سکوت

The Shape of a Good Revenge

A Critique of Fodor’s Non-Conceptualism

Alireza Kazemi

Abstract: Jerry Fodor’s recent defense of the non-conceptual content, in his The Revenge of the Given, deserves a specific attention. Fodor’s strategy has two main parts; a philosophical part and an empirical part. Fodor tries to find out the main property of the conceptual content by means of the philosophical deliberation and after that, he appeals to the empirical psychology to examine whether there is a kind of content that does not meet the mentioned requirement. Since his answer to the second question is affirmative, he happily announces that there is indeed such a thing as non-conceptual content. This article takes issue with the first part of the Fodor’s strategy in his defense of the non-conceptual content. It will be argued that although this strategy has no problem in its own right, the specific way that Fodor adopts in furthering his argument would result in nothing but begging the question against the proponents of conceptualism.

Keywords: Given, Non-Conceptual Content, Conceptualism

Introduction

The debate of non-conceptualism and conceptualism about content is one of the most discussed topics in the contemporary philosophy of mind. One of the reason is the important implications that this debate has for various philosophical problems, both inside and outside the philosophy of mind. These implications include problems such as the possibility of naturalizing the perception and semantics (Balog, 2009, p. 311) and the epistemic justification of perceptual judgements (McDowell, 1994). Empirically minded philosophers, who currently hold sway in the philosophy of mind (specifically, I have people like Fodor and Heck in mind) think that the final verdict in this debate should be issued from the side of cognitive sciences, not as a conclusion of a priori arguments. But fortunately, there is a consensus that

without the philosophical analysis and scrutiny about the nature and properties of the non-conceptual content, the discoveries of empirical sciences would be of no help because it would not be clear whether the empirical fact appealed to, supports the existence of non-conceptual content, properly understood, or merely another kind of the conceptual content.

Jerry Fodor’s The Revenge of the Given contains a roughly new case for the non-conceptual content. The novelties of Fodor’s approach are both in his interpretations of the empirical facts and in his characterizations of the conceptual content. The cogency of the empirical facts in supporting the existence of the non-conceptual content heavily depends on the characterizations previously attributed to the conceptual content by means of philosophical analyses. Of course, finding an exhaustive and exclusive characterization of the conceptual content, is as hard and contentious as the main debate of conceptualism/non-conceptualism itself. The issue at stake is that the characterizations attributed to the conceptual content should not at the outset beg the question against the opponents of the non-conceptual content. This requirement is what I call the shape of a good revenge. My main objective in this paper is to show that the specific approach that Fodor has in mind in taking the revenge of the non-conceptual content from the conceptualists, lacks this requirement and therefore, his strategy does not change the situation in favor of the conceptualism. Consequently, as a result of this problem in the philosophical part of his strategy, his empirical facts too do not support the existence of the non-conceptual content more than the existence of varieties of the conceptual content.

In the first section of this article, I discuss Fodor’s approach in defense of the non-conceptual content in more detail. Richard Heck’s explication of Fodor’s paper, which should be considered best as an extension of that argument, too is utilized in discussing this strategy. In the second section, the outline of McDowell’s account of the nature of the conceptual content and its differences from the non-conceptual content is proposed. This part would be the main material in criticizing Fodor’s characterization of the conceptual content. In the third section, it will be shown that in the light of McDowell’s characterizations of the conceptual content, a proponent of the conceptualism could argue that Fodor’s initial steps in his

approach does not really show the properties of the non-conceptual content, but only different manners of the presence of concepts in the mental content. I finish the paper with a short conclusion.

1. Fodor on non-conceptual content

What is it in virtue of which we call a content conceptual? What is the characteristic feature of the conceptual content? It is by having this characteristic feature that we can, consequently, characterize the non-conceptual content. Therefore, the first step in Fodor’s argument is what I call Asymmetry Principle (henceforth: AP). According to AP, in order to judge about the tenability of the duality of the content (conceptual/non-conceptual), we need an asymmetry, something that one side of the alleged dichotomy has, but the other side doesn’t. Fodor’s suggestion to fill up this role is compositionality. Discursive content, or propositional content, which he equates with the conceptual content, is compositional; it has parts that by a specific combination make the whole proposition. One of the derivative properties of compositionality, according to Fodor, is the existence of a canonical decomposition. There are many ways conceivable to decompose a proposition into its parts, but among them, there is usually a unique way of decomposition which is exactly the reverse algorithm by means of which the whole proposition has been specifically constructed by means of its parts. Compositionality is our best explanation for the miraculously productive nature of the language and discursive content. It is novel and different arrangements and re-arrangements of parts which can rapidly produce a bulk of novel meaningful propositions. But the process of composition and decomposition are not arbitrary. There are canonical ways of the formation and decomposition of the proposition from and into the parts. The parts are not all on a par. There are differences between each part and the way that it systematically contributes to the formation of the discursive content (Fodor, 2007, p. 108).

