Present tense in relative clauses as evidence for sequence of tense in French

The paper accounts for an unexpected embedded present tense, which denotes a future time in French relative clauses. The matrix displays either the periphrastic future or the simple future. In both cases, one can arguably decompose the matrix into a present tense feature and a forward shifter. This move leads to an analysis of the morphology encountered in the relative clause as an instance of Sequence of tense.

Counterfactual biscuit conditionals: A temporal remoteness Approach

Hypothetical conditionals like If you are hungry, your stomach is growling and ‘biscuit’ conditionals like If you are hungry, there is pizza in the fridge have been analysed as sharing the same syntactic and semantic template, differing only in the presence of an additional pragmatic inference leading to the ‘biscuit’ effect in the latter case (Franke 2009: a.o.). However, when considering their counterfactual versions, the two forms differ in the verbal morphological make-up of the consequent clause, which posits a challenge to the unified approach. The present papers develops an analysis of tense and mood morphology within the unified approach where the key idea is that counterfactuals biscuits involve breaking Sequence of Tense and so-called Sequence of Mood. Unacceptable biscuit and hypothetical forms are ruled out via pragmatic competition between weaker and stronger forms and via the Gricean Principle of Manner.

Varia: On ‘speaker ghosting’ and the ‘sequence of tense and mood’ rule in the bulgarian language

The analysis of sentences taking part in the semantico-syntactic ‘X said that [content of that]’ schema, in the second part of which verb forms in all nine Bulgarian tenses are used, demonstrates that five of the types of sentences obtained are non-grammatical, and four are grammatical. In the main cases, with the aorist and the imperfect, which are witness-forms, non-grammaticality is due to speaker ghosting, a phenomenon which has been revealed by the author in previous publications. Non-grammaticality with the future in the past and the pluperfect is due to the fact that the verb forms have non-cancelable content, and not because they are witness-forms, as claimed by some authors. The main conclusion of the study, in contrast to previous conceptions in Bulgarian grammars and in Bulgarian linguistics in general, is that the Bulgarian language has a ‘sequence of tense and mood’ rule.

Towards a New Explanation of Sequence of Tense

Past-under-past embeddings have two readings, a simultaneous and a backward-shifted one. While existing accounts derive these readings via distinct mechanisms, be it by means of an ambiguity at the level of LF or via blocking of a cessation implicature, we propose an alternative account which avoids such ambiguity. For us, the meaning of a past tense morpheme, like -ed, is comprised of two components. Syntactically, every past tense morpheme carries an uninterpretable past feature [uPAST], to be checked by a (single) covert past tense operator Op- PAST carrying an interpretable feature [iPAST]. Semantically, the past tense marker encodes a relative non-future with respect to its closest c-commanding tense node (informally: ‘not later than’), immediately yielding the two distinct readings.

Sequence of optional tense

In English, sentences containing a past tense in the complement clause of a past-marked propositional attitude predicate (e.g, John said Mary was sick) are ambiguous between a simultaneous and a back-shifted reading, in a phenomenon known as Sequence of Tense. In languages like Japanese, only a shifted reading is available for such past-under-past sentences. Two families of theories have been proposed in the literature to account for this variation: structural and pragmatic. Structural accounts rely on a syntactic rule or licensing condition to derive simultaneous readings of embedded clauses. Pragmatic accounts rely on competition between past and present tense in embedded clauses to derive the readings. In this paper, I provide new data from Washo, an optional tense language, to weigh in on these theories. In Washo, both tensed and (past-oriented) tenseless embedded clauses can have simultaneous and back-shifted readings. I argue that structural approaches can account for the Washo generalizations fairly straightforwardly, while pragmatic approaches encounter difficulties. The result is that the distribution of simultaneous readings cross-linguistically is more fruitfully viewed as a syntactic phenomenon rather than a pragmatic one.