This dissertation is devoted to the temporal interpretation of fronting constructions in English and the phenomenon of the Sequence of Tense. It provides and analyzes previously unobserved data from verb phrase fronting constructions in which the simultaneous interpretation of a present tense embedded under a matrix past tense is available. These data are theoretically unexpected and challenging because most theories of English tense disallow this interpretation for Present-under-Past configurations. An account that captures the new data is proposed. It establishes a connection between the simultaneous interpretation of Present-under-Past and the mode of semantic composition between a verb and its complement. It is proposed that the simultaneous Present-under-Past is licensed in a VP-fronting configurations when the direct object of a transitive verb or a CP-complement of an attitude verb compose with their verbs by the rule known as Restrict (Chung & Ladusaw 2004). According to this rule, a verb’s complement acts semantically as a modifier and does not saturate the verb’s internal argument position. In the process of formulating this proposal, a number of other results are obtained. A version of a theory of temporal de re (Abusch 1994) is stated in terms of temporal concept generators (building on Percus & Sauerland 2003). A case is made for the existence of the simultaneous de re Present-under-Past, which is captured in terms of the proposed concept generator-based analysis. I also propose a novel theory of how tense features are transmitted in Sequence of Tense phenomena. Finally, several consequences of the proposed approach to tense are investigated. Among them are the temporal interpretation of CP- and DP-fronting constructions, the predictions for Past-under-Past fronting configurations and the predictive potential of temporal de re attitude reports in comparison with temporal de se reports.