[ previous | contents | next ] |
Conceptual idealism
Twentieth Century Idealism and the Possibility of Metaphysics
We are now in a position to describe and discuss the doctrine of ‘conceptual idealism’. In so doing I want to characterise a certain prevalent approach to philosophy as being roughly Kantian, or at any rate post-Kantian, in interesting respects. Not that every aspect of transcendental idealism has an echo in conceptual idealism, or vice versa. It is rather that the latter can be seen as an heir to Kant’s philosophy. More precisely, conceptual idealism is an heir to the question ‘How is metaphysics possible?’ and to the ‘Copernican revolution’ that formed the basis of Kant’s answer.
The assumption that Kant questions at the beginning of his critical philosophy is ‘that our knowledge must conform to the objects’. Since nothing, Kant claimed, had been achieved on this assumption, it is worthwhile pursuing the path that starts with its converse: that objects must conform to our knowledge[1]. Given the linguistic-analytic turn of the twentieth century, this starting point becomes: ‘things must conform to our concepts’. This is the central tenant of conceptual idealism.
We are in immediate danger of several misunderstandings. Most importantly, the conceptual idealist shares Kant’s aversion to any form of empirical idealism, such as Berkeley’s immaterialism. Indeed, so averse are most conceptual idealists to any philosophy that implies that things are dependent on actually being perceived or thought about, that they eschew the label of ‘idealism’ completely. While this hides the debt that their philosophies owe to Kant, the move is understandable. Witness the difficulty that Kant himself had distinguishing his idealism from that of Berkeley’s. And since the demise of both German and British ‘Absolute Idealism’, the label has fallen on even harder times. Nevertheless, we will call those philosophies ‘idealist’ just in case the ‘fundamental idea of idealism’ (which we formulated in chapter one) can be discerned in them. A doctrine will be called idealist just in case it presupposes that reality is somehow constrained by (or corresponds with) our ability to conceive it. A doctrine will be awarded the title of ‘conceptual idealism’ just in case it maintains that reality is somehow constrained by our publicly shared concepts. Different forms of this idealism can be differentiated by the sense they purport to give to the terms ‘reality’ and ‘concept’.
A second point to note about conceptual idealism is that it is not committed to the claim that objects are necessarily perceivable as Kant’s idealism of time and space implies[2]. The claim is that things are necessarily conceivable, or better still, describable. The latter way of putting the matter has the advantage of making it clear that what is at stake here is not the ability of some individual at some point in time to be able to conceive of a thing (a claim of a kind with empirical idealism). The claim is that things must necessarily fall under some possible concept. How the phrase ‘possible concept’ is cashed out will determine how credible a particular version of conceptual idealism is.
We can illustrate these issues by discussing a version of conceptual idealism that has been presented in an admirably clear and explicit form: the ‘Conceptualism’ of Michael Morris[3]. This is the conjunction of two (related) claims:
(A) There can be interesting metaphysical explanations.
(B) The nature of the objects, properties, and facts to which our concepts correspond is not fixed independently of the nature of the concepts which correspond to them.[4]
What does the phrase ‘not fixed independently of’ mean here? Morris rephrases (B) as ‘there would not have been those objects, properties, and facts, if they had not corresponded to those concepts’. And he expands this a little further with ‘It follows that the nature of the world we think about is at least partly determined by the thoughts we have about it.’ Someone who denies (B) is labelled ‘Platonist’ by Morris, a term he borrows from discussions of Wittgenstein’s views on meaning (that is, Platonism, in Fregean guise, is one of the views of meaning that Wittgenstein attacks in Philosophical Investigations[5]). The third major position is the No-theory view, which denies (A) for non-Platonic reasons. Since the conceptualist proposes (B) partly because she does not see how metaphysics is possible given the assumptions of Platonism, claim (A) is an important part of conceptualism. Above all else, conceptualism is a starting point for metaphysical theorising.
A couple of points about the dependence relation specified in (B) are worth stressing. Firstly, the connection between a concept and a fact is not to be seen as a completely determining relation from the former to the latter. (B) allows for an externalism that goes the other way. The natural kind that I refer to when I say ‘water’, may partly determine what the word ‘water’ means in my mouth. Secondly, there must be a gap between concept and object, on Morris’ account, in order to allow for interesting metaphysical explanations. He goes on to specify a condition for factual equivalence for expressions in order to constrain metaphysical reductions. If there was no ‘space’ between fact and concept (if two concepts could not count as concepts of the same fact), the only condition for factual equivalence could be conceptual equivalence. But this would not allow for interesting reductions. Morris therefore wants to allow for different concepts to count as concepts of the same thing.
