Eva Schmidt

In this article, I argue against Kearns and Star’s reasons-as-evidence view, which identifies normative reasons to ɸ with evidence that one ought to ɸ. I provide a new counterexample to their view, the student case, which involves an inference to the best explanation from means to end or, more generally, from a derivative to a more foundational “ought” proposition. It shows that evidence that one ought to act a certain way is not in all cases a reason so to act. I present a diagnosis of the problem that is brought out by the counterexample.

We argue that, to be trustworthy, Computa- tional Intelligence (CI) has to do what it is entrusted to do for permissible reasons and to be able to give rationalizing explanations of its behavior which are accurate and gras- pable. We support this claim by drawing par- allels with trustworthy human persons, and we show what difference this makes in a hypo- thetical CI hiring system. Finally, we point out two challenges for trustworthy CI and sketch a mechanism which could be used to gener- ate sufficiently accurate as well as graspable rationalizing explanations for CI behavior.

In this chapter, I rebut three incarnations of the epistemological objection put forth by McDowell (Mind and World, Harvard University Press, Cambridge, MA, 1994a) and Brewer (Perception and Reason, Clarendon Press, Oxford, 1999). According to them, only the assumption that perceptual experiences have conceptual content can account for the fact that perception plays a crucial role in justifying belief about the external world. I begin by providing some context to the objections, viz. by presenting the myth of the given that threatens the foundationalist (Sellars, 1:253–329, 1956). Next, I respond to the objection from the logical space of reasons. I argue that nonconceptual content and nonconceptual mental states are themselves elements of the normative space of reasons and thus fit to enter normative justificatory relations with belief. In response to the second objection, the objection from the inferential nature of justification, I argue that nonconceptual and non-propositional scenario contents can be involved in justification, for experiences with such contents can be the starting points of content-sensitive correctness-truth transitions that have beliefs with Fregean propositional contents as their endpoints. In this Modest Nonconceptualist account of perceptual justification, I rely on the external content of perception and belief, which consists in the states of affairs represented. I defend the account against several conceptualist objections. In my defense against the third objection, an objection from cognitive access, I show that nonconceptual states and their contents can be the subject’s own reasons on a mentalism-inspired picture of perceptual justification. Modest Nonconceptualism can thus respect some of the internalist motivations of the conceptualist.

I defend both conceptualists and nonconceptualists against an attack which has been leveled at them by critics such as Byrne (Perception and conceptual content In: Steup M, Sosa E (eds) Contemporary debates in epistemology. Blackwell, Malden, pp 231-250, 2005), Speaks (Philos Rev 114:359–398, 2005), and Crowther (Erkenntnis 65:5–276, 2006). They distinguish a ‘state’ reading and a ‘content’ reading of ‘(non)conceptual’ and argue that many arguments on either side support only the respective state views, not the respective content views. To prepare the ground for my defense, I argue for an understanding of the state view in terms of concept exercise rather than concept possession and provide an overview of versions of conceptualism and nonconceptualism of different strengths. I then argue that conceptualists and nonconceptualists tacitly accept a so-called ‘state-to-content’ principle, show that existing defenses of this principle fail, and provide a new defense of it. It draws on the sources of the nonconceptualism debate, viz. the need to do justice both to the phenomenology of experience and to its epistemological role and to account for the existence of perceptual content and thought content. I argue that epistemological considerations together with considerations from the subject’s perspective support the claim that conceptual thought has conceptual and propositional content, whereas nonconceptual experience has nonconceptual and non-propositional content.

I examine two arguments for nonconceptualism from the phenomenal character of perceptual experience. The idea is that only the assumption that experience content is nonconceptual does justice to the phenomenology of experience. In particular, if experience content is conceptual, we cannot account for its finely grained representational content. The problem is that visual color experience makes differences between shades of a color that are much more fine-grained than our conceptual repertoire allows. Further, conceptualism is incompatible with the situation-dependence of perceptual content Kelly (Philos Phenomenol Res 62:601–608, 2001b). For instance, it is hard if not impossible to make room, in purely conceptual terms, for the difference between perceiving a certain shade of purple instantiated by a steel ball as compared to seeing it instantiated by a wool carpet. As to the argument from fineness of grain, I concede that the conceptualist’s demonstrative strategy against the argument is initially successful. However, it fails in the end because of problems with the phenomenal character of hallucination, which cannot be accounted for by appeal to only demonstrative concepts. As to the argument from situation-dependence, I point out that the conceptualist cannot convincingly account for the perceived presence, at the same time and in the same place, of situation-dependent and independent properties in experience.

