Modal epistemology

Philosophical Studies

Volumn 92, Issue 1-2, 1998, Pages 67-84

  Van Inwagen, P ^a  

^a UNIVERSITY OF NOTRE DAME   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 54749087066     PISSN: 00318116     EISSN: 15730883     Source Type: Journal    
DOI: None     Document Type: Article

References (20)

1. But philosophers have employed many formally valid modal arguments that cannot be so described. Here is a trivial example: it is possible for there to be cases of justified true belief that are not cases of knowledge; hence, knowledge is not justified true belief. Kripke's arguments for the essentiality of origins or the impossibility of unicorns might provide non-trivial examples - at least for those who were willing to say that these arguments were formally valid or could easily be made so.
2. Or, of course, an assertion of non-necessity - since 'it is possible that p' is equivalent to 'it is not necessary that not-p'.
3. For example: we know that there could be a full-scale papier-mâché mock-up of a barn that looked like a real barn from a distance, or that the legs and top of this table might never have been joined to one another.
4. It is plausible to suppose that one can learn from the testimony of others what one could not learn by the exercise of one's own unaided powers. It would be therefore consistent with my thesis for me to affirm, say, that I knew that a perfect being was possible because God existed and had informed me that He was a perfect being - or (to anticipate an example that I shall later discuss) that I knew that transparent iron was possible because the Wise Old Beings from the Center of the Galaxy had assured me that their physics (which surpasses human understanding) had demonstrated this possibility.
5. God, Knowledge, and Mystery: Essays in Philosophical Theology (Ithaca and London: Cornell University Press, 1995), pp. 11-14.
6. Suppose, for example, that we know that it is not possible for water to be a different physical stuff from the physical stuff it is - that no other physical stuff would be water (an example, perhaps, of "basic" modal knowledge); and suppose we know that water is the physical stuff composed of molecules formed by joining a hydrogen atom to hydroxyl radical (a "fact about how the world is put together"); then - or so at least many have argued - we can validly conclude that water is essentially hydrogen hydroxide. For an interesting discussion of the ways in which reason, starting with a stock of "basic" modal knowledge, can extend our modal knowledge, see Phillip Bricker, "Plenitude of Possible Structures," The Journal of Philosophy LXXXVIII (1991), pp. 607-619.
7. "Materialism and the Psychological Continuity Account of Personal Identity," forthcoming in Philosophical Perspectives;
8. review of Richard Swinburne, The Coherence of Theism, The Philosophical Review 88 (1979), pp. 668-672.
9. If there is logical impossibility, there is also logical necessity, for a state of affairs is logically necessary if and only if its negation is logically impossible. The logically necessary is that which can be seen to be necessary on the basis of logical (or semantical) considerations alone.
10. What I have called possibility without qualification, some have called "absolute" or "intrinsic" or "ontological" or "metaphysical" possibility. The first two seem good enough names. I don't find "ontological" or "metaphysical" particularly appropriate tags, however. I don't think that unqualified or absolute or intrinsic possibility is in any clear sense an ontological or metaphysical concept. An analogy is perhaps provided by "truth without qualification" (as opposed, for example, to scientific, metaphorical, approximate, or contingent truth). One might call "truth without qualification" ontological or metaphysical truth, but these wouldn't be particularly appropriate tags.
11. See his "Some Difficulties in Theistic Treatments of Evil," in Daniel Howard-Snyder, ed., The Evidential Argument from Evil (Bloomington and Indianapolis: Indiana University Press, 1996), pp. 206-218.
12. That Hesperus is necessarily identical with Phosphorus can be established by the joint application of the theorem 'x = y → □ x = y' and the semantical thesis that 'Hesperus' and 'Phosphorus' are "rigid designators" - or at least it can if that semantical thesis has itself been "established."
13. "Is Conceivability a Guide to Possibility," Philosophy and Phenomenological Research LIII (1993), pp. 1-39.
14. Yablo rightly distinguishes objectual from propositional imagining. (Consider, for example, the difference between "imagining a tiger" and "imagining that one encounters a tiger.") The kind of imagining that figures in his epistemological thesis is objectual imagining. Perhaps, therefore, it would be better to state his thesis in terms of "imagining a cosmos" or "imagining a universe," since most philosophers - David Lewis, of course, is the notorious exception - take "possible worlds" to be states of affairs or other such proposition-like entities.
