Abstract
The aim of this chapter and the next is to develop a way of looking at Realism and anti-Realism. I should begin with a disclaimer. In these two chapters no original philosophical theses will be advanced, and no arguments, either for or against Realism, will be developed. The aim is simply to develop a new way of looking at Realism: A novel perspective or gestalt will be offered on some familiar philosophical ideas.
This is a preview of subscription content, log in via an institution to check access.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Notes
If the notion of dependence is meaningless, then so must be the notion of independence be meaningless (since, presumably X is independent of Y if it is not the case that X is dependent on Y). But since Realism is defined in terms of independence, it would follow that Realism too is meaningless. And if Realism is meaningless, it clearly cannot be true. It is interesting to note that this simple line of reasoning would appear to be overlooked by Michael Devitt in his Realism and Truth. Devitt seems to regard the notion of dependence as hopelessly obscure, and for the onus tó be on the opponent of Realism to clarify this notion. But it is hard to see how Devitt could regard the notion of dependence as hopelessly obscure and yet be happy with the notion of independence. Indeed, it would seem that - very much contrary to his intentions - Devitt should be seen as a sort of an anti-Realist. In the terminology to be introduced below, he is a Semantic Negative anti-Realist. See, especially, p.16 of Realism and Truth, where he says ‘in so far as [the notion of dependence] is unintelligible, the realist needs no argument to dismiss it.’
It should also be noted, however, that Goodman has (unhelpfully, I think) said that his claim that ‘we make worlds’ is to be interpreted literally. But usually when we make something, such as a table, we causally act on its parts. This type of creation is very different from any sense in which we ‘create’ worlds.
I have encountered philosophers who describe the idea that the world might somehow depend on our theories as ‘guff’ and as ‘linguistic voodoo’. Evidently, what is behind such dismissals is the assumption that what is being claimed is that our theories causally influence the world.
It would offend against the principle of charity to interpret the claim that X depends on the mental as an empirical claim that conflicts with both science and common sense, because such an interpretation would make Realism trivially and undeniably correct. There would, on such an interpretation, be no serious Realism/anti-Realism issue. So, it is not being sufficiently charitable to the philosophers who think there is such an issue.
Hellman mentions this example in his Introduction to the 3rd edition of Goodman’s The Structure of Appearance p.xxviii. Hellman notices the similarity of this case to the view developed in Paul Benacerraf s well-known paper ‘What Numbers Could Not Be’ (The Philosophical Review 74 (1965). The other examples come from Goodman’s Ways of Worldmaking.
See Goodman Of Mind and Other Matters, pp.35–36.
It seems as though this is implicitly acknowledged by Goodman himself. He adds to these examples a footnote asserting that not all different but equally true descriptions arise from carving up an underlying space-time in different ways. However, he does not elaborate on this point. (See his Of Mind and Other Matters, p.36, footnote 5.)
In his paper, ‘Realism and Independence’, Elliot Sober is much exercised by the question of whether the counterfactual approach to Realism can deal with difficulties such as these. Despite considerable ingenuity in tackline the issue, Sober acknowledges that his own approach is not entirely satisfactory.
Or, alternatively, we could consider the classes of possible worlds of which S* and S are true.
See, for example, Johnston’s paper, “Explanation, Response-dependence and Judgement-Dependence” in Response-Dependent Concepts, edited by Peter Menzies (published by RSSS, Australian National University, 1991).
I am not here claiming that it is obvious, or beyond serious argument, that a Realist attitude is appropriate towards ‘gold’. Such a claim would clearly beg the question against global anti-Realists such as Nelson Goodman and Hilary Putnam. It is merely suggested that construing anti-Realism as response-dependence agrees with plausible intuitions about what is real and what is not.
See Ways of Worldmaking, p.111.
We might briefly note here that we need not accept this. Graham Nerlich (in an unpublished paper) has argued that, if space were Riemannian, then we could give clear empirical sense to the claim that an object was in absolute motion. If Nerlich is right, then Goodman’s example fails. One of the pair of statement is simply right and the other simply wrong; or, they are both wrong. Goodman’s claim that they have an equal claim to truth would simply be false.
