In English, sentences ascribing probabIIities may be classed according to whether they have (or permit paraphrase in) the syntactical form Pg, where P is a sentential operator, containing the probabilistic reference, end q is a non-probabilistic sentence. ‘The1·e’s a good chance of snow tonight’ does so paraphrase, for example, but ‘Green snakes are probably harmIess’ rloesrft. It seems to me that this syntactical distinction has considerable si gnilicance: scntcnccs of this operator form correspond to single cass applications ot` probability. Briefly, this is bccausc a single case probabilistic judgment, inter aiia, is one which could iorrn the imnied.iate basis of a choice of betting odds; and any bet is a bet that q, for some sentence q. This explains my terminology: I use the term SP sentence (statement, utterance, etc.) for a sentence (statement, utterance, etc.} of this form} More importantly, it motivates an enquiry as to the semantic role of a probabilistic sentential operator — the characteristic Feature of the meaning of an SP sentence. For it is widely held that the major objection to Frequency theories of probability is their inability to make sense of the single case. An appealing, if not popular, strategy is hence to retain frequencies where they work best, in ‘the general case’; while seeking another, but compatible, account of the single case use of probability (i.e. of the meaning of SP sentences). However, my present aim is not to defend frequencies, but to criticise a particular account of SP utterances: the view that such utterances have a...