SWIF Philosophy of Mind, 28 March 2002. http://www.swif.uniba.it/lei/mind/forums/hutto2.htm

Forums Forum 1 Praetorious's commentary

Have You Heard Any Good Transcendental Arguments Lately?

Reply to Praetorious.

Daniel D. Hutto

Praetorious favourably comments on the central, negative aspect of my project in Beyond Physicalism, but expresses scepticism about the first and final sections in which I offer proposals about how we might, respectively, ‘dissolve’ the phenomenological and metaphysical problems of consciousness. Although she maintains that ‘strong arguments’ could be mounted against the positions set out in both of these sections, in the reply published here she restricts herself to criticising the very idea of nonconceptual experience, as discussed in part I of the book. Before replying to this in some detail, it is worth correcting a remark she makes at the outset, if only for the sake of potential damage limitation. She regards my "…assumptions about the nature of consciousness…[to be] part of a proposal, elaborated in Part III, for an alternative unified metaphysics…". This is strictly untrue. Part I and Part III may appear to be more closely linked than they are because in advocating a modified version of Absolute Idealism as better basis for metaphysics than physicalism, I endorse the idea that, "reality…must be analogous to non-conceptual experience in being non-discursive" (BP, p. 158). Nevertheless, the arguments for thinking this about reality are entirely independent of those for thinking that we ought to recognise the ultimately nonconceptual character of experience. This being so the arguments for these two sections do not stand or fall together.

With this matter aside, I can now concentrate on Praetorious’s worries about the project of the first section. She raises concerns about my apparent attempt, "…to show, by means of examples, that …basic non-conceptual [experiences] do indeed exist". She then spends some time trying to establish that none of the examples given ‘makes clear what precisely makes non-conceptual experiences distinctly nonconceptual". I am happy to concede this latter point without contest. Yet, I never hoped to demonstrate the nonconceptual character of experience by a straight appeal to examples. For within the space of a few paragraphs after introducing these, I also consider interpretations that higher-order theorists would put on such cases. This is because my only ambition in discussing them was to use them as ‘intuition pumps’ in order to establish minimally that, "…non-conceptualism about experience is at least speculatively plausible" (BP, p. 18). Nor were examples my only weapons in making that case. Like many others, I appeal to the importance of recognising a nonconceptual aspect to experience because of its potential in enabling us to understand such phenomena as: perceptual illusions, aspect-dawning, mental imagery and the fined-grainedness of experience. Two things are worth noting at this juncture. Firstly, I do not think any of these considerations gives us decisive reason to endorse non-conceptualism. Secondly, in defending the speculative plausibility of the notion of nonconceptual experience it is not necessary to give positive examples of the phenomena in question or, more strongly still, an account of what ‘makes’ experience distinctively nonconceptual. Minimally, a negative conception of the nonconceptual, e.g. as ‘not conceptual’, is required. If the nature of concepts is clearly determined then this will provide a means of demarcating the non-conceptual, even in the absence of a positive account of the latter (I set out my positive understanding of this general divide as it concerns intentionality in Chapters 4 and 5 of The Presence of Mind).

Still, even bearing in mind my limited purpose in introducing examples, Praetorious thinks that the very idea of nonconceptual experience is so confused and incoherent that any appeal to our intuitions could serve to make it plausible. She argues that it is nonsense to imagine that non-conceptual experiences (which she calls ‘apprehensions’) could have the particular qualitative features they are alleged to have since in every case these must be distinguished by conceptual means. In making this case she conflates two things – what we can say about such experiences and what characteristics they can have per se. Thus she asks: "What would an experience of the smell of a rose be like without roses?" This is either meant to be a question about whether experiences can be separated from their objects or a question about whether they can occur in the absence of conceptual identifications. With respect to the first reading the reply is straightforward. If experiences are type-identified by appeal to their objects then experiences of roses cannot occur in the absence of roses, since they would necessarily be experiences of something else. With respect to the second reading, I completely agree that we cannot conceptually identify ‘the-smell-of-a-rose’ without employing the concept ‘rose’. Yet it does not follow that we could not enjoy such smells or distinguish the-smell-of-a-rose from the-smell-of-a-lilac, even if we lacked such concepts. That they smell different to us (or any creature) could be seen in our reactions to them. The hidden assumption in Praetorious’ argument is that experiential qualities must be non-distinct if not conceptual identified. But this simply begs the question against the non-conceptualist. At this point, it is worth asking: Apart from providing a means of indexing our experiences, in what sense does a conceptual identification describe their qualitative character? Conceptual identifications are useful for many purposes, but other than providing an easy means of identifying the qualities in question, they do not give us direct insight into their character.

