Abstract
A potential explanation of a fact is a hypothesis such that, if it were true, it would explain the fact in question. Let’s suppose that we become aware of a fact and some potential explanations thereof. Let’s also suppose that we would like to believe the truth. Given this aim, we can ask two questions. First, is it likely that one of these hypotheses is true? Second, given an affirmative answer to the first question, which one is it likely to be? Inference to the best explanation (IBE) offers answers to both questions. To the first, it says ‘Yes’—assuming that at least one of the hypotheses would, if true, provide a satisfactory explanation of the fact under consideration. To the second, it says that the hypothesis most likely to be true is the one that scores best on the explanatory virtues: conservatism, modesty, simplicity, generality, and predictive power. Many philosophers have argued against IBE’s answer to the first question. I am interested in an objection to its answer to the second. Many philosophers seem to think that it is unsustainable: they seem to think that even if we assume that one of the competing hypotheses is true, we should not think that IBE will help us to identify it. Or, more carefully, if these philosophers are doing what they appear to be doing—namely, offering critiques of IBE that don’t depend on assumptions about the field of competing hypotheses—then their claim is that IBE will not help us to identify the truth. I believe that this is mistaken: the argument for believing it assumes a model of IBE that we have no reason to accept.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
See Lipton (2004, Chapt. 9) for discussion of this two-stage conception of IBE, plus a defense of an affirmative answer to the first question. For a different defense, see Psillos (2002). The clause about explanatory quality insures that IBE does not permit a person to believe any hypothesis—however bizarre—as long as it’s the best of her options.
First, note that this particular list is due to Quine and Ullian, with ‘generality’ substituted for the more awkward ‘fecundity’ (Quine and Ullian 1978, pp. 64–82). It is not unusual; see Lycan (1988, p. 130) and Lipton (2004, p. 122). Second, note that I will discuss the virtues as though they are non-relational properties of hypotheses; this simplifies my exposition, and I don’t think that anything turns on it. Third, note that I distinguish between generality and explanatory power. Generality is a measure of the scope of the hypothesis—i.e., the number of phenomena to which it applies. Explanatory power is a measure of the overall virtuousness of a given hypothesis, represented by an overall score on the virtues. So I’ll say that hypothesis \(h_{1}\) has greater explanatory power than hypothesis \(h_{2}\) just in case \(h_{1}\) has a better overall score than does \(h_{2}\).
We might worry that it isn’t fair to saddle the critics with the ceteris paribus clause. If, for example, they expect the virtues to co-vary positively, then they might not insist that a change in an individual virtue should have some bearing on the relative probability of two hypotheses. But if that were their view, then it wouldn’t make sense to question the credentials of any particular virtue—which is precisely their strategy. After all, if you think that predictive power is indicative of truth, and that simplicity is indicative of predictive power, then you wouldn’t complain that, for all we know, “[n]ature is lush, prodigal, messy, wasteful, sexy, etc.” (Gunner 1967, pp. 4–5). Rather, you’d hunt for cases in which hypotheses score high on predictive power and low on simplicity, and hope to find enough to undermine any alleged link between the two. So either the virtues co-vary positively, or they don’t. If they do, then the critics need to offer new criticisms. If they don’t, then there is no reason to balk at the ceteris paribus clause.
In other words, no virtue is, all on its own, a reliable indicator truth. Of course, many philosophers insist that the individual virtues are truth-conducive. Richard Swinburne, for example, contends that “it is
Footnote 8 continued
a fundamental a priori principle” that simpler theories are more likely to be true than are more complex ones (Swinburne 2001, p. 102). And there are a number of less radical defenses of simplicity: e.g., Quine and Ullian (1978), Sober (1981), and Kelly (2007).
These are not actually Einstein’s words. The original phrasing is to be found in a letter that Einstein wrote to Max Born in 1926: “Quantum mechanics is certainly imposing, but an inner voice tells me that it is not yet the real thing. The theory says a lot, but does not really bring us any closer to the secret of the ‘old one’. I, at any rate, am convinced that He is not playing at dice” (Einstein et al. 1971, p. 91).
