The Sequence of Tenses

1.1. Traditional viewpoints Tense is usually said to be a deictic category, in that the truth of a tensed sentence is relative to speech time (that is, to the context of utterance). This is clearly the case in independent sentences. Complement clause tenses are not directly linked to Speech Time or Utterance Time (UtT). Instead, they are linked to higher tenses that are linked to ST or to even higher tenses, etc. Therefore, complement tenses are linked to Ut-T only indirectly. One of the most salient aspects of the systematic dependence of the complement clause on the main clause is the fact that the semantic interpretation of tenses in the complement clause depends on the main clause. This semantic dependence of complement tense on matrix tense also has a syntactic aspect, traditionally discussed under the rubric of the sequence of tenses (= SOT) of the Indicative Mood. English is a language that manifests the sequence of the tenses (= SOT), in the sense that the presence of a particular tense in the main clause may place constraints on which tenses are allowed in the subordinate clauses, and also constraints on how tenses in subordinate clauses are interpreted. The simplest formulation of the sequence of the tenses rule says that a past tense in a main clause requires only past tenses in the subordinate clauses. There are two main ways of approaching the SOT: a) The SOT may be viewed as an effect of indirect discourse. When reporting a person's statement under a past verb of saying, the reporter must back-shift all the present tenses in the reported statement to the corresponding past tenses. A recent statement of this position is the Longman Grammar (1999):