In order to avoid the charge of empirical idealism, Morris makes it clear that what he is talking about is not a connection between some psychological particular (a ‘concept’ or ‘thought’ that exists in some mind) and the nature of things. The world existed, and contained many of the things with which we are familiar (rocks, trees, etc.), long before anyone was around to think about it. Conceptualism avoids denying this truism by insisting on an atemporal notion of concepts. A particular concept, for Morris, is that which is common to anyone who possesses that concept. Someone possesses a concept just in case they can have a propositional attitude that involves that concept in its contents[6]. The timeless existence of concepts is then explained as ‘a matter of there being something which it would be to possess that concept’.[7] One important consequence of this notion of timeless concepts is that the sense in which (B) suggests that our concepts ‘determine’ reality cannot be an ‘empirical sense’[8]. It should not be taken that (B) suggests that we ‘construct’ the world with our concepts, or that our concepts ‘carve up’ the world. But this way of putting the matter is not very informative. If ‘determines’ can be given a non-empirical sense, then why not ‘constructs’ and ‘carves up’? And it leaves open the question of what non-empirical sense we can give to ‘determine’.
One apparent answer to this question that Morris has to offer is a reformulation of (B) as the statement that there would not have been those objects, properties and facts if there had not been those concepts. Given the notion of concepts as timeless, however, Morris also concedes that conceptualism (and its denial, Platonism) is thereby committed to the view that there can be subjunctive conditionals with impossible antecedents that are nonetheless non-vacuously true. But this is controversial, to say the least.
Is it meaningful to claim, for instance, that ‘if 2+2 was not equal to 4, then not p’, for any filling in of p? A good reason to think that this is not meaningful is that p, it seems, is not going to affect the overall truth-value of the ‘statement’. It maybe that there is an analysis of subjunctive conditionals that admits of interesting cases of impossible antecedents, though Morris gives none[9]. In any case this way of expressing conceptualism remains obscure, to say the least[10]. Is there not an alternative way of spelling out the difference between it and Platonism? One other formulation given by Morris is worth considering, since it illustrates more clearly what I mean by ‘conceptual idealism’:
[I]f the world is capable of being thought about at all, it must be essential to the world that it is such as to be thought about… it is essential to the world that the world can be made sense of.[11]
Morris points out that this way of putting the matter suggests a link with (A) (the claim that there can be interesting metaphysical explanations) which he goes on to exploit. His understanding of (A) demands an operable condition of adequacy for metaphysical explanations. This operable condition is given in epistemological terms: different concepts of the same object, property or fact are distinguished by different ways of knowing about that object (different ‘modes of presentation’). All this is in line with the roughly Kantian nature of conceptualism that Morris emphasises[12]. However, I wish to pursue a different line of argument from (B) to (A) that replaces epistemological concerns with issues in the philosophy of language. This ‘linguistic turn’ is in line with the development of conceptual idealism in the context of twentieth century analytic philosophy, and it also brings us into line with the discussion of idealism in chapter 1. The idea that ‘it is essential to the world that the world can be made sense of’ can be then put like this: ‘it is essential to the world that it can be described’, where describing something means to bring it under a public concept. That is to say that the determining relation in (B) (‘not fixed independently of’) is to be taken in such a way such that the world (that we can think about at all) is limited by our possible concepts. All objects, properties and facts necessarily fall under some possible concept, since to be an object, property or fact just is to fall under the relevant concept.
I would like to adapt the characterisation of approaches to metaphysics provided by Morris to my own ends. Instead of the term ‘conceptualism’ I will make use of the label ‘conceptual idealism’ as the denial of the approach called ‘conceptual realism’. Any approach that denies the possibility of metaphysical explanations or philosophical theories in one or more areas of philosophical reflection will be regarded as ‘quietist’ with respect to that area. I make this change in terminology in order to avoid any unnecessary commitments to the details of Morris’s own philosophy, and also to avoid falsely attributing commitments to him. The positions are supposed to be quite general, since my aim is to reflect on the nature of philosophical theorising as generally as possible. I call this a ‘topology’ of metaphysics since I intend to map out some fixed points to which particular theories can be seen as relative to, rather than strictly subordinated to. Any metaphysical system can be located on the map by examining how it attempts to deal with the problems of metaphysical theorising that one can identify at this general level.
