Dorit Bar-On

The notion of expression has been put to many uses in philosophy, yet it has received surprisingly little direct theoretical attention. Drawing on certain distinctions as employed in earlier work, this chapter tries to show how, properly understood, the notion of expression can indeed help address standing puzzles in seemingly disparate areas: a puzzle about so-called first-person authority (in philosophy of mind and epistemology), a puzzle about the motivational character of ethical claims (in metaethics), and a puzzle about the origins of meaning (in the philosophy of language). In each case, this chapter argues that failure to draw crucial distinctions has led to undue dismissal of substantive expressivist proposals that could potentially resolve the relevant puzzles.

In analogy with external-world skepticism, content skepticism claims that I do not know the content of my present thoughts, given various possible alternatives supplied by content externalism. Yet, as common sense would have it, my belief that, say, I am presently thinking that there’s water in the glass, is much more secure than my belief that there’s a glass in front of me. This paper argues that the analogy relies on a recognitional conception of our ordinary knowledge of content. There are good reasons to reject this conception. The paper’s second aim is to sketch a neo-expressivist alternative, according to which ordinary self-ascriptions of contentful states are protected from skeptical doubt because they enjoy ascriptive immunity to error. What allows such self-ascriptions to represent secure knowledge of content is the fact that they involve the exercise of an expressive ability.

After presenting Boyle’s appeal to the Sartrean notion of nonpositional self-awareness in explaining Evans’ “transparency fact” concerning self-knowledge, I argue that his explanation suffers a certain instability. To the extent that nonpositional self-awareness is taken to be a matter of first-order ‘transparent’ orientation to the world, Boyle’s suggestion concerning the character of explicit positional self-knowledge is compromised. On the other hand, to the extent that nonpositional awareness is regarded as a form of genuine self-awareness, his explanation overintellectualizes first-order mental states. I conclude by raising questions regarding Boyle’s success in providing a viable alternative to epistemic accounts of basic self-knowledge.

When people avow their present feelings, sensations, thoughts, etc., they enjoy what may be called “first‐person privilege.” If I now said: “I have a headache,” or “I'm thinking about Venice,” I would be taken at my word: I would normally not be challenged. According to one prominent approach, this privilege is due to a special epistemic access we have to our own present states of mind. On an alternative, deflationary approach the privilege merely reflects a socio‐linguistic convention governing avowals. We reject both approaches. On our proposed account, a full explanation of the privilege must recognize avowals as expressive performances, which can be taken to reveal directly the subject's present mental condition. We are able to improve on special access accounts and deflationary accounts, as well as familiar expressive accounts, by explaining both the asymmetries and the continuities between avowals and other pronouncements, and by locating a genuine though non‐epistemic source for first‐person privilege.

Truth pluralism is a metaphysical theory of the nature of truth. The pluralist rejects the deflationist claim that truth is at best a ‘shallow’, insubstantial property. Indeed, the pluralist embraces a plurality of truth properties (such as correspondence, superassertibility, coherence), each appropriate to a different domain (or domains) of discourse. On the face of it, the pluralist will inherit all the main problems of the various traditional substantivist theories of truth. In addition, a strong pluralist, who only recognizes a plurality of truth properties (so that truth emerges not as one but only many), faces a number of problems, such as the problem of mixed discourse. But the moderate pluralism, who acknowledges that there is, in addition to diverse truth properties, a single property of truth (so that truth emerges as both one and many), faces problems of her own. After raising specific difficulties we see with moderate pluralism, we propose a less extravagant way to preserve the metaphysical intuitions that motivate pluralism.