In this chapter, I turn to the claim that we cannot speak of perceptual content unless we assume it is objective content. The conceptualist argues that only conceptual content can meet the requirement of being objective. I start out by presenting the objection from objectivity as it can be found in McDowell (Mind and world, Harvard University Press, Cambridge, 1994a). I then discuss the following replies: First, even if objective perceptual experience requires the perceiver to have an objective world-view, the experience’s own content may be nonconceptual; second, perceptual objectivity can be had in virtue of mere nonconceptual personal-level abilities; third, a weaker kind of perceptual objectivity that does not even require personal-level capacities is substantial enough to provide for genuine perceptual content. The last reply is the one championed by Modest Nonconceptualism. All that genuine perceptual content presupposes is that the world is perceptually presented to the subject. This requirement can be elucidated via a subpersonal account of how the perceptual systems generate representations underlying her experiences that are poised to influence her central behavior-guiding system.

This chapter provides an overview of the structure and purpose of the book. It introduces the philosophical context and motivations of the debate between conceptualism and nonconceptualism. The book is a defense of the nonconceptualist claim that experience is nonconceptual and has nonconceptual content. In particular, it defends what I call ‘Modest Nonconceptualism,’ which is briefly introduced in this chapter. On this view, all perceptual experiences are at least partly nonconceptual, i.e., involve the exercise of at least some concepts. It involves an argument that enables the Modest Nonconceptualist to bridge the gap between the state view and the content view of nonconceptualism. ‘Concept’ talk is taken to be anchored in conceptual abilities that the subject possesses and exercises. Nonconceptual perceptual content is taken to consist in scenario content (see Peacocke, A study of concepts. MIT, Cambridge, 1992), which is both nonconceptual and non-propositional; externally conceived, the content of an experience consists in the worldly states of affairs it represents. The latter content is needed for the Modest Nonconceptualist’s account of perceptual justification. The view claims that the Autonomy Thesis is correct: A perceiver’s experiences may have nonconceptual content even if she possesses no concepts whatsoever. The Modest Nonconceptualist account of the representational content of perceptual experience is based on the subpersonal-level organization of the underlying representational states.

The argument from contradictory contents presented here is based directly on observations about the content of experience. It claims that experience content, if conceptual, allows for contradictions within one and the same content. There are at least two examples of this, the waterfall illusion and the visual experiences of some grapheme-color synesthetes. However, due to a Fregean principle of content individuation, no conceptual contents are contradictory. So experience content is nonconceptual. I motivate a particular version of the argument and defend it against two central objections: First, the objection that there is more than one content involved in apparently self-contradictory experiences and second, the claim that conceptual contents are sometimes contradictory, as is true of some desires. In reaction to the second objection, I argue that we need an explanation of why subjects are not irrational if they do not revise their self-contradictory perceptual experiences, and that the conceptualist has no good account of this. For on his view, just as in thought, conceptual abilities are involved in experience and we are dealing with the same Fregean propositional contents. By contrast, nonconceptualism provides such an explanation, for no conceptual abilities are involved in experience itself, and scenario content is not the kind of content that can be revised under rational pressure.

In this chapter, I clarify the notions of mental content and of concept. I present competing views on these notions and indicate my own position. I introduce content in terms of correctness conditions and distinguish several kinds of propositions, as well as non-propositional scenario content, with which perceptual content might be identified. I relate this discussion to a wide-spread commitment in philosophy of perception to respect the subject’s perceptual perspective in ascriptions of perceptual content. Then I compare views of concepts as Fregean senses, as mental representations, and as cognitive abilities and investigate how they relate to the central idea that concepts are possessed by subjects. I suggest that our talk of concept possession and exercise is anchored in subjects’ abilities for re-identification and for general thought and in their inferential abilities. I clarify how possession and exercise of these three conceptual abilities relate.

