The view that if-clauses function semantically as restrictors is widely regarded as the only candidate for a fully general account of conditionals. The standard implementation of this view assumes that, where no operator to be restricted is in sight, if-clauses restrict covert epistemic modals. Stipulating such modals, however, lacks independent motivation and leads to wrong empirical predictions. In this paper I provide a theory of conditionals on which if-clauses are uniformly interpreted as restrictors, but no covert modals are postulated. Epistemic if-clauses, like those in bare conditionals, restrict an information state parameter which is used to interpret an expressive layer of the language. I show that this theory yields an attractive account of bare and overtly modalized conditionals and solves various empirical problems for the standard view, while dispensing with its less plausible assumption.

认为if子句在语义上作为限制器起作用的观点被广泛认为是全面概括条件句的唯一候选者。该视图的标准实现假设，在看不到要限制的运算符的情况下，if子句限制隐性认知模态。然而，规定这样的模态缺乏独立的动机并导致错误的经验预测。在本文中，我提供了一个条件理论，在该理论上，if子句被统一解释为限制器，但没有假定隐式模态。认知如果-clauses，就像那些在裸条件中一样，限制用于解释语言表达层的信息状态参数。我表明，该理论对赤裸裸的、明显模态化的条件产生了有吸引力的解释，并解决了标准观点的各种经验问题，同时摒弃了其不太合理的假设。