Abstract
Our aim is to show how we can have good reasons for claims about the unobservable, theoretical claims made by science. But: there are objections to the thesis that we can have such reasons. The aim of the early chapters of this book is to critically evaluate those objections. In this chapter we will only be concerned with one of the objections: Humean skepticism about induction.
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
Notes
- 1.
Hume gave his argument against induction in his An Enquiry Concerning Human Understanding, especially Section IV: “Sceptical Doubts Concerning the Operation of the Understanding”. However, what philosophers nowadays generally have in mind when they refer to “Hume’s Argument” perhaps goes rather beyond Hume’s brief treatment of the topic. Some more recent statements of what has come to be known “Hume’s Argument” are Brian Skyrms Choice and Chance, Chapter Two “The Traditional Problem of Induction” and Wesley Salmon An Encounter with David Hume.
In his Popper and After: Four Modern Irrationalists David Stove identifies Hume’s scepticism about induction as the underlying source of an irrationalism about science he discerns in Popper, Kuhn, Lakatos and Feyerabend.
- 2.
The notion that plays this role on the view given here is the notion of the independence of theory from data. But, the argument given in favour of that notion is, again, very closely related to the argument for induction.
- 3.
See for example David Stove The Rationality of Induction, 1986, Oxford, Clarendon Press.
- 4.
See, for example, F. J. Clendinnen “Rational Expectation and Simplicity” in What? Where? When? Why?; Australasian Studies in History and Philosophy of Science, vol 1, (1982), pp.1–25.
- 5.
See Laurence BonJour In Defence of Pure Reason (Cambridge University Press).
- 6.
Hume’s own statement of what has since come to be known as “Hume’s Argument” can be found in Hume’s Enquiry Concerning Human Understanding, paragraph 4.<?spieprPar16?>
- 7.
This account of what is required of a justification of induction follows Skyrms.
- 8.
Edmund Gettier “Is Justified True Belief Knowledge?”. Analysis, 23(6), pp.121–123, (1963).
- 9.
BonJour, op cit, pp.1–6.
- 10.
Hypotheses attributing non-standard metrical properties to space are a source of examples of this sort. Possibly the example of this kind most strongly in conflict with what we believe to be true is the concave Earth hypothesis developed by Mostapha Abdelkader: “A Geocosmos: Mapping Outer Space into a Hollow Earth” in Speculations in Science and Technology. (vol 6, pp.81–89, (1983)). For a discussion, see “Quine on space-time” by J. J. C. Smart, and the reply by Quine, in The Philosophy of W. V. Quine edited by Lewis Hahn and Paul Schilpp (Open Court, 1986). Abdelkader develops a view on which the surface of the Earth is concave, and in which the sun, stars and galaxies are contained within the sphere that is the surface of the Earth. It appears that by modifying the paths taken by rays of light, other motions and the laws of physics, Abdelkader’s view can be made empirically indistinguishable from the view we believe to be true. Surely, Abdelkader’s view is flagrantly unreasonable. However, if it is unreasonable even though empirically equivalent to our view, whatever reason there is to favour our view must be non-empirical.
In his discussion of Abdelkader’s hypothesis, Martin Gardner allows that empirical data could not give us reason to prefer our view to Abdelkader’s. Gardner instead argues that Abdelkader’s view must be rejected because of its great complexity. While this is may be (at least in the present author’s view) correct, it leaves unanswered the question: What, if anything, entitles us to assume simplicity is a sign of truth? In the present context the point to note is that it is far from clear how there might be empirical evidence that simplicity is a sign of truth. The system of physical laws Abdelkader postulates, in all their baroque complexity, will have every bit as good a track record as our laws. So, the empirical evidence could equally well be taken to support the hypothesis that simplicity is a sign of truth, or that baroque complexity is a sign of truth.
- 11.