A conceptual idealist approach to philosophical explanation maintains that
(i) there can be interesting metaphysical explanations;
(ii) there is a necessary correspondence between the concepts we use and the natures of the objects, properties and facts picked out by them; such that,
(iii) insofar as a thing can be thought about at all, it necessarily falls under some possible concept.
By ‘interesting’ in (i) I mean non-trivial and justifiable. That is to say, metaphysical explanations generate knowledge. A metaphysical explanation is one that explains the way things seem (or happen to be conceptualised) on the basis of the way things really are. While the conceptual idealist will help herself to such a distinction, she disregards the idea of a thing-in-itself that cannot be thought about or predicated in judgement. Conceptual realism denies (ii) and consequently (iii). While the conceptual realist maintains (i), he holds that the true nature of reality is described truly by mind independent concepts, such as platonic forms. It is therefore a contingent matter whether or not that description can be understood, or even recognised by us. Quietism denies (i) with respect to a particular area of philosophical discourse. Note that this might simply be a claim about the nature of the discourse, denying the possibility of philosophical explanation in terms of the nature of reality it purports to describe. It is not self-evidently refuting (in the way the verification principle is). This topic will be taken up again in chapter 6. The possibility of quietism with respect to subjectivity will be examined in part 3 of this thesis.
There is a certain conception of reality that does not fit neatly into the topology. According to Kant, the noumenal world, things-in-themselves, are necessarily beyond the reach of the understanding, and therefore knowledge. The conceptual realist wants to maintain a contingent relationship between the understanding and things-in-themselves, the conceptual idealist a necessary one, whereas the Kantian position apparently denies any possible connection. One reason for not immediately granting this position a separate location on the topological map is that it is apt to strike one as nonsense. It is at least arguable that the idea of something of which we can necessarily have no knowledge whatsoever is not, as Kant claimed, intelligible. A second reason for denying it independent status is that, depending on further assumptions, the position collapses into one of the others. Such a position holds that the notion of ‘being’ is independent of our concepts in general, though the latter presumably presupposes the former. If, when pressed, the proponent of such a view wants to admit that, say, the progress of science or philosophy provides us with a glimpse of ‘an actual, mind independent reality’, then he is really being a conceptual realist. If, on the other hand, he maintains that nothing can be known of the thing-in-itself, but that we can nevertheless produce metaphysical explanations within our understanding of the phenomenal world, then the thing-in-itself plays no apparent role whatsoever, and he has turned to conceptual idealism. Finally, if he admits that metaphysical explanation must refer to the thing-in-itself, while maintaining that such a thing must remain mysterious to us, he has offered a general argument for quietism. But I do not wish to argue conclusively for the instability of a Kantian position here. Whether or not it collapses into one of the other three positions under pressure of argument, transcendental idealism at least begins by staking a claim for the middle ground. I use the term transcendental idealism to mean any doctrine that maintains that the world that is knowable or representable at all is necessarily knowable or representable, and that anything else is necessarily beyond the reach of knowledge or representation. While Kant first advanced such a position in part to constrain free wheeling metaphysics, he also did so to defend the idea that certain philosophical justifications are possible. In so far as these justifications involve metaphysical commitments, this position is distinct from quietism.[13]
Conceptual realism denies (ii). This allows for considerable scepticism about our knowledge of Reality. If the connection our knowledge and conceptual scheme have with the true order of things is merely a contingent one, then there is always room for doubt as to the absolute validity of our metaphysical conclusions. If there is more than one conceptual scheme, and some of those conceptual schemes are inaccessible to us lowly humans, it remains a distinct possibility that we will never gain knowledge of the real, mind independent world. One attempt to contain this scepticism is to claim that the mind independent concepts that are manifest in the things we take ourselves to be talking about are involved in bestowing meaning on our words, albeit in a contingent manner. Such a conceptual realist theory of meaning holds that the meaning of our terms is given by the things they refer to. This idea introduces a further complication to our topology. The conceptual realist may make some concessions to conceptual idealism in order to give his thesis some justification. He may allow that certain general considerations (for example, about the nature of representation) will constrain both the shape of our propositions and the nature of those objects and facts he takes to be the referents of those propositions. The existence of internal, necessary relations between the form of propositions and facts is suggested to answer certain sceptical worries about how our thoughts can really be about a mind independent world. This produces the hybrid between conceptual realism and conceptual idealism that denies (ii) but accepts some correlate of (iii). On the one hand, such a theory of meaning will involve abstract objects that are beyond our conceptual reach, in the sense that they cannot be subordinated to our concepts (they cannot be described, other than perhaps by ostensive definition). The objects are presupposed by any description. On the other hand, the nature of those objects, which necessarily coincides with representational form, limits the nature of the world. Such a hybrid would be a form of the transcendental idealism discussed above, and the most notable example is the theory of meaning in the Tractatus. That this particular attempt fails is evidenced by Wittgenstein’s later rejection of it, as will be discussed in chapter 4. But what is most significant about the metaphysics of the Tractatus is that it collapses into quietism.
