Intentionality and the Myths of the Given

Between Pragmatism and Phenomenology

Carl B Sachs

• Publisher: Routledge
• Year: 2015
• Print ISBN: 9781848935075
• eBook ISBN: 9781317317586
• Edition: 1
• Topic: Philosophy
• Subtopic: Philosophy History & Theory

Intentionality is one of the central problems of modern philosophy. How can a thought, action or belief be about something? Sachs draws on the work of Wilfrid Sellars, C I Lewis and Maurice Merleau-Ponty to build a new theory of intentionality that solves many of the problems faced by traditional conceptions.

1 Intentionality and the Problem of Transcendental Friction

DOI: 10.4324/9781315653945-2

Original Intentionality and the Naturalist Challenge

In his insightful (and delightful) ‘Te Intentionality All-Stars’, John Haugeland examines the problem of how there can be original intentionality within a naturalistic world-view. How, he asks, can any part of the natural world bear an intentional relation to any other part? ‘How can there be norms among the atoms in the void?’.¹ I use ‘original intentionality’ here in Haugeland's sense: ‘original’ contrasts with ‘derived’, with the kind of intentional, semantic content that sentences and signs have. Since not all semantic content can be derived, the argument goes, there must be original intentionality. Put otherwise, original intentionality does not mean a particular kind of intentionality but rather ground-level, ‘original’ cases of intentionality that need to be understood in order for other cases of intentional content to be understood.

This is not to say that any particular item has its own intrinsic semantic content independent of all other semantic contents – that, I will show below, is an episode of the Myth of the semantic Given. Rather, it is to say that the entire relational system of semantic contents does not have its content conferred upon it by something else; intentional content is a holistic property of the system, not of any specific part of the system. Nor does “original” here mean ineffable, private, inexplicable, or unexplained. Like Haugeland, I am interested in explaining the place of original intentionality within the broader framework of a naturalistic world-view (subject to caveats about ‘natural’). In that sense, what is original under one set of considerations (the order of understanding) can be a result or product when viewed naturalistically (the order of being). The question is, how to regard original intentionality as something other than a Lever of Archimedes, how to explicate what original intentionality is, and how to explain the place of original intentionality in rerum natura.

On Haugeland's analysis, those who are interested in explaining original intentionality can be divided into three camps, which he calls neo-Cartesianism, neo-behaviourism, and neo-pragmatism. The neo-Cartesians (e.g., Jerry Fodor) seek to explain intentionality in terms of the pattern of relations amongst internal mental states, whereas the neo-behaviourists and neo-pragmatists broaden the scope. The neo-behaviourists (e.g., Willard V. O. Quine, Daniel Dennett) focus on ascriptions of intentional contents by virtue of patterns of interactions between organisms and their environments, particularly in perception and action. By contrast, the neo-pragmatists (e.g., Wilfrid Sellars, Robert Brandom) explicate intentionality in terms of social norms. All three positions purport to account for original intentionality, and to that extent are rivals. But Haugeland allows a position between neo-behaviourism and neo-pragmatism, and more interestingly, for the possibility of being both a neo-behaviourist and a neo-pragmatist:

we might imagine a neo-behaviorist and a neo-pragmatist agreeing that animals and people share a certain primitive sort of intentionality (second-base position), and yet also that a qualitatively ‘higher’ intentionality is possible only for conformists with a culture and language (third-base position). Questions could be raised, of course, about what the two sorts of intentionality have to do with one another – why, in particular, both are sorts of intentionality – but perhaps there would be enough similarities to justify the common term.²

The position taken up here answers Haugeland's challenge by showing just how one can be both a neo-behaviourist and neo-pragmatist about original intentionality by distinguishing between the different kinds of intentionality. The key difference between Haugeland's suggestion and my developed view is that I draw extensively on Maurice Merleau-Ponty's first-person-standpoint descriptions of perception and action, rather than on the third-person-standpoint of the neo-behaviourists. The benefits of construing one kind of original intentionality as both non-social and non-linguistic will be made clear.

Hence, my approach is what I call bifurcated intentionality: that there are two basically different kinds of intentionality, both of which are ‘original’ in the relevant sense. The first is what I call discursive intentionality: the kind of intentionality that we use to characterize the ‘aboutness’ or ‘of-ness’ of thoughts, beliefs, desires, and more generally, anything with propositional content. The second is what I call somatic intentionality: the kind of intentionality that we use to characterize the lived bodily engagements with and comportments towards the world as enactively perceived and practically grasped. I aim to show that both discursive and somatic intentionality must be considered as equally original with regard to the philosophical work they are called upon to perform, because discursive intentionality (the intentionality of propositional discourse) and somatic intentionality (the intentionality of embodied perception) are individually necessary and jointly sufficient for judgements with empirical content:as empirical content or objective purport. To motivate this view, I shall show just why it is that we need both discursive and somatic intentionality, how to remove the obstacles that have prevented us from accommodating both kinds, what further distinctions are needed in order to render the view intellectually satisfactory, and what further consequences this view has for philosophizing about mindedness and experience.

