Navigation – Plan du site
Comptes rendus

Bénédicte GuillaumeA Corpus-Based Study of Since Clauses in Contemporary English. Toulouse : PU du Mirail, 2014, 195 pages.

Texte intégral

1This book proposes an analysis of the properties of since clauses in contemporary English, based on a sample of 526 occur-rences, most of them from the British National Corpus. It is usually considered that since (Old English siÞÞan) used to have strictly temporal meanings and that, like a number of adverbial subordinators (cf. Traugott 1989), it acquired a causal value later on, in this case in Middle English; the aim of the study is to establish whether since should still be regarded as a single marker today (as for Lapaire & Rotgé 1991, for instance, for whom both values relate to the notion of origin), or whether the properties of temporal vs. causal since clauses are so different that there are now two homonymous subordinators (e.g. Deléchelle 1989). The author’s framework is mainly the French TOE theory (Théorie des Opérations Enonciatives), but rather than propose a full account of underlying operations and theoretical implications, she chooses to use some TOE tools to highlight certain phenomena. She also follows Lecercle (1990)’s notion of ‘remainder’: whatever theoretical model linguists devise, they will never account for all occurrences. There is always a remainder, that is, unclear or unstable instances, which are in fact a revealing and constitutive part of language. After detailing her data collection procedure, the author analyses causal since clauses (chapter 1), temporal ones (chapter 2), then the ‘remainder’, that is, ambiguous and hybrid cases (chapter 3).

2For her corpus, the author used the tagging function of the BNC to select the uses of since as a subordinator (vs. preposition or adverb). Out of the resulting 25,988 hits, she randomly selected 500, which ‘seemed to [her] the right size for [her] corpus to be varied enough yet manageable’ (p. 12). When an extract contained several occurrences of since, she included them all, hence the final total of 526. A manual review enabled her to correct a number of tagging errors and a certain lack of consistency in the tags, possibly due to different conceptions of ‘subordinators’ or ‘adverbs’. To her, since is a subordinator if and only if it introduces a clause that contains a finite verb phrase. For each occurrence, she entered the type of clause (temporal, causal or hybrid); its location relative to the main clause; tenses, aspects and modality in the since clause and in the main clause; the presence of negation; coordination; and any other element that might seem noteworthy.

3A major finding is that the two types of since clauses (causal and temporal) show diverging trends as regards their syntactic properties, but that they do not differ systematically. For instance, when a since clause is placed after the main clause, it is typically isolated by a comma if it has a causal value, and has no punctuation if it is temporal; but there are instances of postponed temporal since clauses preceded by commas. Simple tenses are particularly common in the main clause when the since clause has a causal value, whereas they are very rare when it has a temporal value (about 90% of the time, the main clause then has a perfect aspect); but again, there are instances with a simple tense in the main clause and a temporal since clause. Finally, causal since clauses are more often placed after the main clause than before (contrary to what might have been expected), and this is also the case for temporal since clauses, although postponing is even more frequent in the latter case.

4The author concludes from this that the syntactic properties noted in the corpus are not determined by the value of the since clause (temporal or causal), but by other factors. For causal since clauses, the place of the clause relative to the main clause depends on speaker attitudes. Postponing has a backgrounding effect, so that when a since clause is postponed, what is highlighted is the contents of the main clause. A fronted since clause, on the other hand, highlights the cause-effect relationship between the subordinate clause and the main clause (rather than the contents of the since clause). The author notes that the difference, however, is often ‘subtle’. Temporal since clauses are typically postponed because their function is then to provide an aspectual locator, but they have to be fronted when the main clause describes a situation that has evolved since the temporal landmark expressed in the since clause.

5Another finding of is that in a small number of cases, it is impossible to determine whether since has a purely temporal, or a purely causal, value. The context often allows for disambiguation, but sometimes it does not, as in, The Catholic religion has marginalised women, in part, one might say, since it was decided that God was male (p. 181). To the author, such ‘hybrid’ instances are evidence that since should still be regarded as a single marker in contemporary English.

6This book will be of interest to all those, whether students, the general public or scholars, who are interested in the notion of core value in grammar and in the evolution of grammatical words. It is very accessible and provides many examples and micro-analyses. A more explicit, broader state of the art would have helped situtate the author’s research within the existing literature on the topic, base some intuitions on facts, and show how the tools of the TOE framework allowed for improved analyses compared to other approaches. The reader might also have liked to read more about the stages in the shift from temporal to causal value in subordinators – this might have helped to define more clearly the notion of ‘single morpheme’ that the author retains for since. But the general conclusions make very convincing reading and will be of value to researchers.

7Laure Gardelle,
Ecole Normale Supérieure de Lyon

Haut de page

Pour citer cet article

Référence électronique

« Bénédicte GuillaumeA Corpus-Based Study of Since Clauses in Contemporary English. Toulouse : PU du Mirail, 2014, 195 pages. », Corpus [En ligne], 14 | 2015, mis en ligne le 29 août 2017, consulté le 19 septembre 2017. URL :

Haut de page

Droits d’auteur

© Tous droits réservés

Haut de page
  • Logo Revues électroniques de l’université de Nice