Despite the naturalness of this way of proceeding, I am not aware of any author who has attempted to explicate dependence in this way. Of course, this raises the question of why this approach has not been used before. I suspect that the reason is because it seems to be widely assumed that ‘explanation’ must mean ‘causal explanation’ and hence that the claim that some part of the world depends on the Intentional would need to be construed as the Intentional is a cause of that part of the world. But, as we noted in chapter two, it seems pretty clear that the issue that divides the Realists and the anti-Realists does not concern some strange kind of causal relation between the Intentional and the world.
For example, Putnam’s well-known model-theoretic arguments purport, perhaps, to establish a form of verificationism, but certainly do not proceed by analysis of the meaning of ‘true’.
I mention this because I have, on a number of occasions, encountered the argument that all claims of the form ‘W because I’ must be false because an aspect of the Intentional (such as an aspect of language) can never constitute a sufficient condition for the truth of a claim about the world.
Some anti-Realist claims can, perhaps, be interpreted in this way, but we should beward of assuming that they all can. Recall the distinction made earlier in this chapter between Positive anti-Realism and Negative anti-Realism. It is only the Positive anti-Realist who asserts that some aspect of the world is dependent on the Intentional. The Negative anti-Realist denies that we are justified in asserting Realism, but also refrains from making dependency claims. Since some anti-Realist positions may be merely instances of Negative anti-Realism, we ought not to assume that all anti-Realist claims can be construed as dependency claims.
It will be an explanation, analogous to the explanation given above, of why light below a certain position P on the spectrum is red light.
Of course, David Lewis’ way of dealing with a similar problem concerning the notion of causation is to say, roughly, that E causes E* iff, in all nearest possible worlds in which E does not occur, neither does E*. It might be thought that a natural move for us to make here would be to attempt to define explanatory force using Lewis’ notion of “nearest possible worlds.” However, such a move is neither necessary, nor, I think, advisable. It is not necessary because another way of explicating the notion of explanatory force will be given below. (I do not claim that this account could however, furnish us with a non-circular account of causation). It is not advisable because Lewis’ notion of “nearer possible world” is surely highly cntroversial. Nororiously, on Lewis’ account, a world differing from our own in its laws can be closer to our own world than one having the same laws but some different initial conditions. I think it is fair to say this is highly counter-intuitive. It is perhaps also worth observing it would seem to be possible to apply Lewis’ counter-factual analysis of causation to the types of explanations employed here. But since these explanations are not causal, Lewis’ analysis would therefore appear not to have satisfactorily captured the notion of causation.
Although the laws contribute to explanatory force in this particular case, it is not suggested that the presence of such laws is, in general, either necessary or sufficient for explanatory force. It is not claimed that the presence of such laws is necessary, since it is only claimed that the account given here describes one sufficient condition for explanatory force. There may be other sufficient conditions making no mention of laws. Neither is it claimed that the presence of laws is sufficient for explanatory force. It was observed above that we do not seem to be able to explain a necessary truth. It is easy to construct cases in which a necessary truth is derived, by means of true generalisations, from true statements of initial conditions. But we would clearly not regard such a derivation as being explanatory, or possessing explanatory force. (One example of this is the following: Let ‘E’ be some empirically detectable property, and let ‘E(a)’ be a contingently true sentence. Define ‘Q(a)’ to be equivalent to ’P(a) v P(a)’, where P is some arbitrary property. Then ’(Vx)(E(x) → Q(x)’ will be true and ‘E(a), (Vx)(E(x) -* Q(x)) - Q(a)’ will have the form of a D-N explanation. But since Q(a) = P(a) v P(a), Q(a) is necessary, and so we would hardly say that an explanation of Q(a) has been given.
Although it would, of course, be possible to describe a state of the Intentional in terms of descriptions of a state of the Intentional. One way of doing this is as follows: Let I be a description of a state of the Intentional. Then the corresponding state of the Intentional S, is just the set of possible world in which I is true.
It might be wondered in what sense the word ‘possible’ is being used when it is said that it is possible to Tv x. The answer is that each different sense of possible will result in a claim relevant to a different form of anti-Realism. For example, if ‘possible’ is construed as ‘practically possible’, and Tv is ‘verify’, then the corresponding Intentional adjective TA will be ‘... practically possible to verify’. To say that X is practically possible to verify will result in a description of a state of the Intentional relevant to the type of anti-Realist who identifies truth with the practical possibility of verification. But if, for example, ‘possible’ is identified with ‘logically possible’, we will be able to formulate descriptions of states of the Intentional relevant to the anti-Realist who identifies truth with the logical possibility of verification. Different senses of ‘possible’ will be appropriate for different varieties of anti-Realist.
