Pseudo-Imperatives Michael Franke Abstract: Pseudo-imperatives are compound sentences where an imperative sentence is followed by 'and' or 'or' and a declarative sentence. Schematically, pseudo-imperatives are of the form: an imperative I + 'and' | 'or' + a declarative sentence D Following Schwager's (2004) terminology, I will refer to pseudo-imperatives with a conjunction as IaDs and to pseudo-imperatives with a disjunction as IoDs. It is already a matter of interest that imperative sentences and declarative sentences can be grammatically combined in this way. The case becomes even more interesting if we look at the meaning of pseudo-imperatives. My basic concern will therefore be the particular meaning asymmetries of the following paradigm where it is assumed that being killed is unconditionally undesirable and being kissed is unconditionally desirable for the addressee and that this is common knowledge among the interlocutors: (1) a. Close the window and I will kill you. b. Close the window and I will kiss you. c. Close the window or I will kill you. d. ? Close the window or I will kiss you. What meanings are associated with these sentences? - All of the examples in (1) are surely associated with some kind of conditional assertive force. For instance, (1a) and (1c) do not announce an unavoidable murder, and (1b) commits the speaker to a kiss conditional only upon the addressee's closing the window. In particular, the declarative sentence of an IaD makes an assertion only about those situations in which the content referred to by the imperative sentence holds. In contrast to that, the declarative sentence of an IoD makes an assertion only about those situations in which the content referred to by the imperative sentence does not hold. Under the assumption that the addressee does not want to be killed, sentences (1a) and (1c) are, or contain, conditional threats. Under the assumption that the addressee wants to be kissed, sentences (1b) and (1d) are, or contain, conditional promises. Not only do all the sentences in (1) have a particular assertive force, they also have a particular directive force associated with them. This directive force coincides with or differs from the directive force associated with the plain imperative 'Close the window'. In (1b) and (1c) we interpret positively, i.e. the pseudoimperative as a whole has the same directive impact as the plain imperative form which it contains. Borrowing terminology from Clark (1993), I will speak of positive IaDs or POS-ANDs to refer to examples like (1b) and of positive IoDs or POS-ORs to refer to examples like (1c). In contrast to positive interpretations, (1a) is interpreted negatively, since an utterance of (1a) directs the opposite of the plain imperative form which it contains. I will speak of negative IaDs or NEG-ANDs to refer to examples like (1a). It is important to see that the directive force associated with the utterance of a pseudo-imperative depends on the desirability of the declarative sentence which it contains. Finally, we note that (1d) is pragmatically odd, if being kissed is desirable for the addressee. (1a) conjoins an undesirable proposition to yield a negative interpretation, (1b) conjoins a desirable proposition to yield a positive interpretation and (1c) disjoins an undesirable proposition to yield a positive interpretation. So, from symmetry one might expect that in (1d) where a desirable proposition is disjoined we would obtain a negative interpretation. But this is clearly not the case. Also a positive interpretation is odd. In fact, it is not possible to maintain that being kissed is desirable for the hearer and to make sense of an utterance of (1d) in either way. This observation can safely be generalized. Examples which should be NEGORs for reasons of symmetry are pragmatically infelicitous. In particular, there are no negatively interpreted IoDs. Although there are NEG-ANDs, there are no NEG-ORs. This is the main problem which is going to be addressed in this thesis. I will refer to the basic explanandum of this thesis as NEG-OR Problem. Upon closer look, the NEG-OR Problem comprises two aspects which are to be distinguished. For one, we ask why there are no negatively interpreted IoDs. I will speak loosely here and say that the impossibility or non-existence of NEG-ORs has to be accounted for. For another, we ask why IoDs with a positively connoted second disjunct are pragmatically odd. I will say that the pragmatic infelicity of NEGORs has to be accounted for, and what is meant here is that forms which should for reasons of symmetry be negatively interpreted IoDs are pragmatically infelicitous. In a nutshell the problem to be dealt with in the following is this: NEG-OR Problem: The basic task in connection with pseudo-imperatives is to explain (i) why there are no negatively interpreted IoDs and (ii) why IoDs with a positively connoted second disjunct are pragmatically infelicitous. Unfortunately, the literature is not unanimous about what the best description of our intuitions about the meanings of pseudo-imperatives should be. In section 2 I therefore argue extensively for such a description. In particular, I argue that we should acknowledge the full asymmetry in the pattern (1) and distinguish IaDs and IoDs more than we should subsume them under the label pseudo-imperatives. I argue in section 2 that IaDs are assertions of conditionals and that IoDs are speech act conjunctions. It will transpire that if this description of the meanings of pseudo-imperatives is correct we have thereby solved the NEG-OR Problem in both its aspects already at the end of section 2. The critical challenge will be to find further justification for the suggested view on pseudo-imperatives. That means that especially an adequate explanation for the meaning contribution of natural language conjunction 'and' and natural language disjunction 'or' has to be provided. This I try to give in section 3. The basic question to be addressed in section 3 is how it is possible for 'and' and 'or' in pseudo-imperatives to yield the intuitive meanings for which I have argued in section 2.