Now let have another look at these characterizations-compositionality and canonical decomposition, the conjunction of which is the Fodor’s suggestion for the best characterization of the conceptual content. Mental representations with these two characteristics, or conceptual content, have provided an immense possibility of representations of the external world for human beings. But are they the only forms of

representation that we have at our disposal in confronting the world? Fodor does not think so. Iconic representation, he thinks, is a kind of representation (hence content) which does not meet the mentioned requirement, i.e. compositionality & canonical decomposition. By having parts (or pixels), it could be said that iconic representations too are, somehow, compositional. But in contrast to propositions, these parts are all on a par; every part of an icon represents, in isolation, a part of the represented. By invoking the AP, Fodor concludes that there are contents which are not conceptual, hence conceptualism is refuted.

The other favorite empirical fact of Fodor is “the item effect” as a result of which cognitive engagement of the brain in processing, memorizing, and recollecting information from an image (or icon) is less burdensome than discursive contents. It is, according to Fodor, another evidence for the asymmetry principle; there is some kind of content (or representation) which is not conceptual. Richard Heck has suggested that cognitive maps- the geographical representations that animals and human beings have of their environment- could be another candidate of the non-conceptual content. An absent-minded driver who easily reaches his destination through an amazingly busy and sophisticated route, is utilizing his cognitive map of his environment. Pre-lingual infants and animals too seem to possess these cognitive maps of their environment in their bodily interaction with their world. I have no problem with making use of empirical evidence in clarifying and helping to solve philosophical puzzles. But there are many obscurities and possibility of different interpretations of these empiricals fact that make it useless in the debate in question. This objection would become clear in the third section.

2. The Many Faces of the Conceptual Content

John McDowell’s attack to the idea of the non-conceptual content, in his Mind and World, has been considered as one of the best arguments for the conceptualism (Heck, 2007, p. 118). McDowell’s main argument is epistemological in spirit. The argument is that unless we credit experience with the conceptual-content, there is no way to keep experience as a capacity for knowledge- internalistically conceived- without falling prey to the Sellarsian Myth of the Given. In this conception of experience, our perceptual judgments are justified, for a specific subject, because her experience immediately reveals

things to be as they are judged to be in the perceptual judgments. If experience wants to play this specific justificatory role for the subject, McDowell argues, there is no way except than considering it to be “conceptual all the way down”.

To understand why, we have to block the possible alternatives which easily come to mind. If we understand perceptual justification externally, as is common in the contemporary epistemology, there would be no problem. Experience causes, or brings about, perceptual beliefs about the situation of their elicitation. It is not experience, conceived internalistically from the eyes of the beholding subject, which entitles her to her perceptual judgments. Experience stands outside the realm of concepts and propositions, it is the result of the causal experiential confrontation, which are the observational judgments, which brings about epistemic disclosure about the layout of the external world. Another alternative, which is also appealing in its own right, is to hold that experience does provide a source of epistemic justification for perceptual judgments, even if we conceive the mentioned justification internalistically and from the eyes of the beholding subject. But there is no need to credit experience itself with conceptual or propositional content. Experience stands outside the realm of concepts and propositions but it, somehow, discloses the layout of the external world to the beholding subject.

The latter alternative is usually appealed to as a case for the non-conceptual content. The former does not in itself supports non-conceptual content, but it does urges us to devoid experience from the conceptual content. In its most extreme cases, this option could lead to something like Brandom’s stance: “experience is not one of my words” (Brandom, 2001, p. 205). McDowell thinks that Davidsonian coherentism too is in line with Brandom’s position. Now, what is wrong with the first alternative, which McDowell imputes to Evans, Travis and others? It is here that the charge of the Myth of the Given is invoked to put aside this alternative. Myth of the Given, roughly speaking, is that nothing outside the realm of normativity (realm of concepts, propositions, implications, commitments and entitlements) could justify the occurrence of an epistemic episode. In other words, there is no “side-way on view” or “cosmic exile” point of view, to link the causal occurrences outside the realm of knowledge to normative episodes in the realm of knowledge. If experience wants to confer an entitlement or justification to a

beholding subject to hold an empirical proposition, the experience could not be devoid of normative significance. There is no bridge which ties the dumb world of causal-nomological relations to the meaningful world of the normative implications. Sellars and McDowell think that forgetting this amounts to what Moore has called, in ethics, the naturalistic fallacy.