There is one final point about the formulation of conceptual idealism given above that I would like to note. So put, (iii) is a fairly minimal way of disambiguating the dependence relation in (ii). It suggests only that there could be no object, property or fact that has no conceptual form. This is what rules out the given for the conceptual idealist: it doesn’t have the richness of structure that (iii) requires. But the conceptual idealist may have something stronger in mind. (ii) is an instance of the ‘formal dependence’ relation I defined in chapter 1[14]. If one also rejects the idea that the task of philosophy is to find some remotely reductive level of analysis (such as the level of quantum mechanics or sense data), then (ii) can also be taken to imply that the essences of objects, properties and facts that we normally talk about are formally dependent on our ordinary concepts. Such a conceptual idealism holds that radical scepticism is misguided (and the idea that most of our ordinary terms fail to refer does seem intolerable, since it thereby makes language vacuous). It can therefore use ordinary language as a source of metaphysics.
Conceptual realism allows for considerable scepticism about metaphysical theorising. For if it is merely contingent whether our concepts correspond to the true nature of reality, it remains mysterious how we could come to any firm conclusions about that true nature. Experience, at least on the empiricist conception, remains hopelessly ambiguous as a basis for our metaphysical judgements. (Humean scepticism, like all forms of empiricism, is a kind of conceptual realism. Empiricism allows that experience can get between the world and our conception of it). Such considerations may motivate the advocate of (i) to accept (ii), and consequently (iii), in the search for some justification of (i). One argument from (ii) to (i) goes like this. The first step is to argue that the concepts that we use to describe the world are, by and large, the ‘correct ones’. That is, there is no interesting conceptual relativism with regard to truth. This view goes naturally with conceptual idealism because it rejects a world that is independent of our concepts. The second step is to argue that the ‘large features’ of language, which embody our concepts, reveal the large features of reality. The task of metaphysics is thus to decide what those large feature of language are (and of course there is room for disagreement here). Both these steps can be found in the work of Donald Davidson.
In the discussion of idealism in chapter 1 we identified the fundamental idea of idealism as the claim that “what there is must be possibly conceivable or describable by us”. [15] This claim can be given support with an argument to the effect that the notion of what cannot be thought about or described by us, or those like us, makes no sense. The kind of argument in question claims that if we try to make sense of the notion of what we could never conceive, we must use some general notion of something being true (or being the case, or existing etc.), where we could not in principle apply any further concept. The conceptual idealist objects that to conceive of something in such vague terms is not to conceive of it adequately at all. Hence, where we thought we could conceive of a notion that we could not understand, we discover we understand nothing by this empty conception.
Davidson, in rejecting ‘the very idea of a conceptual scheme’, presents an argument to this effect. He asserts that we do not possess a general notion of truth that goes beyond the truth of all possible sentences in any language that we could understand, or that could be translated into a language that we could understand.
