be taken into account in order to make temporal (in) dependence on the verb possible, but only certain specific ones.
She finalizes concluding that her paper offers further evidence to the legitimate of Carlson's (1977) distinction between stages, individuals and kinds.
Hollebrandse's paper "Sequence of tense: new insights form crosslinguistic comparisons" analyses the behavior of tense in complement clauses and relative clauses. Sequence of tense is the case in which an event in a main clause overlaps in the time with the event of its complement clause and a simple past morpheme, as in (5) linguistically expresses both events.
(5) John said that he was ill
Non-sequence of tense is the case in which the simultaneity between main and embedded clauses is expressed by using a present tense in the embedded clause, as Japanese in (6).
(6) Taroo-wa Hanako-ga byooki-da-to it-ta 'Taro said that Hanako was sick (at the time)'
Hollebrandse focuses on the distinction of sequence/non-sequence of tense in the languages. There are two important factors to the interpretation of tense in embedded event: complementation and imperfectivity. The author proposes a classification of languages in four types:
- Type A (English, Dutch) for which both the imperfectivity and complementation requirement are necessary; - Type B (Japanese) for which neither of them holds; - Type C (Polish, Russian) for which only complementation requirement holds; - Type D (Italian?) for which only the imperfectivity requirement holds. For this type, Hollebrandse notes that the data are still unclear.
In "Tense in indirect speech and thought: some proposed modifications" Vandelanotte provides a corpus study explaining relative and absolute tense in indirect speech of thought. Tenses in these constructions are fundamentally different from tenses in a 'non-reported' clause. She argues the fact that 'original speaker' or Sayer/Cognizant (Declerck 1991) in indirect speech or thought is important for a correct understanding of tense in speech or thought reports (e.g. the original speaker is tied to the deictic center separated from the reporting speaker).
(7) John said he would be late
(8) He said I am gay but that's not true
In (7) 'would' expresses the future in the Sayer/Cognizant's past. But in (8) the present expressed by 'am' is intensionally the Sayer/Cognizant and not the Speaker's.
Vandelanotte suggests the addition of a second Sayer/Cognizant timer-line in Declerck's framework (1991).
ASPECT
In his paper "Point of time" Arozio compares the imperfective paste tense form Imperfetto and present tense in Italian, considering the durative nature. He argues if we attempt to the distribution of durative adverbials in Italian, we capture some interesting tense proprieties of languages. Given the pattern of distribution of the durative temporal adverbials, Imperfetto and Presente verb forms are the morphological spell-out of two special tenses introducing point of time (cf. Heim 1994). Since Imperfetto and Presente introduce points of time they do not combine with per- and in- adverbial, given that these adverbials require the time introduced for these tenses to have some length.
(9) ?? Mario era malato per due giorni 'Mario was-IMP sick for two days'
On other hand, Imperfetto and Presente do combine with da- adverbials since these adverbials introduce an extended new interval abutting the reference time.
(10) Mario era malato da due giorni 'Mario was-IMP sick for two days'
The author concludes by suggesting crosslinguistic comparisons in further investigation of the points of time accounts.
Asnes's paper "Incompatibility between telicity and homogeneity in French" discusses the paradox 'homogeneous object-DPs vs. telic VPs', dealing with (apparent) counterexamples the claim that telicity entails heterogeneity and atelicity homogeneity. She analyses French quantifiers plusiers 'several', quelques 'some' and X N au plus 'at least x N'.
The paradox is that while a VP composed of a homogeneous V and of a heterogeneous object is interpreted as telic, there are cases where a VP may be telic although the combinations of the verb and the object has a homogeneous reading.
(11) a. Jean a dessiné plusiers/quelques cercles en une heure 'John drew several/some circles in an hour' b. Jean a lu trois livres au plus en deux mois 'John read at most three books in two months'
(12) a. Jean a dessiné trois circle en une heure 'John drew three circles in an hour' b. Jean a lu au moins trois livres en deux moins 'John read at least three book in two months'
The VPs in both (11) and (12) are telic because they are compatible with the time-span adverbials. The telic VPs in (11), as opposed to those in (12), are said to possess the subinterval property under the influence of their homogeneous object, and to satisfy the criterion of divisibility. It John drew some circles in an hour, it may be also true that he drew some circles in half an hour. This case, Asnes argues, should not be considered as a paradox given that DPs such as plusiers, quelques and X N au plus have a heterogeneous reference; they do not really satisfy the criteria of divisibility and cumulativity. And, the telicity of the situation denoted by VPs is implied by the quantifier's heterogeneity.
