Abstract
We often think or say that someone was wrong about something but almost right about it or close to the truth. This can mean more than one thing. Here, I propose an analysis of the idea of being epistemically close to the truth. This idea plays an important role in our practice of epistemic evaluation and therefore deserves some detailed attention. I start (section1) with an exposition of the idea of getting things right by looking at the main forms of reliabilism about true belief and belief acquisition. The focus on reliabilism is justified because (almost) everyone is a reliabilist in a basic sense. Section 2 develops a notion of closeness to the truth in two steps. Section 3 mentions some ways in which this notion is useful, one having to do with the Gettier problem.
This is a preview of subscription content, log in via an institution to check access.
Similar content being viewed by others
Notes
For related issues see, e.g., Sorensen 2016.
I will thus not go much at all into the topic of truthlikeness (see, e.g., Oddie 2016). This notion can be explained or construed in several different ways the details of which need not concern us here. But one can give some ostensive definitions or examples. For instance, an utterance of “It is 3 pm now” when it is only 2:59:32 pm at the moment of utterance is closer to the truth than an utterance of “It is 1:30 pm now”. Or, if Newton’s theory is not strictly true, it is (so it seems) much closer to the truth about laws of motion than Aristotle’s views on the subject. Truthlikeness is thus a semantic notion having to do with “closeness” or similarity between contents; it is not an epistemic notion concerning our relation to such contents. It is important to keep these different meanings of “close to the truth” apart.
This is a very basic and rough form of the view but we need not concern us with whistles and bells here. The same holds for all the other conditions and explanations that I will present below. – If safety requires truth of the belief in the actual world, then we can also simply say that safe belief is knowledge.
We can and need to leave aside questions concerning the individuation of kinds of beliefs.
According to Nozick (see 1981, 172–196), a true belief is knowledge (if and) only if it meets two conditions, a variation and an adherence condition:
(Belief-Sensitivity-v) A true belief that p is sensitive-v just in case in all (or most) close possible worlds in which not p, the subject (using the same method as in the actual case) doesn’t believe that p;
(Belief-Sensitivity-a) A true belief that p is sensitive-a just in case in all (or most) close possible worlds in which p, the subject (using the same method as in the actual case) does believe that p.
Again, one might want to add conditions of method sensitivity:
(Method-Sensitivity-v) A method used in the acquisition of a true belief that p is sensitive-v just in case in all (or most) close possible worlds in which p or a proposition of the same kind as p is false the subject does not, using that method, come to believe that proposition;
(Method-Sensitivity-a) A method used in the acquisition of a true belief that p is sensitive-a just in case in all (or most) close possible worlds in which p or a proposition of the same kind as p is true the subject does, using that method, come to believe that proposition.
I am using the term “sensitivity” like “tracking” and for both conditions, in contrast to Nozick’s use of the term. I am not aiming at the most correct interpretation of, say, Nozick here but rather at capturing the core of the idea of sensitivity.
This requires something different for knowledge:
(Belief-Reliability) A true belief that p, resulting from the use of a certain method M, is reliable just in case the probability of ending up with a true belief about whether p is high, given that one uses M.
Here is a corresponding condition for method-reliability:
(Method-Reliability) A method used in the acquisition of a true belief that p is reliable just in case use of the method for finding out whether p or a proposition of the same kind is true has a high probability of leading to a true belief about the matter.
It is possible that two methods M1 and M2 are both inversely method-safe but one (M1) is much “closer to the truth” in the sense of truthlikeness (see fn.2 above) than the other. For instance, my way of estimating spatial distance (M1) might be slightly off and systematically lead to a small deviation from the truth whereas my way of estimating the success of political candidates (M2, driven, say, by wishful thinking) systematically leads me to grossly incorrect beliefs about the matter. It is obvious that M1 is better in an important respect than M2. However, here I am not dealing with the virtues of truthlikeness (see above) but with the different virtue of being close to the truth in the sense discussed here. Thanks to a referee who pressed me on this point.
We can describe the above bean counting case also by using notions of inverse sensitivity:
(Inverse Belief-Sensitivity-v) A false belief that p is inversely sensitive-v just in case: In all (or most) close possible worlds in which p is true, the subject (using the same method as in the actual case) doesn’t believe that p;
(Inverse Belief-Sensitivity-a) A false belief that p is inversely sensitive-a just in case: In all (or most) close possible worlds in which not p is true, the subject (using the same method as in the actual case) does believe that p.