See for example Laurence BonJour “Externalist Theories of Empirical Knowledge” Midwest Studies in Philosophy 5, 53–57. (1980). In this paper BonJour the case of “Clairvoyant Norman”. BonJour argues that, despite the fact that clairvoyant Norman’s beliefs are formed in a way that reliably leads to truth, we would not say he had good reason for his beliefs. The imaginary example of “Sally” and “Wally” in the text to follow is of the same general sort as BonJour’s “Clairvoyant Norman”, but it will also be argued this imaginary case has an advantage over the example of “Clairvoyant Norman” .
- 12.
See Alvin Goldman Epistemology and Cognition, Harvard University Press, (1986), pp.111–112.
- 13.
The argument just given, although based in part on Bonjour’s “Clairvoyant Norman”, is designed to bring out how reliabilism has difficulty in accommodating the idea that a function of rationality is to act as guide to truth about the less accessible parts of reality. The argument given also uses ideas developed by Keith Lehrer and Cohen “Justification, Truth and Coherence”, Synthese, 55, pp. 191–207. Lehrer and Cohen consider Descartes’ “evil genius” hypothesis. They argue that Descartes’ hypothesis creates a difficulty for reliabilism. More specifically, they argue that a person misled by Descartes’ evil genius would have just as much reason as us to believe in the existence of an external, material world. But, of course, the method by which the person deceived by Descartes evil genius arrives at their beliefs does not reliably produce truths. Hence, the way in which a person deceived by Descartes’ evil genius constitutes a counter example to reliabilism.
It is suggested that perhaps the argument given here might have an advantage over that given by Lehrer and Cohen. It is open to dispute whether the person deceived by Descartes’ demon, or a brain in a vat, would have any beliefs about material objects at all. This has been argued, for example, by Hilary Putnam. If this is correct, the brain in a vat does not falsely believe the proposition “There are material tables.”, rather, when the brain in a vat says to itself the sentence “There are material tables”, the sentence does not refer to material tables. If this is granted, it becomes unclear whether the brain in a vat really does have a vast number of false beliefs. Presumably the same may hold of a person deceived by Descartes’ genius. Perhaps, if the brain’s beliefs are interpreted as being merely about its own experiences, its beliefs might be true. And so it seems to be at least an arguable thesis that the brain in a vat is not a counter-example to reliabilism at all. But the argument presented here would not appear to be confronted with that difficulty. The brain in a vat argument does not cast in to doubt the either the meaningfulness or the falsity of Wally’s beliefs.
- 14.
For a critical discussion of BonJour’s position, see for example Jose Zalabardo “BonJour, Externalism and the Regress Problem”, Synthese, 148, 1, pp.135–169. (2006)
- 15.
This is not to deny that we still need an understanding of how we can come to know, for example, analytic truths a priori. But such cases seem far less “mysterious” than would a priori knowledge of something like “Mount Everest is over 29,000 feet high.”
- 16.
See BonJour, In Defense of Pure Reason, (Cambridge University Press, 1998), pp.15–19.
- 17.
BonJour criticizes traditional rationalism by arguing that there are cases of propositions that certainly seemed to be supported by a priori insight, but which we now believe to be not just defeasible, but empirically defeasible. One example is Euclidean geometry. We now believe, not only that the axioms of Euclidean geometry might be false, but that there is empirical evidence that in the vicinity of heavy bodies such as the Sun they are in fact false.
BonJour also rejects the radical empiricism of Quine. For Quine, the evidence for apparently a priori assertions, such as those of logic and mathematics, is ultimately purely empirical. The assertions of logic and mathematics, for Quine, lie very close to the centre of our “web of belief”. As such, they are not vulnerable to falsification by any empirical observations we might make: only assertions close to the observational periphery are vulnerable to such falsification. Michael Devitt suggests that this apparent immunity to empirical falsification is the reason why we are (mistakenly, in Devitt’s view) inclined to regard the statements of logic and mathematics as a priori.