I am not saying that there is no way for the proponent of the Simple Additive Model to avoid saying that Einstein reasoned correctly: you never have to say anything in philosophy. My claim is just that our judgment about Einstein fits better with an instance of the Context-Sensitive Schema than it does with the Simple Additive Model. That much, I think, is fairly plausible.
If you prefer, you can generate the (Rough) Lakatosian Model quite easily; it will serve just as well for present purposes. However, one advantage of the (Rough) Kuhnian Model is that we might regard it as a way to clarify how scientists strikes the balance between dogmatism and openness that, arguably, Kuhn took to characterize science and explain its success.
Day and Kincaid are not clear about the relation between decontextualized virtues and their contextualized counterparts, so the relation may be weaker still.
There are other major differences between my proposal and Day and Kincaid’s: e.g., they have a very different picture of the structure of IBE. Recall that I understand the competing hypotheses as potential explanations: each one must be such that, if it were true, it would explain the data. We then evaluate the hypotheses based on their scores on the virtues. But Day and Kincaid don’t think about IBE this way: on their view, there are various reasons that you can offer in favor a view, of which being explanatory is only one. So, they see being explanatory as a reason that competes with the reasons generated by the explanatory virtues. On their view, the best explanation might not be recommended by our total evidence; on the view that I’ve sketched, this can’t happen.
If it were dubious whether one of the competing hypotheses is likely to be true, then this jump would not be as problematic: in this situation, we would be relying on the virtues to make up for potential deficiencies in the field. But we are supposing that the critics are offering an independent challenge to IBE, and therefore a challenge that grants that the truth is likely to be among the competing hypotheses.
This is not to suggest that there is no stage of the game at which corroboration matters.
For further considerations in favor of particularism, see Lemos (2004).
Someone might object that the G. E. Moore shift allows people to preserve beliefs that are clearly unreasonable. Suppose that Miss Cleo (a psychic) tells Tom that he will be lucky in love this week. Tom believes her and runs to tell his friends the good news. They are less excited than he had hoped. They suggest that his belief is reasonable only if (a) Miss Cleo offers falsifiable predictions and (b) those predictions are reliable. Since she fails on both counts, Tom should not hold out hope for a romantic encounter. Tom replies by making the G. E. Moore shift: he reasonably believes that Miss Cleo is right, but his belief does not meet his friends’ conditions; so, those conditions are not necessary for reasonable belief. Now, I agree that Tom has made a mistake in this situation. However, it’s not clear to me that the mistake is making the G. E. Moore shift. Suppose that it really is more plausible to Tom that Miss Cleo is correct than that his friends have hit on the right criteria for reasonable belief in her testimony. In such circumstances, what else is he supposed to do? Confidence is usually correlated with high subjective probability; so, by Tom’s lights, it is more probable that Miss Cleo is on target. And if he has that belief, then surely he would be irrational to opt instead for claims that he judges to be less probable. Of course, Tom is wrong to assign such a high subjective probability to Miss Cleo’s testimony. But I have yet to see a defense of the view that IBE arguments are on par with the testimony of self-proclaimed psychics, nor even that they are in the same neighborhood. Arguably, there are (virtually) no contexts in which psychics (qua psychics) provide you with good reasons to believe that p. Things are not nearly so grim for IBE.
Thanks to Walter Edelberg, Bill Hart, and Colin Klein for comments on the early drafts of this paper; thanks to Burkay Ozturk and an anonymous reviewer for remarks on the later ones.
References
Benacerraf, P., & Putnam, H. (Eds.). (1983). Philosophy of mathematics : Selected readings (2nd ed.). New York: Cambridge University Press.
Bueno, O., & Shalkowski, S. (2004). Modal realism and modal epistemology: A huge gap. In E. Weber & T. De Mey (Eds.), Modal epistemology (pp. 93–106). Brussels: Royal Flemish Academy of Belgium.
Chisholm, R. (1973). The problem of the criterion. Milwaukee: Marquette University Press.
Cornman, J. (1980). Skepticism, justification, and explanation. Dordrecht: Reidel.
Dawes, G. (2013). Belief is not the issue: A defense to the best explanation. Ratio, 26, 62–78.
Day, T., & Kincaid, H. (1994). Putting inference to the best explanation in its place. Synthese, 98, 271–295.