Research on the evolution of language is often framed in terms of sharp discontinuities in syntax and semantics between animal communication systems and human language as we know them. According to the so-called “pragmatics-first” approach to the evolution of language, when trying to understand the origins of human language in animal communication, we should be focusing on potential pragmatic continuities. However, some proponents of this approach (e.g. Seyfarth and Cheney Animal Behavior 124: 339–346, 2017) find important pragmatic continuities, whereas others (e.g. Origgi and Sperber 2000) find sharp discontinuities. I begin (in Section 1) by arguing that this divergence is due to the fact that the proponents implicitly rely, respectively, on two different views of pragmatics, corresponding to different conceptions of what is involved in context-dependence – one “Carnapian”, the other “Gricean”. I argue that neither conception is fit to serve the purposes of pragmatics-first approaches to the evolution of language. In Section 2, I examine a recent formal “semantic-pragmatic” analysis of monkey calls, due to Philippe Schlenker et al. (in, e.g., Linguistics and Philosophy 37 (6): 439–501, 2014, Trends in Cognitive Science 20 (12): 894–904, 2016a, Theoretical Linguistics 42 (1–2): 1–90, 2016b), which appears to improve on the Carnapian and Gricean conceptions. However, I argue that the appearances are misleading and that the S-P analysis is no better suited than Carnapian analyses for the purposes of those seeking to establish human-nonhuman pragmatic continuities. Understanding why this is so will point the way toward my preferred, genuinely intermediate conception of pragmatics (as defended in Bar-On Biology & Philosophy 36 (6): 1–25, 2021), which – I argue in Section 3 – is better fit for these purposes. Drawing on recent discussions of chimpanzee communication, I briefly indicate which aspects of extant primate call communication – both gestural and vocal – could potentially count as pragmatic according to this conception.

Expressivism, broadly construed, is the view that the function of utterances in a given area of discourse is to give expression to our sentiments or other (non-cognitive) mental states or attitudes, rather than report or describe some range of facts. This view naturally seems an attractive option wherever it is suspected that there may not be a domain of facts for the given discourse to be describing. Familiarly, to avoid commitment to ethical facts, the ethical expressivist suggests that ethical utterances (e.g., “Gratuitous torture is wrong”, “What John did was morally good”) do not serve to ascribe ethical properties to objects, actions, persons, or states of affairs. Instead, they simply function to give voice to certain of our sentiments (or ‘pro/con’ attitudes). Along similar lines, philosophers have entertained versions of expressivism about the aesthetic, the modal, the mental, what is funny, even about theoretical science and knowledge ascriptions.

Ordinarily, if a person produces a nonreflective, ‘unstudied’ self-attribution of a present mental state – an avowal – we do not presume that they have produced the avowal on some specific epistemic basis; and we do not expect them to know how they know the self-attribution to be true. This no-‘how’ character of basic self-knowledge is puzzling, given that we regard avowals as manifesting factual, and indeed privileged, knowledge. I am here interested in views that accommodate both the baseless, no-‘how’ and the factual, privileged character of basic self-knowledge. I argue that leading constitutivist views, which embrace both, fail properly to meet the doxastic (as opposed to justificatory) requirement on basic self-knowledge, thereby failing to preserve its genuinely factual character. I then argue that an alternative, neo-expressivist approach is better placed to meet the doxastic requirement, as well as being at least as well-placed as constitutivist views to address the justificatory requirement on baseless self-knowledge.

Suppose, familiarly, that you and a friend have landed in an alien territory, amidst people who speak a language you do not know. Upon seeing you, one of them starts yelling, seemingly alarmed. You say to your friend, “She thinks we want to hurt her. She's scared. We must seem very strange to her.” Your friend, who is facing you, says, “No, I think she's actually trying to warn you: there's a snake right above your head, on that tree. You see the sling in her hand? I think she's going to try to shoot it down.“On a prevalent view, much discussed in recent years, you and your friend have engaged in a mini-theoretical enterprise. Using certain observations of the alien's behaviour as your data, and deploying certain generalizations and principles concerning human behaviour, you advance hypotheses regarding the internal psychological states which issued in her observed behaviour.

Recent discussions of animal communication and the evolution of language have advocated adopting a ‘pragmatics-first’ approach, according to which “a more productive framework” for primate communication research should be “pragmatics, the field of linguistics that examines the role of context in shaping the meaning of linguistic utterances”. After distinguishing two different conceptions of pragmatics that advocates of the pragmatics-first approach have implicitly relied on, I argue that neither conception adequately serves the purposes of pragmatics-first approaches to the origins of human linguistic communication. My main aim in this paper is to motivate–and begin to articulate–an intermediary conception whose scope is narrower than Carnapian pragmatics but broader than Gricean pragmatics. To do so, I first spell out what I take to be the key insight offered by proponents of the Gricean approach concerning the emergence of linguistic communication, namely, its being communication ‘from a psychological point of view’. I then develop this insight using key elements from the anti-Gricean ‘biosemantic’ account of linguistic communication due to Ruth Millikan, Philosophical Perspectives 9, Ridgeview Publishing, Atascedero CA, 1995, Millikan R Varieties of Meaning. Mass.: The MIT Press, Cambridge, Millikan, Beyond concepts: unicepts, language, and natural information, Oxford University Press, Oxford UK, 2017, and elsewhere). I argue that the intermediary pragmatics-first approach that I propose, which draws on both Gricean and Millikanian resources, would be better equipped to serve the purposes of those who search for potential precursors of human linguistic communication in animal communication.