For BonJour, we do in fact have some positive reason to believe the laws of logic and mathematics. How might a Quinean do justice to this? For a (Quinean) radical empiricist, any evidence that accrues to the laws of logic and mathematics lying in the centre of the web must ultimately come from the observational periphery. But Quine also says the interior of the web is underdetermined by its observational periphery. The actual content of the interior of the web is determined, not just by empirical input, but also by considerations of simplicity and conservatism. But now it seems: if we are to have good reason to believe the contents of the interior of the web, we must also have some sort of reason to think the simplicity, for example, of our web is an epistemic point in its favour. What kind of reason might we have for this? It seems it cannot be purely empirical reason, for the reasons given in the previous section of this chapter. So BonJour concludes a radical empiricist Quinean cannot do justice to the idea we have reason for the (empirically undeterdetermined) contents of the interior of our web. So, the radical empiricist Quinean cannot do justice to the idea we have reason for the laws of logic and mathematics.
- 18.
By “plausible numbers” it is meant “numbers of crows that are in fact realistically likely to be in Geelong and in the rest of the world. So, there are presumably at least hundreds of crows in Geelong and at least millions elsewhere. Note also that this assumption does not in any question-begging way involve making an assumption that will assure us of our desired conclusion. The conclusion argued for here could, perhaps, be qualified to something like: “On the assumption that that there are “sufficiently many” crows in the world, it is rational to prefer enumerative induction.” Also, “sufficiently many” does not here have to mean “very many”. As the subsequent passage of the main text points out, the claim of the argument will still be true provided there is at least one crow in Geelong and one crow elsewhere.
- 19.
Perhaps strictly speaking the considerations just given only support the thesis that it is a priori highly likely that it is unlikely that we should have blindly chosen this.
- 20.
The expression “The Principle of Indifference” is due to J. M. Keynes A Treatise of Probability, (MacMillan and Co. 1921), especially Chapter IV “The Principle of Indifference”.
There are, of course, some refinements that can be made to the principle, for example, that the sum of the probabilities of P1, …, Pn cannot be greater than one, and if P1, …, Pn exhaust all the possibilities, the sum must be exactly one. But as these refinements do not concern the issues with which we are here concerned, we will ignore them.
- 21.
See van Fraassen Laws and Symmetry (Oxford Clarendon Press, 1989), esp. pp.307–309.
- 22.
For an overview, see Hájek, Alan, “Interpretations of Probability”, The Stanford Encyclopedia of Philosophy (Winter 2012 Edition), Edward N. Zalta (ed.), URL = <http://plato.stanford.edu/archives/win2012/entries/probability-interpret/>. Especially section 3.1.
- 23.
This objection is made by J. Meixner and G. Fuller against BonJour’s justification of induction. See Mexiner and Fuller “BonJour’s A Priori Justification of Induction” in Pre-Proceedings of the 26th International Wittgenstein Symposium. S. Kostenbauer (ed) (2008)., pp.227–2
- 24.
See, for example, Karl Popper The Logic of Scientific Discovery, p.364.
- 25.
Hans Reichenbach “The Pragmatic Vindication of Induction” reprinted in T. J. McGrew, Marc Alspector-Kelly and Fritz Alhoff The Philosophy of Science: An Historical Anthology (Willey-Blackwell, 2009), p.366–371.
- 26.
F. J. Clendinnen “Rational Expectation and Simplicity” in What? Where? When? Why? Australasian Studies in the History and Philosophy of Science edited by Robert McLaughlin (Springer, Dordrecht, 1982), pp.1–25.
- 27.
See Explaining the Success of Science (Acumen, 2014), esp. pp.70–81.
Author information
Authors and Affiliations
Rights and permissions
Copyright information
© 2018 Springer Nature Switzerland AG
About this chapter
Cite this chapter
Wright, J. (2018). The Skeptical Arguments Against Realism I: Inductive Skepticism. In: An Epistemic Foundation for Scientific Realism. Synthese Library, vol 402. Springer, Cham. https://doi.org/10.1007/978-3-030-02218-1_2
Download citation
DOI: https://doi.org/10.1007/978-3-030-02218-1_2
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-02217-4
Online ISBN: 978-3-030-02218-1
eBook Packages: Religion and PhilosophyPhilosophy and Religion (R0)