My false bean belief failed to meet either of these conditions. One can also characterize my false belief as failing to meet inverse method sensitivity conditions:
(Inverse Method-Sensitivity-v) A method used in the acquisition of a false belief that p is inversely sensitive-v just in case in all (or most) close possible worlds in which p or a proposition of the same kind as p is true the subject does not, using that method, come to believe that proposition;
(Inverse Method-Sensitivity-a) A method used in the acquisition of a false belief that p is inversely sensitive-a just in case in all (or most) close possible worlds in which p or a proposition of the same kind as p is false the subject does, using that method, come to believe that proposition.
Finally, we can also express things in terms of inverse reliability:
(Inverse Belief-Reliability) A false belief that p, resulting from the use of a certain method M, is inversely reliable just in case the probability of ending up with a false belief about whether p is high, given that one uses M;
(Inverse Method-Reliability) A method used in the acquisition of a false belief that p is inversely reliable just in case use of the method for finding out whether p or a proposition of the same kind is true has a high probability of leading to a false belief about the matter.
Here are the parallel conditions for the other forms of reliabilism:
(Sensitive-v Belief-Closeness) A false belief that p is sensitively-v close just in case: In all (or most) close possible worlds in which p is true, the subject (using the same method as in the actual case) does believe that p;
(Sensitive-a Belief-Closeness) A false belief that p is sensitively-a close just in case: In all (or most) close possible worlds in which not p is true, the subject (using the same method as in the actual case) does believe that not p;
(Sensitive-v Method-Closeness) A method used in the acquisition of a false belief that p is sensitively-v close just in case in all (or most) close possible worlds in which p or a proposition of the same kind as p is true the subject does, using that method, come to believe that proposition;
(Sensitive-a Method-Closeness) A method used in the acquisition of a false belief that p is sensitively-a close just in case in all (or most) close possible worlds in which p or a proposition of the same kind as p is false the subject does, using that method, come to believe the negation of that proposition;
(Reliable Belief-Closeness) A false belief that p, resulting from the use of a certain method M, is reliably close just in case the probability of ending up with a true belief about whether p is high, given that one uses M;
(Reliable Method Closeness) A method used in the acquisition of a false belief that p is reliably close just in case use of the method for finding out whether p or a proposition of the same kind is true has a high probability of leading to a true belief about the matter.
This requires a good metric for closeness of possible worlds which, some say, is very hard to get (if at all possible).
What if Smith is confused about Jones or Fords (say, typically mistakes Jones for Miller and Fords for Buicks)? Is he still thinking or talking about Jones or some Ford? We can leave this complicated question aside here since it doesn’t affect the main point here and only complicates our view of Gettier cases. Thanks to a referee here.
It is not clear at all whether Smith is safely or sensitively belief-close. – In the case of reliabilism, one could also, arguably, use belief-closeness as a condition. I am leaving this complication aside here.
References
Baumann, P. (2016). Epistemic contextualism. A defense. Oxford: Oxford University Press.
Bernecker, S. (forthcoming). The problem of explaining knowledge from falsehood.
Dretske, F. I. (1971). Conclusive reasons. Australasian Journal of Philosophy, 49, 1–22.
Gettier, E. L. (1963). Is justified true belief knowledge? Analysis, 23, 121–123.
Goldman, A. I. (1992). Discrimination and perceptual knowledge. In A. I. Goldman (Ed.), Liaisons. philosophy meets the cognitive and social sciences (pp. 85–103). Cambridge: MIT Press.
Nozick, R. (1981). Philosophical explanations. Cambridge: Harvard University Press.
Oddie, G. (2016). Truthlikeness. The Stanford Encyclopedia of Philosophy (Winter 2016 Edition), Edward N. Zalta (eds.), URL = https://plato.stanford.edu/archives/win2016/entries/truthlikeness/.
Russell, B. (1948). Human knowledge: Its scope and limits. New York: Simon and Schuster.
Sorensen, R. (2016). Fugu for Logicians. Philosophy and Phenomenological Research, 92, 131–144.
Sosa, E. (1999). How to defeat opposition to Moore. Philosophical Perspectives, 13, 141–154.
Warfield, T. A. (2005). Knowledge from falsehood. Philosophical Perspectives, 19, 405–416.
Acknowledgements
Thanks to Sven Bernecker for conversations on this topic and to referees for comments on earlier versions of this paper.
Author information
Authors and Affiliations
Corresponding author
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Rights and permissions
About this article
Cite this article
Baumann, P. Close to the Truth. Philosophia 48, 1769–1775 (2020). https://doi.org/10.1007/s11406-020-00201-w
Received:
Revised:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11406-020-00201-w