According to ethical neo-expressivism, all declarative sentences, including those used to make ethical claims, have propositions as their semantic contents, and acts of making an ethical claim are properly said to express mental states, which (if motivational internalism is correct) are intimately connected to motivation. This raises two important questions: (i) The traditional reason for denying that ethical sentences express propositions is that these were thought to determine ways the world could be, so unless we provide an analysis of ethical terms in terms of natural properties, are we not committed to the Moorean conclusion that ethical sentences describe sui generis “nonnatural” ways the world could be? (ii) If we reject the claim that motivational attitudes constitute any part of the semantics of ethical sentences, are we not committed to denying the intuition behind motivational internalism after all? This chapter argues for negative answers to both of these questions.

Philosophers are often interested in explaining significant contrasts between ordinary descriptive discourses, on the one hand, and discourses – such as ethics, mathematics, or mentalistic discourse – that are thought to be more problematic in various ways. But certain strategies for ‘saving the differences’ can make it too difficult to preserve notable similarities across discourses. My own preference is for strategies that ‘save the differences’ without sacrificing logico-semantic continuities or committing to deflationism about truth, but also without embracing either truth-pluralism or global expressivism.I motivate my preference by examining, as a test case, mentalistic discourse. I begin by reconstructing three philosophical puzzles that have led philosophers to think of mentalistic discourse as problematic (Section 2). These puzzles concern the semantic, epistemological, and metaphysical status of contrasts between first-person present-tense attributions – ‘avowals’ – and all other ordinary contingent attributions. I then briefly present my own,neo-expressiviststrategy for addressing the puzzles (Section 3). Unlike traditional ‘simple expressivism’ (which is the analogue in the mentalistic realm of ethical emotivism), neo-expressivism is not committed to avowals’ being non-truth-apt or having non-propositional meanings. And yet it does not require embracing either deflationism about truth or global expressivism. It preserves continuities between mentalistic and other discourses while allowing us to capture discontinuities. Moreover, it is possible to apply the neo-expressivist framework in other areas where the notion of expression is deemed explanatorily useful, as illustrated by consideringethical neo-expressivism(Section 4). In the final section (5), I make more general comments on truth and meaning and tease out some of the commitments of the approach I advocate.

So-called basic self-knowledge (ordinary knowledge of one's present states of mind) can be seen as both 'baseless' and privileged. The spontaneous self-beliefs we have when we avow our states of mind do not appear to be formed on any particular epistemic basis (whether intro-or extro-spective). Nonetheless, on some views, these self-beliefs constitute instances of (privileged) knowledge. We are here interested in views on which true mental self-beliefs have internalist epistemic warrant that false ones lack. Such views are committed to a form of disjunctivism about basic self-knowledge. We begin by presenting an influential disjunctivist view about perceptual knowledge (Pritchard 2008, 2012, and elsewhere) and articulate a problem for it. We then consider two versions of disjunctivism about basic self-knowledge – one 'constitutivist', the other 'neo-expressivist' – and argue that both can avoid an analogue of this problem for self-knowledge. However, we give reasons for preferring the disjunctivism yielded by neo-expressivism. We conclude by considering briefly whether an acceptable disjunctivism about mental self-beliefs can point the way toward a sensible disjunctivism about perceptual beliefs.

After offering a characterization of what unites versions of ‘expressivism’, we highlight a number of dimensions along which expressivist views should be distinguished. We then separate four theses often associated with expressivism – a positive expressivist thesis, a positive constitutivist thesis, a negative ontological thesis, and a negative semantic thesis – and describe how traditional expressivists have attempted to incorporate them. We argue that expressivism in its traditional form may be fatally flawed, but that expressivists nonetheless have the resources for preserving what is essential to their view. These resources comprise a re‐configuring of expressivism, the result of which is the view we call ‘neo‐expressivism’. After illustrating how the neo‐expressivist model works in the case of avowals and ethical claims, we explain how it avoids the problems of traditional expressivism.