Devitt, M. (2011). Are unconceived alternatives a problem for scientific realism? Journal for General Philosophy of Science, 42, 285–293.
Einstein, A., Born, M., & Born, H. (1971). The Born–Einstein letters: Correspondence between Albert Einstein and Max and Hedwig Born from 1916–1955, with commentaries by Max Born. London: Macmillan.
Fraassen, V., & Bas, C. (1980). The scientific image. New York: Oxford University Press.
Fraassen, V., & Bas, C. (1989). Laws and symmetry. New York: Oxford University Press.
Fumerton, R. (1980). Induction and reasoning to the best explanation. Philosophy of Science, 47(4), 589–600.
Goodman, N. (1979). Fact, fiction, and forecast (4th ed.). Cambridge: Harvard University Press.
Gunner, D. L. (1967). Professor Smart’s ‘Sensations and brain processes’. In The identity theory of mind. St. Lucia: University of Queensland Press.
Hacking, I. (1984). Experimentation and scientific realism. In J. Leplin (Ed.), Scientific realism (pp. 154–172). Berkeley: University of California Press.
Harman, G. (1968). Enumerative induction as inference to the best explanation. Journal of Philosophy, 65(18), 529–533.
Josephson, J. R., & Josephson, S. G. (Eds.). (1996). Abductive inference: Computation, philosophy, technology. New York: Cambridge University Press.
Kelly, K. (2007). A new solution to the puzzle of simplicity. Philosophy of Science, 74(5), 561–573.
Lemos, N. (2004). Common sense : A contemporary defense, Cambridge studies in philosophy. New York: Cambridge University Press.
Lipton, P. (2004). Inference to the best explanation (2nd ed.). International library of philosophy. New York: Routledge.
Lycan, W. (1988). Judgement and justification. New York: Cambridge University Press.
Marinov, M. (1988). Inference to the best explanation: Van Fraassen and the case of the ‘Fifth Force’. International Studies in the Philosophy of Science, 3(1), 35–50.
Moore, G. E. (1963). Four forms of scepticism. In Philosophical papers. New York: Collier Books.
Moser, P., Mulder, D., & Trout, J. D. (1998). The theory of knowledge: A thematic introduction. New York: Oxford University Press.
Psillos, S. (1996). On Van Fraassen’s critique of abductive reasoning. The Philosophical Quarterly, 46, 31–47.
Psillos, S. (1999). Scientific realism : How science tracks truth, Philosophical issues in science. London: Routledge.
Psillos, S. (2002). Simply the best: A case for abduction. In A. C. Kakas, & F. Sadri (Eds.), Computational logic: From logic programming into the future, LNAI 2408 (pp. 605–625). Berlin: Springer.
Quine, W. V., & Ullian, J. S. (1978). The web of belief (2nd ed.). New York: Random House.
Rowe, W. (1979). The problem of evil and some varieties of atheism. American Philosophical Quarterly, 16(4), 335–341.
Shalkowski, S. (2010). IBE, GMR, and metaphysical projects. In B. Hale & A. Hoffmann (Eds.), Modality : Metaphysics, logic, and epistemology (pp. 167–187). New York: Oxford University Press.
Sober, E. (1981). The principle of parsimony. British Journal for the Philosophy of Science, 32(2), 145–156.
Stanford, P. K. (2006). Exceeding our grasp : Science, history, and the problem of unconceived alternatives. New York: Oxford University Press.
Swinburne, R. (2001). Epistemic justification. New York: Oxford University Press.
Thagard, P. (1978). The best explanation: Criteria for theory choice. The Journal of Philosophy, 75(2), 76–92.
Thagard, Paul. (2006). Evaluating explanations in law, science, and everyday life. Current Directions in Psychological Science, 15(3), 141–145.
Weintraub, R. (2012). Induction and inference to the best explanation. Philosophical Studies doi:10.1007/s11098-012-0034-1.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Fischer, R.W. Why it doesn’t matter whether the virtues are truth-conducive. Synthese 191, 1059–1073 (2014). https://doi.org/10.1007/s11229-013-0309-x
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11229-013-0309-x