In "Degree scales and aspect" Caudal proposes a more detailed and principled account of the interaction between scalar expressions and adverbs. The term 'scalar expressions' includes degree modifier (Paradis 1971) and adverbial particles. At least two different levels of situational aspect information should be distinguished: 'stage' composed by expressions that give information about the stage structure of the eventuality, and 'scalar structure' composed by scalar expressions. The author suggests a third level, a mixed category that combines the feature [+stage] and [+scalar].
In Reichenbach's framework (1947) the Past Progressive (PPROG) of English and the Imparfait (IMP) of French report an eventuality E that temporally contains the reference time R. Molendijk's paper "The imparfait of French and the past progressive of English" shows that the IMP has a much largest distribution then PPROG.
(13) a. Quand Pierre entra, 9 heures donnaient. b. When Peter came in, the clock was striking 9. c. Quand Pierr entra, 1 heure sonnairt. d. ?? When Peter came in, the clock was striking 1.
IMP occurs in frequentative contexts (as in 13 a,c IMP is independent of the number of strikes of the clock), contrary to what can be said about PPROG (it seems natural if the clock strikes more than once). PPROG cannot be used to express perfect simultaneity if R is an instant since it is durative. Other property of PPROG is that the reference time of the PPROG- sentences must be explicitly mentioned. These facts Molendjik argues can be explained by claiming that IMP-sentences simply state the eventuality is true at R, whereas the use of PPROG implies that the eventuality is not simply true at R but also ongoing at this moment of time.
In "Morphological and telicity aspect with accomplishment VPs in Greek" for the necessity to separate the two aspectual domains, predicational/telicity aspect and grammatical/viewpoint aspect in Greek. The grammatical aspectual distinction perfective/imperfective is morphologically encoded in the verb, while the verb and its object complement determine the lexical aspectual distinction.
She shows that the complement of verbs of creation (write, build) and verbs of consumption (eat, drink) that can be appear bare in Greek can affect the aspectual interpretation of VP in Greek.
(14) a. I Maria troi portokali the Maria eats orange.ACC 'Maria eats/is eating an orange'
a' I Maria troi ena portokali the Maria eats an orange.ACC. 'Maria eats/is eating an orange'
(14)b. I Maria grafi gramma ston Yanni the Maria writes letter.ACC to-the Yanni 'Maria writes/is writing a letter to Yanni'
b' I Maria grafi ena gramma ston Yanni the Maria writes a letter.ACC to-the Yanni 'Maria writes/is writing a letter to Yanni'
(14) c. I Maria xtizi spiti fetos the Maria builds house.ACC this year 'Maria builds/ is building a house this year'
c' I Maria xtizi ena spiti stin eksoxi the Maria builds a house.ACC in-the-country 'Maria builds/ is building a house in the country'
Sioupi concludes that (a)telicity and (non)delimitedness are not synonymous in Greek: (a)telicity depends on the determiner (a VP which consists of a DP argument with an indefinite determiner is interpreted as telic, while if the same VP has as its object a bare singular count name it is atelic); (non)delimitedness is determined by perfective/imperfective distinction aspect (the eventuality in imperfective aspect is non- delimited, while in perfective aspect is delimited, independently of DP complement).
Verkuyl's paper "How (in-) sensitive is tense to aspectual information?" focuses two questions:
(i) How to deal with the opposition between grammatical aspect (perfective/imperfective) and lexical aspect (Aktionsart)? (ii) Whether or not aspectual information penetrates into the tense system and if so, how deeply?
About the question (i) Verkuyl observes that the distinction perfective/imperfective in Slavic languages do not have a direct counterpart in Germanic or Romance languages. In these languages (Germanic and Romance) grammatical aspect marking is conflated with the tense marking and the tenses seem also to carry some aspectual meanings. He also discusses the traditional association between the telicity (T = terminative aspect) and the perfectivity (pfv). There are empirical evidences from English, French and Polish that T is not equal pfv. And to response the question (ii) Verkuyl's point is that the tense is insensitive to aspect (lexical and grammatical).