As I said, both candidates are appealing in their own rights. Brandomian, Davidsonian alternative escapes from the charge of the Myth of the Given by pushing the experience off the scene. It is only the resultant perceptual judgments, caused by the external world in appropriate situations which is the source of our perceptual judgments. Experience, in itself, does not provide a first-personal, internal, disclosure of the world which itself justifies the subject’s perceptual judgments, again, from the disclosure which is already at her disposal. The position of people like Evans and Travis, on the other hand, tries to save the intuition of the perception as an internal and first-personal capacity for knowledge. The problem of the former is its deflationary conception of experience, while the problem of the latter is its ignorance about the problems imposed by the Myth of the Given. This is, as McDowell metaphorically says, “a seesaw” between two unpalatable extremes. It is here, in response to this felt tension that McDowell proposes his case for the conceptualism. If the experience is “conceptual all the way down”, if the rational, normative capacities of human beings are in operation in experiential disclosure of the world, then experience would not lie outside the realm of reason and normativity, and therefore, there would be no charge of falling prey to the Myth of the Given. Still, unlike Brandom and Davidson, experience could be the disclosure of the external world to the beholding subject, from her first-personal and internal point of view.

The upshot of McDowell’s argument could be summarized as follows: conceptualism is the only candidate that tenably preserves the intuition of seeing experience as a capacity for knowledge-internalistically and first-personally conceived, while avoiding the problems incurred by accepting the Myth of the Given. Otherwise, the importation of epistemic significance form the experience to the perceptual judgements would be highly problematic. This is why McDowell’s argument is sometimes called the importation argument against the non-conceptual content (Heck, 2012, p. 117).As it should be clear by now, conceptualism for McDowell is the claim that upper-level rational normative capacities of

the human beings are, essentially and always, working on their experience, and that is how experience can disclose states of affairs to the human beings. Non-conceptualists, as Heck describes them, come in two strands; some of them, like Evans, claims that experience is totally devoid of the presence of concepts, or they holds, as Peacock does, that “there are always non-conceptual elements present in the contents of visual perception, but conceptual elements may also occur” (Heck, 2012, p.129). McDowell thinks that anything less than his proposal, cannot provide an epistemic and justificatory role and therefore is not entitled to be called a kind of content at all, because a blind content is not a kind of content at all. In other words, there is only one kind of content; conceptual content. Fodor, Heck and others, have usually focused exclusively on this part of McDowell’s line of thought; his importation argument for the conceptualism. But McDowell’s project has a further wing which is overlooked by most of the non-conceptualists and, I think, could block most of their arguments for the existence of non-conceptual content.

This overlooked aspect of McDowell’s project concerning experience is this: what is the mechanism and process through which higher order rational and normative faculties of the human beings work on the experience? Doesn’t McDowell’s position imply that experience is not different from propositional content? If not, how can McDowell show the differences between experience (conceived conceptually) and other paradigmatic cases of the conceptual contents (e.g., judgments and discursive capacities). Surprisingly, McDowell’s answers to these questions are not less discussed by himself, but they are usually ignored by the proponents of the non-conceptual content. It is one of the main claims of this article that taking these considerations into account undermines most of the arguments proposed for the existence of the non-conceptual content, especially those of Fodor and Heck. The upshot of considering this usually ignored aspect of McDowell’s conceptualism could be summarized in this slogan: there is only one kind of content- conceptual content- but there are indeed varieties of the conceptual content. In other words, there are various ways in which concepts take part in mental contents. More discussions should be anticipated in the next section, but as a promissory note, the claim is that this point heavily undermines what in the first section I called Fodor’s AP; finding an asymmetry between propositional content and other empirically established forms of content, and concluding that there is actually non-conceptual content.

