Abstract
At the end of Chapter II we noted four desiderata for an explication of Realism. It was considered desirable that an explication (1) furnish us with a unified way of looking at the Realism/anti-Realism issue, (2) provide us with a way of giving a clear meaning to the idea that the world might in some sense depend on us, (3) explain the element of truth in the other explications of Realism and (4) help us to actually resolve the Realism/anti-Realism issue. In this chapter it will be argued that the theory of dependence given in the previous chapter enables us to develop a conception of Realism that can at least satisfy the first three of these desiderata. The task showing that it is can satisfy the fourth desideratum — enabling us to resolve the Realism/anti-Realism issue — will take us into the second half of this book.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
Notes
See Goodman Of Mind and Other Matters,p.36, footnote 5.
According to the later Putnam, truth is an epistemic notion. He sees it as an ‘idealisation of rational acceptability’. But there are in Putnam’s writings two different ways of spelling this out. In his papers, ‘Realism and Reason’ and ’Models and Reality’, he says that a theory is epistemically ideal if it correctly predicts all possible observation sentences, and is simple, plausible, conservative, etc. It seems as though what Putnam has in mind here are the conditions under which ideal physics, or, perhaps, a Quinean ’system of the world’ would be ideal. But in other places, for example in his Reason, Truth and History,he says that a sentence is true if it would be confirmed under epistemically ideal conditions. That would seem to be an account of the conditions under which a single sentence would be true. Moreover, since it only seems to make sense of a fairly observational sentence being ’confirmed under epistemically ideal conditions’, it is, perhaps, intended as giving the conditions under which a relatively observational, single sentence must be true.
Although Putnam himself does not put things this way, it is natural to see these two characterisations of ‘idealised rational acceptability’ as together entailing a, broadly speaking, Quinean view of the structure of our knowledge. Quine conceived of our system of beliefs as a ‘web’ or ’field of force’ which impinged on the world only at its sensory periphery. The inside of the web of our beliefs, however, is underdetermined by the sensory periphery. On Quine’s view it is simplicity and conservatism, together with the sensory periphery, that determines the nature of the ‘interior’ of our web of beliefs. Quine, of course, only saw this as an account of our actual system of beliefs and as a (naturalised) account of the conditions under which it is rational to accept a belief But I suggest we can see Putnam as offering this Quinean view as giving us an account of the conditions under which beliefs are true. A belief at the observational periphery is true iff it would be accepted under epistemically ideal conditions. A more theoretical belief is true iff it is a part of the ideal (i.e., maximally simple) theoretical system that explains the true observational beliefs, that is, the ones that would be assented to under epistemically ideal conditions. Quine, of course, rejected this as an account of truth since he believed it to be a consequence of the underdetermination of theory by data that there could be incompatible such systems. Putnam’s reply to this objection appears to implicit in be his celebrated ’model-theoretic’ argument against Realism, discussed in chapter nine below.
Putnam’s argument for this is discussed in Chapter X, below.
See, for example, his paper ‘Realism and Equivalence’, reprinted as ch. 2 of volume 3 of his Collected Philosophical Papers Realism and Reason. However, it is not clear whether the thesis concerning which Putnam expresses agnosticism is the thesis that there could be two cognitively non-equivalent but empirically ideal theories which are both epistemically ideal relative to one (our) conception of the epistemically ideal, or whether there could be conceptions of the epistemically ideal sufficiently different to entail that cognitively inequivalent theories are both epistemically ideal. While Putnam would perhaps be right to be an agnostic concerning the first thesis, the second thesis is surely obviously correct. (See Chapter three, section 6.2.)
Here I am assuming that sense can be given to the notion of two theories being of the same subject-matter. If it is believed that this is an inadmissably unclear notion, it can instead be assumed that the two theories are both global theories. (See Chapter III, section 2.3.)
We will not concern ourselves with the structure of Putnam’s arguments here. They are, however, discussed in Chapter IX.
It is perhaps worth noting that this line of thought also indicates that Quine ought to be seen as a Semantic Negative anti-Realist. Although Quine has not explicitly written a great deal about the Realism/anti-Realism issue, the doctrine of the indeterminacy of translation would appear to be in conflict with Realism. The Realist, on the view advocated here, says that when the sentences in one true theory T of a particular domain are correctly interpreted, they are equivalent to the sentences of any other true theory of that domain. So the Realist requires the notion of correct interpretation of theoretical sentences. But if translation is, as Quine asserts, radically indeterminate for theoretical sentences then there will evidently be no such thing as the uniquely correct interpretation of a theoretical sentence: there will be a number of different interpretations, which will not be equivalent, “in any sense of equivalence no matter how loose”. So Quine’s position would appear to be implicitly one of semantic Negative anti-Realism.
Although the situation with Dummett is further complicated by the fact that he appears to be ‘exploring’ rather than advocating anti-Realism. (See, for example, the ‘Preface’ to Truth and Other Enigmas). I have even been informed by some Oxford-based philosophers that Dummett is actually a (theistic) Realist. We should also note that some developments of Dummettian anti-Realism have more affinities with Semantic Negative anti-Realism. These developments are discussed below.
Or, rather, he seems to regard the claim that the world might depend on human language, or human investigations, as having only metaphorical meaning. (See, for example, The Logical Basis of Metaphysics,p.15.)