The criterion of a conceptual scheme different from our own now becomes: largely true but not translatable. The question whether this is a useful criterion is just the question how well we understand the notion of truth, as applied to language, independent of the notion of translation. The answer is, I think, that we do not understand it independently at all.[16]
Now Davidson’s concern here is to reject conceptual relativism: the idea that we can make sense of there being a conceptual schemes that is not our own, to which truth would be relativised. His claim is not that our conceptual scheme is the only true conceptual scheme, for, as he put it, ‘even monotheists have a religion’. If it does not make sense to talk of different conceptual schemes, it does not make sense of all conceptual schemes being one. Rather, Davidson would reject the very idea of a conceptual scheme at all: he argues against the idea of a conceptual scheme that is somehow separate and independent of the empirical content (‘experience’, ‘the given’) to which it is then applied:
[C]onceptual schemes (languages) either organize something, or they fit it… As for the entities that get organized, or which the scheme must fit, I think again we detect two main ideas: either it is reality (the universe, the world, nature), or it is experience (the passing show, surface irritations, sensory promptings, sense-data, the given).[17]
Davidson describes the dualism of scheme and content as the third and final dogma of empiricism, and in rejecting it he sees himself as severing his ties with that tradition. But this dualism is also to be found in Kant (‘the given’), so Davidson is also a long way from Transcendental Idealism. Indeed, we can see some important differences. From his denial that we can make sense of the concept of truth independently of the concept of translation, it seems that Davidson is committed to reject not only the given, but also the (Kantian) idea of the thing-in-itself. To speak of reality free of our conceptual scheme, however generally, would be a misuse of language that returns to the dualism of scheme and content.
As already discussed, both the rejection of both the given and the noumenal world is an essential part of conceptual idealism. In the case of the latter, the rejection is explicit and straightforward. Conceptual idealism is committed to the idea that things are not independent of our concepts of them, and the Kantian idea of a thing-in-itself is just the notion of things considered independently of our concepts. The rejection of the given also follows from the rejection of things independent of our concepts, for the given is supposed to be that which enters our experience yet cannot be described. Davidson, in rejecting the given – that to which our conceptual scheme is supposed to be applied[18] – is stating his conceptual idealist credentials.
The second step in the argument from (iii) to (i) is to spell out the connection between language and reality to establish conceptual investigation as a method of metaphysics. Davidson’s contribution to this debate is set forth in ‘The Method of Truth in Metaphysics’. The central idea is a simple one:
In sharing a language, in whatever sense required for communication, we share a picture of the world that must, in its large features, be true. It follows that in making manifest the large features of our language, we make manifest the large features of reality. One way of pursuing metaphysics is therefore to study the general structure of our language.[19]
He then goes on to argue for the importance of such a method, before describing what he takes that method to be. Firstly, he has to erect the notion of a shared picture of reality to replace the ‘very idea of a conceptual scheme’ that he rejected earlier. The connection between the two notions can be easily discerned in his argument for a shared world-view. ‘Those who understand one another’s speech must share a view of the world’ (correct or incorrect) because ‘we damage the intelligibility of our readings of the utterances of others when our method of reading puts others into what we take to be broad error.’ In other words, the common picture is required to interpret others utterances, and it is assumed that another person’s utterances can be translated into words that we understand. The salient difference between a conceptual scheme and a world picture would seem to be the fact that the latter can be considered (largely) true or false, rather than something to which truth must be relativised. But Davidson goes on to argue that we can make little sense of our shared picture being largely false[20]:
[O]bjective error can occur only in a setting of largely true belief. Agreement does not make for truth, but much of what is agreed must be true if some of what is agreed is false… too much actual error robs a person of things to go wrong about.[21]
There follows the rather curious argument of the omniscient interpreter:
[H]e attributes beliefs to others, and interprets their speech on the basis of his own beliefs, just as the rest of us do. Since he does this as the rest of us do, he perforce finds as much agreement as is needed to make sense of his attributions and interpretations; and in this case, of course, what is agreed is by hypothesis true. But now it is plain why massive error about the world is simply unintelligible, for to suppose it intelligible is to suppose there could be an interpreter (the omniscient one) who correctly interpreted someone else as being massively mistaken, and this we have shown to be impossible.[22]
I will not defend or dispute this argument here, though it is worth making a couple of comments on it. Firstly, the argument assumes that the omniscient interpreter can interpret the speaker. This is not adequately explained by merely pointing out that the interpreter is omniscient. By hypothesis, the interpreter does not use his ‘all-seeing eye’ to ascertain the beliefs of the speaker, for if he did that, there would be no need to interpret at all. The assumption is that the interpreter does not have a conceptual scheme that is incommensurable with the speakers. Thus this argument rests on the previous commitment (in step 1 above) that truth is not relative to a conceptual scheme. The second point worth making is that Davidson’s view excludes the kind of radical scepticism that Descartes contemplated. It cannot be that our beliefs about the world diverge en mass from the objective order of things, for the notion of truth only has meaning for us against the background of the actual judgement that we make and in general agree on. If we are radically wrong about the meaning of our words, nothing we say can make sense, and hence Cartesian scepticism is not possible. This rejection of an objective order somehow beyond all our ordinary judgements about the world is, of course, just the rejection of conceptual realism that marks Davidson as a conceptual idealist. Conceptual realism on the other hand, to the extent that it allows for a notion of truth external to our conceptual scheme, allows that we might be massively in error. Hence brains-in-a-vat scepticism is a real problem for conceptual realists[23].