Given that I hold it is normal practice for us to give descriptions of our experiences, this brings us to yet another of her serious concerns. She asks, "…how non-conceptual experiences could ever be indexed and conceptually schematised or accounted for at all, unless it is assumed that the conceptual schemes and categories used for such accounts matched with those of the experiences being accounted for" (emphasis mine). Thus, to take the simplest sort of case, how can I accurately describe my experience by employing the predicate ‘red’ if my seeing red is essentially nonconceptual? It is worth considering just what this critique amounts to before addressing the question of whether and in what sense accurate description requires ‘matching’. Consider that shortly after making this remark, Praetorious claims, "…this is not of course to deny that, generally, it is perfectly possible to conceptually schematise or linguistically to account for phenomena which are themselves nonconceptual or non-discursive". For example, she tells us that bread rolls and red patches are rightly to be regarded as nonconceptual. Given this, it is possible to raise exactly the same concerns about how such things can be both conceptually identified and yet regarded as nonconceptual. We might puzzle over how a conceptual description could ever apply, in the sense of ‘matching’, to something nonconceptual at all? Why should it be easier to learn to talk of and describe things than it is for us to do so with respect to experiences? I argued in Chapter four of Beyond Physicalism that, as long as we rightly observe the distinction between first- and second-order statements, descriptions of both kinds involve the very same activity (BP, p. 132). For example, when we describe features of a scene, we are at the same time describing to the way we see it, highlighting that which we find important or interesting to our listener. Such descriptions presuppose that the other has the same facility with language that we have and, as such, that they understand the basic points of reference. My interest in nonconceptual experiences is predicated on a concern with what underlies such basic understanding. Consequently, how we come to establish such essential reference points for language can be no more or less puzzling than how a contentful ‘schematisation’ would be achieved so as to enable us to conceptualise phenomena (unless it is pre-supposed, à la Kant, to be already given). In this context, I see no philosophical advantage in drawing a distinction between the subjective and the objective. The problem, if it is one, applies just the same on either construal. I claim that our conceptual grasp of things, and our capacity to describe them, is in part based on our peculiar way of experiencing them, which in the order of our cognitive development pre-dates our capacity to draw a subjective-objective distinction. Had we been creatures of a different sort – with different kinds of sensory apparatuses and different basic experiences – our concepts of such things would have developed differently (presupposing that we were intentionally directed at the very same items).

It is important, however, that in noting these humble – but necessary – grounds for conceptual development I am not in any way committed to the idea that our nonconceptual experiences determine the nature of our concepts. In my view concepts are holistic and should be identified with the inferential connections they accrue in correct usage. The fact that our experiential concepts are grounded on nonconceptual responses does not place heavy restrictive constraints on their development. To understand the concept of ‘green’ is to be able to draw a number of important conclusions about it in relation to a whole set of other concepts – which includes not only other colours, but many other things as well. What should not be ignored is that part of its identity depends on non-inferential and non-conceptual capacities and abilities to discriminate, that we share with others of our kind. Therefore, I do not have a direct answer to Praetorius’ question, "…how we could ever develop any observational concepts which did match with any of our experiences of colours..?". I do not expect anything like a perfect match or isomorphism between nonconceptual experiences and our mature concepts. For me, there is an important degree of independence between them, even though the latter builds upon the former, through training and teaching. In insisting that we recognise the existence of nonconceptual experiences, I have sought only to say what has to be in place to make such a meeting of minds, upon which conceptual development rests, possible in the first place.

Of course, none of this will seem possible if one presupposes that the only way demonstratively to index something is by means of conceptual identification of some kind. Minimally, the thought is that to identify something presupposes that we must be able to identify it as something of this or that kind. I, however, do not hold this view. It is crucial to my account that we can be intentionally directed at objects without the aid of concepts (although care should be taken not to confuse experiences with objects). I defended this position in The Presence of Mind by enlisting a modest biosemantics, designed to show how just this was possible. The learner quite literally has, initially, ‘no idea’ or ‘no concept’ of what ‘greenness’ is in the first stages of the process of acquiring such a concept. Hence, despite my use of intuition pumps, I do not wish to claim that we can properly imagine or recover this state of mind – nor is advocating any such attempt an aim of my project.