We can bring out the way in which a thing’s being evidence for the truth of a sentence depends on the Intentional by considering the following example. Cosider the sentence, ‘A kangaroo hopped past here five days ago.’ Under certain circumstances, some depressions in the sand might constitute evidence for this claim. To a skilled Aboriginal tracker, these depressions might constitute good evidence for the truth of the sentence. But if the skills of the Aboriginal tracker were lost, then the depressions in the sand might cease to be evidence for the truth of the sentence: they might become ‘mere’ depressions in the sand. A particular thing or state of affairs E constitutes evidence for the truth of some sentence S only relative to some human skills, or beliefs or techniques.
The same point can be made with the example of the five rabbits in my backyard. We can imagine certain circumstances in which a photograph, clearly showing five rabbits, was taken to constitute conclusive evidence that there were five rabbits in my backyard on such and such a date. But we can also imagine possible circumstances (such as a ‘nuclear winter’ which plunged the world back into another Stone Age) in which, although the photographs themselves were not destroyed, knowledge of the significance of photographs was. Perhaps in this postnuclear-winter age, knowledge of the means by which photographs were produced was lost, and so a photograph provided no more evidence for the state of affairs it depicted than did the work of an imaginative artist. In such a case, there would cease to be any evidence for the truth of a claim because of a change in the state of the Intentional.
As presented in Chapter III, truth-constitutive laws are conditions of the form (Vp)(E(p) —* T(p)) is true). As a biconditional, (DT) evidently entails such a conditional.
See, for example, the ‘Preface’ to Truth and Other Enigmasespecially pp.xxv-xxvii and The Logical Basis of Metaphysicsp.15.
The expression ‘Dummett’s Metaphor Thesis’ comes from Michael Devitt’s discussion of Dummett in his Realism and Truth,chapter 14, esp. p.264.
Actually, this is questionable: it appears to depend on the assumption that “ - is conclusive evidence for - ” is a transitive relation -, that is, that if A is conclusive evidence for B and B is conclusive evidence for C, then A is conclusive evidence for C.
The term ‘constructivism’ was, as far as I know, first applied to a position in the philosophy of science by Richard Boyd in his ‘On the Current Status of Scientific Realism’. Devitt also uses the term when discussion Kuhn and Feyerabend in his Realism and Truth,especially chapter 9. van Fraassen’s ‘constructive empiricism’ is discussed below.
See Kuhn The Structure of Scientific Revolutionsp.121.
See Kuhn op. cit.p.117.
See Kuhn, op. cit.,p.118. One puzzling feature of this quote is that Kuhn feels that it is a principle of economy that leads us to say that Lavoisier worked in a different world. But, prima facie,it seems odd to say that we can conform to a principle of economy by postulating a multiplicity of worlds. One possible reason Kuhn has for saying this can be represented as the following argument:
We need to postulate a separate world corresponding to each paradigm.
We do not need to postulate a single world underlying all paradigms, and of which they are all versions.
So: The principle of economy dictates that we ought to postulate a separate world for each paradigm, but not postulate a single underlying world.
This interpretation of Kuhn explains his puzzling appeal to economy. But if it is correct, it also brings out a striking similarity between Kuhn and Goodman. Both philosophers postulate worlds corresponding to paradigms (in Kuhn’s case) or true versions (in Goodman’s). But more significantly, both philosophers counsel us not to postulate a single world underlying all the paradigms/true versions, on the grounds that to do so would be explanatorily otiose.
I take this as obvious, but we can say (1) It does not provide us with a clarification of the intuitive idea of Realism, that the world has its features independently of the Intentional; (2) It would not appear to give us a unified way of looking at various Realist and anti-Realist doctrines; (3) It does not clarify the notion of dependence; and (4) It does not explain the element of truth that exists in the other explications of Realism.
In Chapter II it was argued that there is another shortcoming with the explication of Realism as the doctrine that truth may transcend verification - and, indeed, with all explications of that appeal to epistemic notions: they may not always help us to resolve the Realism/antiRealism issue. But here we are only concerned with the issue whether this explication correctly divides doctrines into Realist and anti-Realist, rather than with how useful it is in enabling us to tell which is true.
I say ‘one’ say because it might be suggested that one way of rejecting the idea that truth is non-epistemic is to deny that truth is anything at all, and adopt a ‘redundancy theory’ of truth.
Evidently, it is assumed here that ‘Truth is verifiability’ is advanced as a necessary truth, but I do not know of any philosophers who would wish to claim it was only contingent.
If W depends on some aspect of the Intentional I then, on the view advocated here there must be (i) A truth-constitutive law L that enables us to derive W from I and (ii) the putative explanation employing L must have explanatory force. So, if W is independent of the Intentional (i.e., of all aspects of the Intentional), it must be the case that either there is no truth-constitutive law enabling us to derive W from any description of the Intentional, or any such putative explanation lacks explanatory force, or both. But in general there will always be some truth-constitutive law enabling us to derive W from some description of the intentional, even if the truth-constitutive law is, for example, ‘If p bears the correspondence relation to the world, p is true’. So, what will be denied by the Realist is that there is an explanation with explanatory force.
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). The Explanatory Priority View of the Realism/Anti-Realism Issue. In: Realism and Explanatory Priority. Philosophical Studies Series, vol 71. Springer, Dordrecht. https://doi.org/10.1007/978-94-017-2844-7_4
Download citation
DOI: https://doi.org/10.1007/978-94-017-2844-7_4
Publisher Name: Springer, Dordrecht
Print ISBN: 978-90-481-4823-3
Online ISBN: 978-94-017-2844-7
eBook Packages: Springer Book Archive