Having made the case for the study of language as a method for metaphysics, Davidson presents his own slant on conceptual investigation. His method involves a theory of truth as a constraint on metaphysical theorising. Rather than using metaphysical considerations to decide issues in the philosophy of language, he suggests we do things the other way around: a ‘comprehensive theory of truth’, he says, ‘makes its own unavoidable demands.’ This should not blind us to the fact, however, that using such a method is based on some major metaphysical assumptions.
The central features of his philosophy of meaning and truth are well known, and for the most part they do not concern us. Conceptual idealism is not restricted to the kind of theory of meaning that Davidson favours. But there is one feature of this theory that is relevant to the discussion of conceptual idealism in general. Davidson argued in ‘The Very Idea of a Conceptual Scheme’ that we have no notion of truth that goes beyond the sentence of our language that we take to be true. The notion of truth, he claims, is dependent on the notion of translation into a familiar idiom, and this is dependent on a theory of meaning. The most important restriction on this theory of meaning for Davidson is that it must be learnable[24]. It must explain how an infinite number of sentences can be potentially understood by a finite mind. The notion of truth is therefore tied to what is potentially learnable by finite minds like ours. This makes his philosophy somewhat more obviously idealistic. What is real for Davidson is restricted to what is learnable (in terms of competence, not performance based criteria). This can strike one as counter-intuitive, to say the least.
The intuitions this conflicts with are good old-fashioned conceptual realist ones. They are the intuitions that have been defended by the likes of Nagel and Williams. Nagel argues that we can easily imagine that there are minds that are superior to ours in an analogous fashion to the way our mature adult minds are superior to the undeveloped minds of twelve year olds. The vast majority of twelve-year-olds would not be able to grasp the concept of relativity as it features in theoretical physics. We could imagine a species of beings whose mental capacities were restricted in a similar way, and who could therefore not grasp sentences and concepts that we take to be meaningful. If one of these beings (Realist Junior) were to postulate that there were concepts and truths beyond their comprehension, we would have to agree with him. And if one of these beings (Idealist Junior) were to disagree, arguing that they had no notion of truth beyond what they could comprehend, we would be inclined to object. There are truths that are (by hypothesis) not understandable by the race of mentally-nine-year-olds. So is it not comprehensible that we might be in the same position with regard to some superior alien minds? Might some alien race have concepts that were beyond our comprehension?
To be consistent with his anti-relativism, Davidson must allow that the sentence uttered by the junior idealist is translatable into a sentence in our idiom with the same truth-value. Since the sentence ‘there are no truths beyond their comprehension’ is clearly false in our idiom, Davidson cannot allow this as the correct translation while maintaining that the junior idealist is right. The possibility of claiming that the junior Platonist is right but conceptual idealism is correct in our idiom is merely chauvinistic given the possibility of the superior alien minds. The one plausible claim open to the conceptual idealist is that the meaning of the sentence ‘there are no truths beyond our comprehension’ is exactly the same whether spoken by a junior, a human or an intellectually superior alien. On this account, talk of ‘comprehension’ is somewhat misleading. Davidson is committed to claiming that all three groups have the same conceptual scheme (or rather, that there is no such thing as a conceptual scheme, but we shall put that point to one side as a mere terminological dispute). Admittedly, some of the concepts of this ‘scheme’ are not fully grasped by all members of this three-tiered group of language speakers, but that is just to point out a limitation in the relative performance of their minds, not their potential capacity. Furthermore, given the possibility of trustworthy communication between the groups, a member of one of the lower two tiers has epistemological access to the truths discernible by their intellectual superiors. Just as I can know the difference between gold and fool’s gold by consulting an expert, so the race of nine-year-olds can consult a theoretical physicist on matters pertaining to relativity. What prevents them grasping these matters more directly is merely their limited attention span, lack of general intelligence and mental acuity, and so on.