In "The aspectual readings of the progressive form in Brazilian Portuguese" Wachowitz observes the progressive form in Brazilian Portuguese (BP). This construction is formed with auxiliary 'estar' and the present participle of main verbs + '-ndo'. Differently of English -- in which progressive cannot be combined with stative verbs --, in BP the progressive form is applied to all verbs.
(15) a. João está sabendo quatro lições de francês. b. * John is knowing four French lessons. c. A árvore está vivendo. d. * The tree is living.
When the progressive in BP is associated at stative verbs, it expresses a permanent state-of-affairs interpretation. And with non-stative verbs, the BP progressive can carry not only episodic but also habitual reading.
Based on Verkuyl's aspectual calculus, Wachowitz purposes that at level of inner aspectuality '-ndo' introduces durativity; adverbials or context determine episodic vs. habitual readings at level of outer aspectuality.
The last paper on aspectual domain, Xiao & McEnery's "Situation aspect: a two-level approach" analyses English and Chinese corpora. The authors propose a two-stage approach in which the situational aspect is determined firstly in the sentence, based on the verb class (six classes of situational aspect are distinguished: activities, semelfactives, accomplishments, achievements, individual-level states and stage-level states). At second level the aspectual value of sentence is modified by arguments, by adjuncts and by viewpoint.
The data-corpora evidence points that the situational aspect is to a great extent language independent, whereas viewpoint aspect is very language specific.
MODALITY
In the paper "Mood choice and sentence interpretation in Spanish" Ahern treats the alternation between indicative and subjunctive mood in Spanish.
(16) María te há dicho que {está/estés} preparado. a. Maria has told that you are ready (indicative). b. Mariahas told you to be ready (subjunctive).
The principal questions are what the features this group of predicates in Spanish has in common that can license the appearance of other one of the verbal moods in their argument clauses and what effects of mood choice on the interpretation of the sentence.
She concludes that the choice of moods is an expression of propositional attitude, which puts a restriction on the undetermined meaning of the predicate of the main clause.
Declerck's paper "The relation between temporal and modal uses of indicative verb forms" points that the use of past tense in English is governed by the some mechanism as that a non-modal ones. He describes three mechanisms that trigger a modal reading in English. (i) Shift of domain from one absolute sector to other;
(17) I didn't realize that the man {was/is} not be trusted.
(ii) Shift of temporal focus;
(18) The stranger doesn't understand English.
(iii) Shift of temporal perspective;
(19) a. That morning Hamlet decides to kill his father (past time treated as if it were present) b. I hear you have promoted (pre-present time treated as if it were present) c. Next time I want to be informed at once if there is a problem. (post- present time treated as if it were present)
The third shift gives rise to non-factual, epistemic (tentative) or contrafactual readings. The author concludes that the same mechanisms underlie the modal and temporal uses of tense and he shows that the future tense is not a separate mood, but a tense with modal implications.
In "On the nature of the epistemic readings of the Italian modal verbs: the relationship between propositionality and inferential discourse relations" Rocci discusses the epistemic readings of the Italian modal verbs dovere 'must' and potere 'may'/'can'. The uses of modal verb potere in personal constructions are related with a generic/alethic reading, whereas the uses of modal verb dovere (doxastic-evidential) are consequences of the non-propositional states of this modal.
Finally, the purpose of Werner's "The temporal interpretation of some modal sentences in English (involving a future/epistemic alternation)" is to capture the difference between the future and epistemic/present readings of modals. The author suggests that the temporal interpretation of modals is the result of the interaction between general interpretation and the modal base.
While not all of papers follow the same semantic theory, the papers in this volume are an interesting and valuable contribution towards a tense- aspect-modality description of languages.
REFERENCES
Declerck, R. (1991). Tense in English. Its structure and use in discourse, London: Routledge.
Heim, I. (1994). Comments on Abusch's theory of tense, manuscript, MIT.
Musan, R. (1997). On the temporal interpretation of noun phrases, New York/London: Garland.
Reichenbach, H. (1947). Elements of symbolic logic, New York: Macmillan.
Thieroff, R. (1992). Das finite Verb in Deutschen: Tempus -- Modus -- Distanz, Tübingen: Narr.
ABOUT THE REVIEWER:
ABOUT THE REVIEWER
Raquel Freitag is a linguistics student at the Federal University of Santa Catarina, Brazil. Her field of research (Ph. D. dissertation in progress) covers the tense-aspect-modality of Preterito Imperfeito (imperfective past) in a functional approach of spoken Portuguese in the South of Brazil.