The Naturalistic Challenge to Intentionality

This work is animated by a commitment to what I call transcendental naturalism: the view that transcendentally-specified roles must have empirically-specifiable role-players.³ I aim to use ‘transcendental’ in a minimal sense, avoiding transcendental arguments or substantive metaphysical doctrines with ‘transcendental’ in the name. By ‘transcendental’, I mean the activity of identifying those items in a basic inventory of our cognitive capacities: what conceptual and perceptual roles must be filled in order for a cognitive system to count as having a mind similar to the kinds of minds that we have, and thereby to have the kinds of experiences that we have (however loosely and contextually characterized is the ‘we’ in this usage). Within this loose construal, philosophical methods as different as Lewis's ‘reflective analysis’, Sellars's comparison of the manifest and scientific images, and Merleau-Ponty's phenomenological descriptions of embodied perception will all count as different implementations of transcendental philosophy.

But transcendental philosophy is not, I think, sufficient: we can, on the other hand, interpret the insights and discoveries of the natural (and social) sciences as providing specifications of the role-players for various transcendentally-specified roles. So I take ‘intentionality’ to be a transcendental notion, rather than a straightforwardly empirical one – but with the caveat that the theory of intentionality that we arrive at should be, at the end of the day, at the very least fully consistent with the natural sciences– it should not commit us to positing the existence of any entities that are not required for any other explanatory purposes. Rather than hastily attempt to ‘naturalize intentionality’, I want to inquire more carefully into what sort of thing intentionality is, such that it could be naturalized (or perhaps not). To see the value of doing so, consider the following line of thought:

1. The concept of original intentionality requires that the basic units of thought are sentence- and term-shaped representations;
2. If original intentionality is real, then it must be empirically identifiable;
3. Hence the most plausible candidate for being the bearer of original intentionality is the primate (perhaps even human) brain;
4. But cognitive neuroscience shows that brains do not represent their environments in sentence-shaped representations – there are no propositions in the prefrontal cortex;
5. So, there is no such thing as original intentionality.

As Alex Rosenberg confidently summarizes this view:

The illusion of original intentionality has its origin in the fact that while the brain stores information in nonpropositional data structures of some kind, it extracts and deploys the information in temporally extended processes, such as noises and marks – eventually speech and writing – and it is these together with the conscious states that they result in that generate the illusion of propositional content.⁴

I am uncertain what to say about (5) save that I find this conclusion so ‘counterintuitive’ that I would be hard-pressed to locate a common ground with those who think that (5) is correct. Yet I find no problem with either (2) or (4); the problem lies entirely with (1) and (3). Against (1), I urge a distinction between two different kinds of intentionality, discursive and somatic, so that the identification of intentional content with propositional content does not go unquestioned. Against (2), the two basic kinds of original intentionality are ‘at home’ in linguistic communities and in embodied animals, respectively. This allows that intentionality may be ‘naturalized’, with the crucial proviso that we do not have at present a sufficiently fine-grained comprehension of the intricate causal interplay between brains, bodies, and environments that would be needed to fully specify how discursive and somatic intentionality are causally instantiated. (Whether it is possible or probable that we may eventually have such a comprehension is beyond the scope of the present investigation). In other words, Rosenberg-style eliminativism about intentionality does not follow from neuroscience alone; it follows from neuroscience conjoined with a specific picture of what intentionality must be like. It is that picture which I aim to dislodge by offering a series of minor corrections to those who have preceded me in this enterprise.

One constraint on the account of intentionality to be developed here is that it must avoid what is called, following Sellars, ‘The Myth of the Given’. Yet here at once a whole net of problems arise as to what is supposedly Given, just why the Given is supposedly a Myth, how the Myth of the Given should be avoided, and whether our commitment to avoiding the Myth entails that we ought eschew ‘nonconceptual content’. The interpretation of the Myth of the Given is compounded by according undue emphasis to Sellars's ‘Empiricism and the Philosophy of Mind’, at the expense of the comprehensiveness of Sellars’ philosophy as a whole, even though Sellars himself insists at the outset:

This framework [the Given] has been a common feature of most of the major systems of philosophy, including, to use a Kantian turn of phrase, both ‘dogmatic rationalism’ and ‘skeptical empiricism’. It has, indeed, been so pervasive that few, if any, philosophers have been altogether free of it; certainly not Kant, and, I would argue, not even Hegel, that great foe of ‘immediacy’. Often what is attacked under its name are only specific varieties of ‘given.’ Intuited first principles and synthetic necessary connections were the first to come under attack. And many who today attack ‘the whole idea of givenness’ – and they are an increasing number – are really only attacking sense data. For they transfer to other items, say physical objects or relations of appearing, the characteristic features of the ‘given.’ If, however, I begin my argument with an attack on sense-datum theories, it is only as a first step in a general critique of the entire framework of givenness.⁵

Accordingly, I want to be bear in mind ‘a general critique of the entire framework of givenness’ through the current project. Our understanding of the Myth has been hampered in two different ways. Firstly, critics of the Myth (and even a few defenders) have emphasized the empiricist version of the Myth at the expense of rationalistic, Kantian, and even Hegelian versions.⁶ These versions of the Myth differ according to what is said to be given: sense-data, universals, real connections, the typology and structure of mental representations, and so on. That is, the Myth has a much wider scope than usually understood. Secondly, the Myth has been construed as merely an epistemological thesis rather than, as I argue here, also a cognitive-semantic thesis. While this is understandable, it matters because Lewis (as I shall argue in Chapter 2) does not commit the Myth if the Myth is just an epistemological mistake – nor, importantly, does Sellars himself think that Lewis’ error was an epistemological one. Rather, I will argue, Sellars’ criticisms of Lewis’ notion of ‘the given’ is not epistemological, but cognitive-semantic – a kind of reflective inquiry closely related to what Hegel called ‘logic’ and to what Wittgenstein called ‘grammar’.