The author defends nonconceptualism, the claim that perceptual experience is nonconceptual and has nonconceptual content. Continuing the heated and complex debate surrounding this topic over the past two decades, she offers a sustained defense of a novel version of the view, Modest Nonconceptualism, and provides a systematic overview of some of the central controversies in the debate. An explication of the notion of nonconceptual content and a distinction between nonconceptualist views of different strengths starts off the volume, then the author goes on to defend participants in the debate over nonconceptual content against the allegation that their failure to distinguish between a state view and a content view of (non)conceptualism leads to fatal problems for their views. Next, she makes a case for nonconceptualism by refining some of the central arguments for the view, such as the arguments from fineness of grain, from contradictory contents, from animal and infant perception, and from concept acquisition. Then, two central objections against nonconceptualism are rebutted in a novel way: the epistemological objection and the objection from objectivity. Modest Nonconceptualism allows for perceptual experiences to involve some conceptual elements. It emphasizes the relevance of concept employment for an understanding of conceptual and nonconceptual mental states and identifies the nonconceptual content of experience with scenario content. It insists on the possibility of genuine content-bearing perceptual experience without concept possession and is thus in line with the Autonomy Thesis. Finally, it includes an account of perceptual justification that relies on the external contents of experience and belief, yet is compatible with epistemological internalism.

I discuss an argument for non-conceptualism based on animal and infant per- ception. Crudely put, some animals and infants who possess no concepts nonetheless have perceptual states with non-conceptual content. Perceptual experiences of adult humans have the same kind of content as the experiences of animals and infants, so the content of the perceptual experiences of adult humans is also non-conceptual. I defend this argument against potential attacks from the conceptualist. I argue that there are indeed creatures which possess no concepts, but have perceptual experiences, and I attack McDowell’s view that we share perceptual sensitivity with animals and infants, but not genuine perceptual contents.

In this paper, I discuss the conceptualist claim that we cannot speak of perceptual content unless we assume it is objective content. The conceptualist argues that only conceptual content can meet the requirement of being objective, so that the view that perceptual experience has nonconceptual content is not tenable. I start out by presenting the argument from objectivity as it can be found in McDowell. I then present the following objections: First, perceptual objectivity cannot be due to the perceiver’s conception of objectivity; and second, even nonconceptual capacities of the individual cannot and need not be appealed to in order to account for objective perceptual content

This chapter contains the results of my discussion in the book. I first review the success of the respective arguments for nonconceptualism and defenses against conceptualist objections. Five of the six arguments presented in favor of nonconceptualism are successful, and three of them are strong enough to support the claim endorsed by Modest Nonconceptualism that every perceptual experience has at least some nonconceptual content. On the other hand, none of the conceptualist objections to nonconceptualism is fatal. So, nonconceptualism is victorious against conceptualism. Finally, I provide a brief sketch of the Modest Nonconceptualist view that has emerged.

In this chapter, I discuss arguments for the claim that a subject can both have an experience with a certain content and not be in possession of all the concepts needed to specify this content. If she does not possess all the relevant concepts, then she cannot exercise them. So, she can undergo such an experience without being required to exercise all the concepts needed to specify its content. The argument from memory experience goes back to Martin (Philos Rev 101:745763, 1992). Since we can extract new information from memories of previous experiences when we acquire new concepts, the content of these previous experiences cannot have been fully conceptual. The argument from animal and infant perception presupposes that some subjects who lack concepts of any kind nonetheless have perceptual experiences with the same kind of content as human perception. So, the content of human perception must be nonconceptual just like the perceptual contents of these subjects. The third argument, the argument from concept acquisition (Roskies, Philos Phenomenol Res 76:633659, 2008; Noûs 44:112134, 2010), shows that we cannot explain how subjects acquire some of their first concepts, particularly perceptual-demonstrative concepts, unless we assume that experience content is nonconceptual. The question of whether a subject can have a conscious perceptual experience only if she is able to cognitively appreciate its content is a recurrent theme in the chapter; it is answered in the negative by Modest Nonconceptualism.