15. Compare my discussion of the "green cheese" case in the review of Swinburne's The Coherence of Theism cited in n. 7.
16. There is a wonderful article about this that I never miss a chance to commend: George Seddon, "Logical Possibility," Mind 81 (1972), pp. 481-94.
17. See also my "Ontological Arguments," Noûs 11 (1977), pp. 375-95.
18. (Reprinted in the collection cited in note 5.) This essay contains a discussion of the "transparent iron" case. See also Paul Tidman, "Conceivability as a Test for Possibility," American Philosophical Quarterly 31 (1994), pp. 297-309.
19. As to the last of these: would it not do simply to imagine a world in which vast numbers of innocent sentient creatures fry on red-hot griddles at every moment at which they exist? But this state of affairs is incompatible - this modal term is to be understood in its "absolute" sense - with the existence of an omniscient, omnipotent, morally perfect being. (Or let's assume so. That some state of affairs involving absolutely inexplicable suffering is incompatible with the existence of such a being is the other premise of the possibility argument whose crucial modal premise is the possibility of such a state of affairs.) To imagine a world in which this state of affairs obtains, therefore, is to imagine a world in which there is no such being. And what would justify one in believing that one had imagined a world in which there was no such being? To imagine an enormous griddle on which vast numbers of innocent creatures are tortured pointlessly is not to imagine a world, a whole coherent reality. The absence of objects "external" to this situation but absolutely incompatible with it must somehow be a part of one's imaginings if one is to have imagined a world in which it obtains. An omnipotent and omniscient being, or so I should imagine, would have to be invisible and omnipresent. How does one go about imagining the absence of an invisible and omnipresent being - as opposed to failing to imagine the presence of such a being? I see no answer to this question. The suggestion that one can imagine an enormous griddle on which vast numbers of innocent creatures are tortured pointlessly is thus very like Hume's contention that one can imagine something's coming into existence without a cause. It is easy enough to imagine something's coming into existence and not to imagine a cause of its coming into existence; it is something else, and rather more difficult, to imagine something's coming into existence and also to imagine the absence of any cause of its coming into existence. But this is only the first word on this topic, not the last. The issues the example raises are too complex to be resolved in a single footnote. As things stand, it might well be objected that if the argument of the preceding paragraph were correct, a parallel argument would show that one couldn't imagine a chair's having been two feet to the left of where it was at noon today, since one couldn't imagine, say, the absence of an omnipotent being who, for some location, decreed in every possible world in which the chair existed that it be in that location at noon today. This is a very good point, but to address it, I should need to leave the topic of modal epistemology and discuss the pragmatics of argument, the purposes for which arguments are offered and the occasions on which various dialectical "moves" are proper. A hint: although I know that the chair could have been two feet to the left of where it was at noon (and therefore know that Spinoza was wrong), it would have dialectically improper of me to present this piece of modal knowledge to Spinoza - or to someone who really did believe that God assigned to each physical object a spacetime trajectory that was invariant across possible worlds - and to have claimed thereby to have refuted his position.
20. Stated with more care, my thesis is as follows. Consider those propositions whose truth-values cannot be determined by logic and reflection on the meanings of words or by the application of mathematical reasoning. Among those, consider those whose truth-values are unknown to us or which are known to be false. If the only way to determine whether a proposition in this category is possible is by attempting to imagine a world we take to verify this proposition, then we should be modal skeptics: while we shall certainly know some propositions of this type to be possible, we shall not be able to know whether the premises of our illustrative possibility arguments are true; and neither shall we be able to know whether it is possible for there to be transparent iron or naturally purple cows.