In its most general form, the Myth of the Given is the idea that a fully adequate cognitive semantics can identify cognitive-semantic contents simply as such, independent of the role(s) that they play in perception, thought, and action. Thus construed, the Myth of the Given constrains any account we can construct of what is necessary for any being to count as sapient (to be able to play the game of giving and asking for reasons) as well as sapient-and-sentient (to be able to play the game of giving and asking for reasons with regard to perception and action). The traditional defender of the Given holds that the Given is not a Myth by arguing that there are contents of cognitive experience that do not require awareness of any other cognitive-semantic contents in order to be known as those contents. Those who hold that the Given is a Myth argue that all cognitive access to cognitive-semantic contents is itself mediated by their role in perception, thought, and action; consequently, identification of anything as Given either covertly depends on our ability to deploy cognitive-semantic contents generally, or that the purported contents do not actually play any cognitive-semantic role. One central problem, for those who think that the Given is a Myth after all, is whether avoiding the Myth requires adjuring from any commitment to ‘nonconceptual content’, much-vexed notion as that is. If so, we shall need to know under what specifications nonconceptual content would be consistent with that criticism, and under what specifications the commitment to nonconceptual content would be an instance of the Myth.

The challenge thereby posed is this: how can one accept nonconceptual content without committing the Myth? Is it coherent that nonconceptual content could be a kind of intentional content, or must all intentional content be conceptually structured? The concept of nonconceptual content in recent analytic philosophy of perception has a complicated history.⁷ Consequently, several scholars have urged that the very debate between conceptualism and nonconceptualism rests on undiagnosed equivocations. Though I agree with that line of thought, I am unsure that the present diagnoses of those equivocations have gotten it entirely right. To see why, consider the following supremely condensed summary of the debate based on recent contributions by Richard Heck, Jeff Speaks, T. M. Crowther, and Josefa Toribio.⁸

Heck argues that the key idea behind ‘nonconceptual content’ is that there is a kind of mental content that is representational but nonconceptual by virtue of failing to satisfy Gareth Evans's Generality Constraint: that a subject cannot conceive of a is F if she cannot also entertain the thought that a is G and that b is F. However, Speaks and Crowther both argue that this is an insufficient characterization of nonconceptual content based on the arguments brought forth by both nonconceptualists (Peacocke, Evans, Heck, Kelly) and conceptualists (McDowell). Based on both Speaks and Crowther, Toribio puts the contrast between ‘content nonconceptualism’ and ‘state nonconceptualism’ – capturing both Speaks's distinction between ‘absolute nonconceptual content’ and ‘relative nonconceptual content’ and also Crowther's distinction between ‘compositional content’ and ‘possessional content’. According to Toribio, content nonconceptualism holds that ‘for any perceptual experience E with content C, C is nonconceptual_c if C is essentially different in kind to the content of belief’,⁹ whereas state nonconceptualism holds that ‘for any perceptual experience E with content C, any subject S and any time t, E is nonconceptual_s if it is not the case that in order for S to undergo E, S must possess at t the concepts that a correct characterization of C would involve’.¹⁰ One note of caution: since Crowther distinguishes between compositional and possessional conceptualism and conceptualism, one could hold both compositional nonconceptualism and possessional conceptualism or both compositional conceptualism and possessional nonconceptualism.

In these terms, I take ‘nonconceptual content’ to be personal-level representational cognitive-semantic content that does not conform to the Generality Constraint. Each of these points deserves separate elucidation: NCC is personal-level, not a feature of subpersonal neurocomputational processing; it is re...

Table of contents

• Cover Page
• Half Title Page
• Series Page
• Title Page
• Copyright Page
• Table of Contents
• Acknowledgements
• Introduction: Why a New Account of Intentionality?
• 1 Intentionality and the Problem of Transcendental Friction
• 2 The Epistemic Given and the Semantic Given in C. I. Lewis
• 3 Discursive Intentionality and ‘Nonconceptual Content’ in Sellars
• 4 The Retreat from Nonconceptualism: Discourse and Experience in Brandom and McDowell
• 5 Somatic Intentionality and Habitual Normativity in Merleau-Ponty’s Account of Lived Embodiment
• 6 The Possibilities and Problems of Bifurcated Intentionality
• Conclusion
• Appendix: Is Phenomenology Committed to the Myth of the Given?
• Works Cited
• Notes
• Index