Already in Mind and World, McDowell proposes an important distinction between experience and the discursive practice of judgment (even though both of them are conceptual for him). Experience is the involuntary actualizations of the conceptual capacities of the subject. The very same conceptual capacities that are voluntarily exercised in an act of judgment, are passively drawn into operation in experience (intuition, perception). This insistence on the involuntariness of the experience is utilized to explain the strong intuition of the passivity and determinateness of the content of experience. It seems that in experience, the normative freedom that is felt in undertaking a commitment or trying to justify a proposition, is not present and it is not because experience is a mere causal confrontation with the world (also called non-conceptual content!) but because the upper-level rational capacities of the subject are passively and involuntarily drawn into operation. Acquiring the capacity to intuit the world, being privileged to be in a position to be revealed and disclosed states of affairs in the external world, is something that we acquire by being initiated into the natural language (the realm of concepts). In other words, acquiring the natural language- or the capacity to work with the concepts- and intuiting the world through experience are co-constitutive- two sides of one coin. The alleged unidirectional mechanism, starting from a concept-free confrontation with the non-conceptual content which leads to acquiring concepts and the conceptual content, is a Myth because it is exactly an episode in what Sellars has castigated as the Myth of the Given. Moreover, now that we have been initiated into space of reason (the realm of concepts and the natural language), the disclosure of the world to us (through the involuntary extraction and excavation of our upper-level rational capacities) has become something like the Aristotelian second nature; a kind of entrenched habit acquired by a normative upbringing (McDowell, 1994, p. xx).

The second important asymmetry between experience and propositional content (which both of them are conceptual for McDowell) is a distinction which is absent in Mind and World, and McDowell has proposed in his later writings (esp. 2009). In Mind and World, McDowell seems to urge that the content present in the experience is actually propositional. But it is neither required for his theory to be successful, nor what he admits now. Experience (intuition, perception) is conceptual all the way down, but it is not

propositionally structured. A propositionally structured experience, is a representation-as not a mere representation. It directly reveals a particular state of affair to the subject. By denying that experience is propositional, McDowell has to tell us why he insists it to be still conceptual. The importation argument again works here. But we still need an account of how experience could be conceptual but not propositional. McDowell’s explanation is vague here. Therefore, I prefer to propose an explanation by myself which I think is compatible with McDowell’s project. Propositions and inferential relations are the elements of the natural language. But by being initiated into the space of reasons (space of implications) we human beings have acquired a great amount of conceptual capacities; deictic expressions (demonstratives)¹, proper names, quantifiers and etc. Robert Brandom has developed a comprehensive theory in the philosophy of language (semantic inferentialism) according to which these conceptual, seemingly isolated elements of the natural language are shown to be dependent upon and late-comer with respect to the inferentially-articulated content. In other words, although deictic expressions such as here, now, this, and that car, are not inferentially articulated (like ‘the car is red’), they are emerged as a result of mastering the game of giving and asking for reasons (game of implications and inferential relations which are propositional) (Brandom, 2001, pp. 123-57). By being able to work with propositional content, this seemingly conceptual capacities are emerged consequently. This theory helps us imagine what McDowell could say regarding the mentioned question; how can intuition be conceptual but not propositional.

McDowell also introduces another point of asymmetry for experience which is not important for my critique of Fodor (Cf. McDowell, 2009, p.259).

3. Fodor through the Lens of McDowell

At this juncture, it is easy to criticize Fodor’s strategy from the eyes of a conceptualist. The characteristic feature of the propositional content, does not exhaust all conceptual contents. Even though propositions and judgments are the paradigmatic cases of the conceptual content, they are not the whole

1. Specifically, Sellars (1967, pp. 1-31) and McDowell (2009) have insisted on the role of the demonstratives in characterizing the conceptual content present in the experience.

story. The specific way that concepts take part in propositions (compositionally & by forming a canonical decomposition), is not the only way of their presence. Even if we give priority to propositions and inferentially-articulated content (as Brandom does in his philosophy of language), there is no reason to believe that the characteristic feature of the propositional content exhausts the realm of conceptual content. There is nothing wrong with the AP in itself, but the specific asymmetry that Fodor suggests is nothing but begging the question against McDowell.

Now let turn to the empirical facts interpreted by Fodor and Heck as evidence for the non-conceptual content. The lighter cognitive burden impose by memorizing a phone list, with respect to memorizing, say, an image, does nothing to substantiate, or even support, the AP. Perceiving an image, discriminating its parts, or memorizing its general and specific features, could be essentially concept-involving even though it is less burdensome with respect to working with propositions and judgments. The reason, simple as it is, is that involuntariness that McDowell appeals to, and the second nature (entrenched habits brought about in subject by a normative upbringing) could easily explains the alleged easiness of working with icons and pictures². About the cognitive map of Heck the solution is different. We should distinguish between the absent driver and the pre-lingual infants. About the absent driver, the interpretation is not necessary or even the best one. Conceptual capacities of the subject is working, maybe unconsciously or through his access-consciousness, in helping him find her way in the streets. The complexity of the process of navigating could even suggest that we have to accept that propositional beliefs and knowledge of the driver is being, somehow, utilized in this situation. But about the case of animals and pre-lingual infants this road is blocked. Animal, devoid of conceptual capacities, can interestingly cope with the environment. Find their way back to their home after migration, escape from the predator but not a prey. This capacities seems mysterious without appealing to some kind of content or disclosure of the world through their perceptual faculties. Surprisingly, McDowell does not have any problem with these cases. It is the experience which has cognitive significance in justifying (internalistically) the perceptual judgments which should be already conceptualized. The importation argument deals with epistemic standards which