This can be illustrated with the following example. it is true that ‘It is necessary that 32 = 9’. But substituting for ‘32’ the co-referential expression ‘the number of the planets’ yields ‘It is necessary that the number of planets = 9’. There are some interpretations of the notion of necessity such as that developed by C. I. Lewis, on which this latter sentence is not true.
Actually, this is a slight oversimplification. There are perhaps some construals of the notion of necessity on which ‘necessary’ can be defined in terms of Intentional verbs. ‘Prove’ is evidently an Intentional verb, since it gives rise to referentially opaque contexts. (To prove that \14 = 2 is not to prove that V’4 -= the number of moons of the heaviest planet in the galaxy.) Therefore ‘provable’ is an Intentional adjective, and ‘provable a priori’ is a conditional Intentional adjective. But it might be held that there is a sense of ‘necessary’ in which ‘X is necessary’ means ‘It is possible, at least in principle, to prove X a priori.’ So, on the account advocated here, to say that something is necessary in this sense would be to give a description of a state of the Intentional. But this is surely an entirely appropriate outcome, since to say that something is necessary in this sense is to make a claim about what we can prove and hence about what we can know. Statements about what we can prove are also just the type of thing that would be likely to figure in the Initial Conditions of an anti-Realist’s dependency claim. So, it is appropriate that statements of the form ‘It is necessary that p’, where ‘necessary’ is construed in the sense under discussion, should be counted as descriptions of states of the Intentional. There are, however, some other interpretations of the notions of necessity, such as, for example, Kripke’s metaphysical necessity, on which it would appear it should not count as an Intentional term. And as far as I can see, on the account offered here, Kripkean ‘necessity’ is not an Intentional term.
Condition (b) appeals to the notion of a state of the Intentional IA different from I. We can define criteria of identity for states of the Intentional as follows: Let “It is ‘P’d that p ” and “It is ‘P*’d that p*” be descriptions of states of the Intentional. They describe the same state of the Intentional just in case the Intentional descriptors ‘V“ and ”‘P“ can be correctly given the same definitions in terms of Intentional verbs, and ”p“ and ”p*“ have the same sense. Otherwise they describe different states of the Intentional.
It may be objected that the situation just described is not one in which some theory other than the actually ideal theory is ideal,but is rather a situation in which some theory other than the actually ideal theory would be judged to be ideal by people who had different epistemic values to our own. It may be objected that the theory that would truly be ideal would remain unchanged, no matter how epistemic values might change. But such a view is surely absurd. The epistemically ideal theory is the theory that is maximally rational to believe, and rationality is a normative notion. That which is epistemically ideal is that which we ought to believe. To say that that which ought to be believed by some investigators is independent of their epistemic values is absurd.
But it does not seem quite right to use the expression ‘in virtue of in causal explanations. For example, it does not seem quite right to say ‘The fire occurred in virtue of the spark’ or ‘The avalanche occurred in virtue of the explosion’.
For example, Putnam in Meaning and the Moral Sciences, part IV, “Realism and Reason”
Of course, not everyone will agree that mathematical truth is more fundamental than our having accepted a particular mapping between set theory and arithmetic. But the dependency theorists claim that we can make e.g. 2 = {ø, {ø}} true by accepting one particular mapping rather than another, would only be comprehensible if the truth of this mathematical claim was constituted by our acceptance of this mapping, and that our acceptance of the mapping is more fundamental than the truth of the claim that 2 = {ø, {ø}}.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 1997 Springer Science+Business Media Dordrecht
About this chapter
Cite this chapter
Wright, J. (1997). A Theory of Dependence. In: Realism and Explanatory Priority. Philosophical Studies Series, vol 71. Springer, Dordrecht. https://doi.org/10.1007/978-94-017-2844-7_3
Download citation
DOI: https://doi.org/10.1007/978-94-017-2844-7_3
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4823-3
Online ISBN: 978-94-017-2844-7
eBook Packages: Springer Book Archive