<p> This thesis is an attempt to defend the somewhat outrageous view that the only things we need consider to be real are experiences. A serious objection to this view is that we seem to talk of things other than experiences, and that such talk is apparently unavoidable. But in Chapter One it is argued that if non-experiential talk can be interpreted in what is called a 'non-representational' way, then the apparent reference to non-experiential entities can be considered illusory. In Chapter Two a method of doing this is outlined. The key concept here is that of the 'acceptability of a sentence relative to a set of experiences.' Roughly, it is proposed that a semantic theory be a system of rules that determines this property for any assertoric sentence and any set of experiences. In Chapter Three the concept of truth is discussed in terms of the ideas already introduced. It is argued that we should recognize two kinds of truth - what are called 'redundancy truth' and 'empirical truth'. Empirical truth is more inportant for our purposes, but we have to recognize that there may be some sentences that are neither empirically true nor empirically false. Although this might be thought to necessitate a revision of classical bgic, it is argued that this is in fact not so.</p> <p> The last chapter contains an evaluation of the traditional empiricist conception of experience and attempts to defend the following: the epistemological priority of experiences; the withholding of ontological status from physical things; and the rejection of solipsism. There is also a comparison (though only a limited one) between the present theory and the verificationisrn of M.A.E. Dummett.</p> / Thesis / Master of Arts (MA)