This brings me to my strongest reason for wishing to defend non-conceptualism. It is not possible coherently to deny the commitment to the nonconceptual without presupposing that our concepts are somehow innate or a priori. Since she denies the consequent, it is not surprising that in discussing this very point Praetorious introduces a curious distinction between ‘linguistic’ concepts and non-linguistic concepts. Likewise in her book, Principles of Cognition, Language and Action, she talks freely of, "…the knowledge and concepts involved in our non-linguistic determination and identification of things" (Praetorious, 2000: 155). Yet, since she rightly denies the possibility of a Fodorian language of thought and the causal-computational accounts of content, it is unclear how we are to understand such ‘non-linguistic concepts’. However, if she wishes to eschew non-conceptualism I agree that something of this sort is needed in order to play the role of enabling the kind of agreement between people that makes conceptual learning and development possible. Our disagreement is about what fills this role. As I said above, unless we presuppose the existence of at least some innate concepts I see no alternative but to hold that concepts develop from nonconceptual capacities of some kind. This is a simple point of logic. Hence, it would seem that by appealing to non-linguistic concepts Praetorious is postulating some kind of ‘conceptual knowledge’ that comes prior to that learnt in the social context. If so, we are owed more details about the nature and origin of this special class of concepts and form of knowledge. If not, it is unclear how it differs from that provided by our regular linguistic concepts. The situation she, or anyone who adopts this line, faces is more paradoxical and embarrassing than that faced by the advocates of non-conceptualism. In the primitive learning situation, when a child is learning its first concepts, how can we say - prior to its mastery of them - that the teacher and the student agree in how they identify things? What would it be to say that they agree in their deployment of non-linguistic concepts or ‘rudimentary notions’? On the other hand, unless this is so, we might wonder how is such learning possible at all? Isn’t this kind of common ground presupposed in the way we co-ordinate the very activity of teaching concepts? I suggest that the common ground is nonconceptual and that, normally, we perform the conceptual identifications and charitably extend these to learners as placeholders for the purposes of describing their responses, even though we know that they are – as yet – unable to make such identifications themselves.

This brings me back to another of Praetorious’ complaints against my defence of non-conceptualism. She warns against any attempt to seek to understand how our conceptual capacities can be, "…reduced to, deduced from or explained in terms of processes and states which are more fundamental than, and hence do not presuppose the existence of referentiality and truth…". Yet, here again, we are agreed. My postulation of nonconceptual experiences is not a consequence of such explanatory ambitions. My only aim is to outline the necessary conditions for conceptual development. In effect, I am interested in developing a kind of transcendental argument. Thus, my remarks concerning the need to provide an explanation of the existence of experiences, which she quotes from page 5 of the Précis, are wholly directed at the commitments of physicalists and can only be understood in the context of my polemic against them. So taken in that context, she is precisely right to question whether there is any explanatory payoff to postulating nonconceptual experiences. However, as a philosopher, I will settle for understanding, not explanation.

Finally, what of the charge that any attempt to recognise that there is a non-conceptual core to experience is necessarily to talk ‘nonsense’ on the grounds that it is impossible to account, "…for such more fundamental processes and states without describing them, and therefore without implying the existence of referentiality and truth" (emphasis mine)? I have nowhere tried to describe the nonconceptual qualities of experiences, per impossible, without using our ordinary concepts. That would indeed be a nonsensical activity. But it is not nonsense to conceptually describe how things appear to us (and thereby to index our experiences). Nor is it nonsense to suggest, on the basis of arguments such as those sketched above, that there is a need to recognise that experiences are nonconceptual at root and that making nonconceptual discriminations is a condition for being able to learn to use concepts and give descriptions at all.

References:

Hutto, D.1999. The Presence of Mind. Philadelphia/Amsterdam: John Benjamins.

Praetorious, N. 2000. Principles of Cognition, Language and Action. Amesterdam: Kluwer.

© 2002 Daniel Hutto