2. McDowell has pointed out to a similar comment in McDowell, 2009, p. 265

animals and pre-lingual infants obviously lack. Therefore, all that McDowell claims is that whatever mental mechanism which guides animal behaviors, it is not the same as epistemically significant perceptual disclosure of the world to human beings. More specifically, McDowell does not seem to deny even a totally externalistic account of justification for animals and pre-lingual infants (Cf. McDowell, 2016). All that McDowell claims and tries to substantiate, is that the experience which purports to ground our perceptual judgments in an internalistic way, should be “conceptual all the way down”.

In an approach which has something in common with Richard Heck’s cognitive maps, Carl Sachs (2014) has recently argued that what he calls somatic (i.e. bodily) intentionality can be the lacuna in bridging the conceptual content and the worldly states of affairs. Somatic intentionality, which is a different kind of content from semantic (propositional and conceptual) intentionality, is the bodily-sensorimotor engagement of sentient beings (including animals and human beings) in their environment. This engagement and aboutness , is motivational in characters. To be solicited or motivated by an object is “to enter into a set of actual and possible sensorimotor actions with regard to that object by virtue of ones innate and acquired sensorimotor abilities” (Noe, 2004, quoted from Sachs, 2015). This new kind of intentionality, Sachs argues, is forgotten by people like McDowell and even most non-conceptualists. Surprisingly, Sachs too insists on the non-decomposable character of somatic intentionality (Ibid, p. 208), unlike compositionality that we observe about semantic intentionality. The answer to this line of thought is not much different from what we said about Heck’s and Fodor’s. Empirical facts are interpretable and cannot substantiate a particular position in the debate. Philosophical deliberations too (non-decomposability and etc.) do not take into account the full range of varieties of the conceptual content, therefore AP is not established. Causal and natural relations which are absolutely necessary for mind’s confrontation with reality, and the sensorimotorial and motivational structures of sentient creatures are not underestimated by a conceptualist at all. The claim is that these necessary natural conditions, which seem to be wholly in the realm of cognitive sciences and evolutionary biology to discover and discuss, do not enter into the realm of mental content at all. If content has any epistemological significance in the lives of the rational agents, then according to the importation argument, it could not be totally devoid of

the concepts. Unconscious and natural mechanisms of sentient creatures, whether explainable through cognitive sciences or evolutionary biology, prove nothing against the conceptualist stance; there is no epistemically significant content for rational agents which is not already immersed in concepts.

The important morale is that empirical evidence is helpful in defending the existence of the non-conceptual content iff conceptualism is totally hopeless in making sense of the mentioned empirical evidence. If explaining some empirical facts aren’t mysterious or problematic from a conceptualist point of view, therefore it cannot give more weight to the non-conceptualist stance.

Conclusion

Working with AP is, in my view, a good strategy in investigating whether there is such a thing as non-conceptual content. But the main argument of this paper has been that we should be more careful about finding an asymmetry between conceptual content and other empirically established kinds of content. There are some subtleties, varieties and complexities in the realm of conceptual content which should be taken into account in working along these lines. Moreover, a tenable piece of evidence for the non-conceptual content should take care of these complexities, varieties and subtleties of the conceptual contents too. Empirical psychology is a great source of help in the conceptualism/non-conceptualism debate, but only if we clearly know what we are looking for among empirical evidence. In this way, I hope that this article not only showed the weakness of a specific approach in defending the non-conceptual content, it also explicated the shape of a good strategy in refuting the conceptualism.

References:

1. Balog, K. (2009). Jerry Fodor on non-conceptual content. Synthese 170: 311-320.

2. Brandom, R. (2001). Articulating Reasons: An Introduction to Inferentialism. London, England: Harvard University Press.

3. Fodor, J. (2007). The Revenge of the Given. In B. P. McLaughlin & J. Cohen (Eds.), Contemporary Debates in Philosophy of Mind. Oxford, UK: Blackwell Publishing.