The utterance of a speaker depends on two directions of fit. The first can be characterized as a mind-to-world-relation. Here, the speaker has the intention of conveying some states of affairs about the actual world. This direction of fit implies that the speaker wants to share some epistemic notions. If she is successful in doing so, the hearer will gain a true belief. This class of utterances is descriptive.
The second is a world-to-mind direction of fit of the speaker’s utterance. Here the speaker wants to convey some of her desires to the hearer, who acts in a particular way in order to fulfill the speaker’s desire. The intention is fulfilled if the hearer gains a new desire to act in order to fulfill the speaker’s desire (Sperber & Wilson 1986). This kind of direction of fit is unimportant for the following account. Here I shall focus on descriptive utterances.
Before I address some implications of epistemic intentions, I shall focus on the separability thesis. This thesis addresses the problem of an asymmetry of interests between hearer and speaker. Since the interests are not identical, the speaker could deceive the hearer. And the other way around: the hearer could distrust the speaker even though she utters a true sentence. Sperber et al. (2010) claim that some amount of distrust is a stabilizator in the evolution of human communication, which they call epistemic vigilance. Imagine that humans believed almost everything they were told. Since not every speaker has the propensity to speak the truth, hearers would have a lower amount of knowledge because they would have no tool for distinguishing a reliable testimony from a non-reliable one. Communication would be very imprecise, because knowledge agency would be less successful. Hence, epistemic vigilance is a pre-condition for cooperative communication, because both speaker and hearer check the reliability of knowledge transition. I agree with Sperber et al. (2010) that epistemic vigilance is a feature of a source and the content of information.[3]
Let me go back to the mind-to-world-relation of fit. If the twofold account given by Sperber & Wilson (1986) is correct, the hearer has gained a true belief (about some state of affairs). To count as knowledge, we have to ask whether the true belief is justified.[4] What could count as a justification? Some social epistemologists would say that the testimony of the speaker is sufficient to count as a justification. This is the thesis of an anti-reductionism in social epistemology which contains the claim that the speaker must not rely on other sources of knowledge such as perception, inference or memory to justify her belief (Coady 1992). A reductionist would say that the testimony cannot count as knowledge without relying in addition upon other sources of knowledge (Fricker 1995).
In Millikan’s (2005) account of an intentional conventional sign (which is the content of an intentional mental representation), she assumes that speaker and hearer are cooperative devices that have co-evolved. The relationship between sender and receiver can be characterized as beneficial in the long term. Millikan proposes a framework, in which the descriptive representations describe a mind-world-direction, whereas the directive representation describes the world-to mind relation. The long term beneficial communicative agency between sender and receiver characterizes a function of reproduction of conventional signs. The direct proper function in this particular case is that the hearer gains a new belief. Millikan (1984) is well known for this teleosemantic account that can deal with misrepresentations. In such a case, the proper function remains unfulfilled. It is unfulfilled if the speaker fails to cause a new belief in the hearer (Millikan 1984, 2005). But the hearer could also be responsible for the unfulfilled proper function if she mistakenly judges the speaker to be untrustworthy. The hearer is also a constitutive part of the cooperative devices that establish the direct proper function (Millikan 1984, 2005).