Davidson is already committed to the claim that any sentence used by the aliens is translatable into English, and into the language of the Juniors for that matter[25]. The translation maybe unbearably difficult to understand, but it must be possible. No doubt there are some theoretical concepts understood by a small elite of actual humans. These concepts will be forever out of the grasp of the masses, but they are nevertheless, in principle, understandable. The defence of the conceptual idealist might be that at most our failure to grasp the concepts of an imaginary alien race is comparable to the failure of some of us to master the subtleties of theoretical physics.
While this response does just enough to meet the realist objection formally, it hardly settles the matter. The conceptual realist is going to continue to insist that the aliens could have concepts that we could not grasp in principle, and it is difficult not to feel some sympathy with the intuitions behind this thought. On Morris’ account of concepts, a concept exists just in case there is something that there is to have a propositional attitude involving that concept. Could there not be something that there was to believe p, even though human minds could not instantiate those conditions? The matter seems irresolvable, since to prove his case the conceptual realist must come up with an example of a concept that we cannot understand, and this, by the very nature of the required example, he cannot do. For what is there to show that there is a concept at all there, if not that we understand it? Of course, the realist can complain that the idealist has not spelt out what it means to say that all concepts are ‘in principle’ understandable by us. But we can imagine possible answers. The Davidsonian response would presumably be that the concept is expressible in our language, allowing that the resulting sentence maybe so long as to render it practically ungraspable[26]. In any case the realist is no better off, for he cannot spell out what it would be for a concept to be ‘in principle’ ungraspable. He can merely gesture in the direction of other kinds of minds. We must concede at least this much to Davidson: the concept of a conceptual scheme, applied at a very general level, is not entirely clear.
In any case the response I have proffered for the conceptual idealist does not go to the heart of his claim. His point is rather that what we take to be meaningful and concept-involving behaviour is bound up in our own use of concepts. The fact that we cannot make sense of some behaviour would amount to evidence that the behaviour was not intelligent.
But what about the following situation: we come across a strange group of creatures, who appear to be communicating with each other, though all our best efforts fail to throw any light on what they are communicating. The noises they make seem to bear no consistent relation to the things they do. If we gag them however, their behaviour is thrown into confusion (as our behaviour might be if we were prevented from communicating). That is to say, we have evidence that they are indeed communicating in a language that we could not understand. Merely pointing out that nothing “could count as evidence that some form of activity could not be interpreted into our language that was not at the same time evidence that that form of activity was not speech behaviour,” will not suffice, as Davidson admits. He concedes that to be convincing, this should be the result of an argument.
To the extent that Davidson does go on to present an argument, it involves rejecting the idea of a pre-conceptual ‘something’ that is independent of the conceptual scheme, the mysterious given or thing-in-itself. He claims that the various ways that have been proposed to make sense of the idea of interpreting or organising the world involve some notion of the world that is beyond comprehension. But this is just to restate the belief that there can be nothing beyond our conceptual scheme. (Or rather, that thinking of the world as something interpreted or organised by our conceptual scheme is a mistake. The notion of a ‘conceptual scheme’ is not meaningful at this abstract level.) So finally we come up against the intuitions that mark the fundamental distinction between conceptual realism and conceptual idealism. It would seem that the debate is irresolvable without recourse to the very assumptions that are at issue.[27]
But there is a simple objection to conceptual idealism that cannot be so easily deflected as begging the point at issue. Put simply, it is this: there are some things with which we are acquainted, that necessarily evade any effort to fully conceptualise them; namely, sensations. The traditional Cartesian view is that sensations are inherently private in a philosophically significant sense. Since my sensations are only known to me, and are known only by ‘acquaintance’, they remain forever beyond the reach of public discourse. There is something ‘ineffable’ about them. No matter how sophisticated my descriptive (public) language, nothing I say seems to capture the true essence of ‘the way things feel to me’. Of course, it comes as no surprise that conceptual idealism is incompatible with the Cartesian view of experience. Transcendental idealism involves, as we saw in chapter one, a rejection of this picture of the mind, and conceptual idealism reinforces that rejection. Cartesian philosophy is a form of conceptual realism, and it is the apparent insolubility of Cartesian and Humean scepticism that lends the first support to conceptual idealism. But the notion of sensations being private and ineffable is a very hard one to shake off. Surely it is true that I can only know how something feels by actually feeling it, and no mere concept can stand in place of that direct acquaintance. So is there not something very real that I could never conceptualise in a public manner? Does this not show conceptual idealism to be wrong?
