One of the main findings of Jacob’s paper is a detailed elaboration of the differences between Millikan’s (1984, 2004, 2005) communicative agency and the Gricean (Grice 1957, 1969, Sperber & Wilson 1986) account of speaker’s meaning and intention. Jacob argues that the Gricean mindreading thesis, the separability thesis, and the ostensive nature of communication are not supported by Millikan’s account of the direct perception of speaker’s intention, which supports a non-inferential model of the understanding of intentional signs. Furthermore, the Gricean account is incompatible with Millikan’s claim that speaker and hearer are cooperative devices; the claim that the prediction of another’s behavior could be explained through reliance on socially established conformities and conventions and that modern developmental psychology could get along without any theory of mind.
A very influential account of naturalizing the content of intentional mental representations is Millikan’s teleosemantic framework (Jacob 2010). According to this view, the content of intentional mental representations can best be naturalized by relying on the history of the biologically-selected functions of these representations, namely the direct proper functions (cf. Millikan 1984, 1989). Interestingly, Jacob focuses on Millikan’s concept of communicative agency, which is strongly connected to the teleosemantic framework, and argues that there are several aspects of the Gricean individualistic account of meaning that are more plausible than Millikan’s when it comes to explaining social communicative agency. The most illuminating finding of Jacob’s paper is the modern and precise presentation of the actuality of the Gricean separability thesis and the mindreading account, because it is explanatorily fruitful not only for philosophy of language, but also for social cognition, social epistemology, informal logics, and the relation between these different studies.
I generally agree with Jacob’s main findings, nevertheless I will address some further issues of Millikan and Grice’s account with regard to philosophical problems in social epistemology. Before I respond to these in detail, I first focus on the Gricean account and its implications for social epistemology.
The well-known Gricean (Grice 1957) account of the meaning of an utterance focuses on analysis of the speaker’s meaning in a conversation. A speaker S means something unnatural if she intends something by the utterance of a sentence.[1] Let us suppose that the speaker is a politician with a specific agenda and with a propensity for aggressive propaganda. She utters the following:
(1) Speaker: “Our party will ensure that taxes go down”.
The speaker S means (1) iff S utters (1) with the intention that a hearer H will gain the belief that S’ party will make sure that taxes go down, if (i) the hearer H recognizes the speaker’s intention (1) and (ii) because of that she gains the belief that S’ party will make sure that taxes go down, (iii) since she recognizes that the speaker’s intention is exactly that (cf. Grice 1957, 1969).
Since Gricean meaning is individualistic and subjective it is important to note which underlying cognitive states constitute this meaning. As Jacob puts it—relying on Sperber & Wilson’s (1986) interpretation of the Gricean account[2]—there are three main assumptions upon which the psychological theory of meaning is based, namely the separability thesis, the mindreading thesis, and the asymmetry between an informative and communicative intention. Together with Jacob I shall focus on Sperber and Wilson’s account, which argues that the Gricean theory can be summarized as a reciprocal process of intentions. The informative intention is an intention of a speaker who wants to inform a hearer about some state of affairs. In order to be successful, the speaker has also the intention that the hearer recognize the informative intention (Grice 1957). This means that at first the hearer has to understand the informative intention. If she understands it, the communicative intention of the speaker has been fulfilled. But the informative intention will only be fulfilled, if the speaker is trustworthy: a necessary condition for accepting a speaker’s utterance. In effect, the hearer gains a new belief. The assessment of her trustworthiness depends on the hearer of that intention (Sperber & Wilson 1986). She must admit that the speaker has to be reliable in order to be trustworthy, or, to put it in Jacob’s (this collection, p. 4) words, “the addressee must further accept the speaker’s epistemic or practical authority”. But a question arises: on which kind of epistemic practices does the hearer have to rely in order to accept the speaker as an epistemic authority? I will address this question in the following commentary.
In order to answer it, I will argue that (i) a hearer could commit a local ad hominem attack, a process of inductive Bayesian reasoning that secures epistemic vigilance. Roughly, an ad hominem attack is an argument that considers rather personal properties of an utterer than the argument itself. (ii) A fallacious kind of the personal attack is the global ad hominem attack, which undermines every testimony of a speaker because of its personal traits. (iii) The Gricean account of mindreading could better account for an inductive inference model than Millikan’s direct-perception-account. (iv) Practices of ad hominem attacks, I will argue, support the Gricean separability thesis, while Millikan’s cooperative devices account is less plausible.
The structure of this commentary will be as follows: first, I focus on Grice and Millikan’s framework and its implications for social epistemology, namely the problem of epistemic reliability (cf. section 2). In section 3 I shall present Lackey’s account of a social epistemological dualism, a hybrid theory in which Lackey tries to connect the most plausible findings of social epistemological reductionism and anti-reductionism. Then I argue that the Gricean account of informative intentions and Lackey’s positive reason component could lead to the personal attack or ad hominem argument (cf. section 4). In section 5 I argue that there are two possible commitments of ad hominem arguments, to be specific, the global and the local ad hominem attack (cf. section 5.1, section 5.2). In the Gricean account of the mindreading thesis is compatible with the drawn picture of our social epistemological practices, because it supports the inductive inference model, while Millikan’s account of direct perception could not account for this. The Gricean separability thesis fits nicely with the positive reasons component and the reliance on ad hominem arguments, while Millikan’s account of speaker and hearer as cooperative devices is less plausible (cf. section 6).