This is not the traditional ‘problem of privacy’. The traditional problem is a sceptical problem in the philosophy of mind. It comes about by reflecting on the nature of experience with the thought that ‘you can’t see what I can see’. This leads one into philosophical problems that it is hard to free oneself from. It leads to scepticism about other minds, or even about mind-independent reality in general. Whatever the difficulties involved with such philosophical reasoning, they are not the problems faced by the conceptual idealist. The traditional problem of privacy is set up using something like the Cartesian conception of mind. It presupposes that the world is distinct from my impression of it. The problem for the conceptual idealist is that this traditional problem, while it seems to make sense at least as a philosophical problem, cannot even be expressed without making concessions to conceptual realism. Conceptual realism makes sense of scepticism about other minds, while conceptual idealism simple denies there is a problem.
The problem for conceptual idealism can be made vivid by considering two related questions in the philosophy of mind that lead one to believe that the world necessarily outstrips our conception of it. Nagel raised the first question when he asked, “What is it like to be a bat?”[28] Simply contemplating the mind-boggling difference between a bats form of life and our own can bring one to the conclusion that such a different form of subjectivity must be completely alien to us. And this is such that it suggests that the bat’s subjectivity must necessarily be missed out of our objective world-view. The point is not that we do not yet have the relevant objective facts about the bat’s perceptual and nervous systems. The intuition that is invoked by this example is that there are some facts (some ‘subjective facts’) that are fundamentally inaccessible to creatures that do not navigate by sonar. Call this the ‘inter-species problem of subjectivity’.
In addition to this, a further problem that privacy presents to the conceptual idealist might be called the ‘inter-personal problem of subjectivity’. For while considering alien forms of life may vividly demonstrate the limits of our conceptual scheme, the problem arises much closer to home. My conception of what it is like to experience, say, red, has been learned through my own experience: by looking at red things. But nothing about those experiences demonstrates that they are just like the way other people experience red. Sure, we may agree on what things are to be called ‘red’, and have similar reactions to its presentation (e.g. the tendency to describe it as a ‘warm’ colour and to act more aggressively in red environments). But how do I know that it feels the same for you as it does for me? I only have my own experience to go on. (The answer that I have the same reactions does not settle the matter: for how do I know that it feels the same way to have those reactions). And nothing that we say about our experience will settle the matter. For what we say will eventually be rooted in our experience, and any amount of agreement will always leave room for a systematic difference between the way it seems to me and the way it seems to you. And this is because there seems to be something missed out of our inter-subjective ways of describing things. It seems that, contra to the claims of the conceptual idealist, there does indeed seem to be something that is missed out by our conceptual scheme. And that is the very thing that makes that conceptual scheme possible: subjectivity.
The tempting answer for the conceptual idealist to give is that Wittgenstein has already dealt with the illusion of privacy, most famously in the “private language argument”. But it is easier to pay lip service to a philosophy as subtle and complex as Wittgenstein’s than to demonstrate its validity or persuasive power. In the next part of this thesis I will look closely at the ideas in Wittgenstein’s later philosophy that may give support to the conceptual idealist, and may have influenced present day idealism. I will argue, however, that they can only be understood (and are therefore only persuasive) as part of Wittgenstein’s philosophy as a whole. And that philosophy is not idealist, but quietist. Eventually I will argue (in part 3) that it is not possible to solve the problem of subjectivity without giving up the claim to substantial metaphysical theories.
[ previous | contents | next ] |
[ previous | contents | next ] |
| © July 2001 |