My disembodied existence is conceivable (thinkable without apparent logical contradiction by me and beings like me). But does it follow that my disembodied existence is possible? Sydney Shoemaker floats the suggestion that this inference is invalid, resting as he thinks on a confusion of epistemic with metaphysical possibility. (Identity, Cause, and Mind, p. 155, n. 13.) Shoemaker writes, "In the sense in which I can conceive of myself existing in disembodied form, this comes to the fact that it is compatible with what I know about my essential nature . . . that I should exist in disembodied form. From this it does not follow that my essential nature is in fact such as to permit me to exist indisembodied form."
We need to think about the relation between conceivability and epistemic possibility if we are to get clear about the inferential link, if any, between conceivability and metaphysical possibility. Pace Shoemaker, I will suggest that the inference from conceivability to metaphysical possibility need not rest on a confusion of epistemic with metaphysical possibility. But it all depends on how we define these terms.
1. What is Epistemic Possibility?
In ordinary English, epistemic uses of 'possible' are rife. I inquire, "Is Jones in his office?" The secretary replies, "It's possible." I am not being informed that Jones' presence in his office is consistent with the laws of logic, or with the laws of nature; there is no question about the logical or nomological possibility of Jones' being in his office. I am being informed that Jones' presence in his office is consistent with what the secretary knows: it is not ruled out by anything she knows. It's possible for all she knows. Of course, if the secretary knows that Jones is in his office, or knows something that (she knows) entails that he is in his office, then Jones' presence in his office will be logically consistent with what she knows; but in that case she will not say that it is possible that he be there. She will say, "He's there." So 'possible' in its epistemic use conveys both consistency with what one knows and ignorance. When I say that such-and-such is epistemically possible, I am saying that it is possible for all I know, but I don't know all about the matter in question. Letting 'S' range over states of affairs and 'P' over persons, we define
D1. S is epistemically possible for P =df (i) S is logically consistent with what P knows; (ii) S is neither known by P nor known to be entailed by anything P knows.
The reason for clause (ii) is that epistemic uses of 'possible' indicate ignorance. 'It's possible that Jones is in his office,' said by the secretary implies that she does not know whether or not he is in his office. If she knew that he was in his office, and said what she said, then she would not be using 'possible' in the epistemic way it is used in ordinary English.
It is clear that my existing disembodied is a state of affairs consistent with what I know and also one that is neither known by me nor known to be entailed by anything I know. So my existing disembodied is epistemically possible for me.
2. What is Ontological (Real, Non-Epistemic) Possibility?
What is possible but not epistemically let us call 'ontologically possible' and define as follows:
D2. S is ontologically possible =df S is objectively or absolutely possible.
The possibility of the ontologically possible does not vary with any finite subject's knowledge or ignorance. Clearly, if S is epistemically possible it does not follow that S is ontologically possible. Suppose Jones dies on the way to his office for his scheduled office hours. If the secretary is unaware of this development, then it will be epistemically possible for her that Jones be in his office despite the objective impossibility of his being there.
Ontological possibility subsumes both nomological and broadly logical (metaphysical possibility).
3. What is Conceivability?
We now define 'conceivable':
D3. S is conceivable for P =df (i) P entertains S attentively; (ii) P finds no logical contradiction in S.
Conceivability as defined does not entail ontological possibility. Reasons why were given earlier. It follows that conceivability does not entail any type of ontological possibility such as metaphysical (broadly logical) possibility. But conceivability may provide defeasible evidence of metaphysical possibility. But this is not our present topic.
4. The Difference Between Epistemic Possibility and Conceivability
Our present topic is the difference between conceivability and epistemic possibility. That they are different can be seen if we reflect on propositions which, if true, are necessarily true. The strong or binary Goldbach Conjecture is that all positive even integers greater than or equal to 4 can be expressed as the sum of two primes. This conjecture is such that either it or its negation is necessarily true. But we don't know which. Yet both are conceivable. I bring before my mind the proposition, attend to it carefully, and find no logical contradiction. Then I bring the negation of the proposition before my mind, attend to it carefully, and find no logical contradiction. I conclude that both the proposition and its negation are conceivable. But only one of them is epistemically possible. For only one of them is consistent with what I know. If the Goldbach Conjecture is true then, because it is necessarily true, it is consistent with what I know but its negation (since necessarily false) is not, and vice versa.
Conceivability and epistemic possibility are distinct for a second reason. What I know to be actual is conceivable to me, but what I know to be actual is not epistemically possible for me. (I am of course assuming the adequacy of D1, something that some will contest.) Since I know that I exist, it is conceivable to me that I exist; but it is not epistemically possible for me that I exist. On the contrary, it is epistemically necessary for me that I exist! For that I exist is entailed by something I know, namely, that I exist. Clause (i) of (D1) is satisfied, but not clause (ii): it is not possible for all I know that I exist, for I do know that I exist. From 'I exist,' I can infer 'It is possible that I exist' and 'It is conceivable that I exist.' But I cannot infer 'It is epistemically possible that I exist.' For since I know with certainty that I exist, there is no room here for the ignorance which is an essential ingredient in epistemic possibility.
If S is epistemically possible for P, then P must be partially ignorant about S. For epistemic possibility feeds on ignorance. It is because of Smith's not knowing that Zorn's Lemma is equivalent to the Axiom of Choice that their nonequivalence is epistemically possible for him. There are no epistemic possibilities for an omniscient being or for one who knows all about a certain subject matter. But every ontologically possible state of affairs is conceivable to an omniscient being. For God, all things conceivable are possible, and conversely; but nothing is epistemically possible. Here again we see that epistemic possibility and conceivability 'come apart.'
My disembodied existence is both epistemically possible and conceivable. But because epistemic possibility and conceivability are distinct, one who takes conceivability as evidence of ontological possibility need not, pace Shoemaker, be confusing epistemic and metaphysical possibility. Nevertheless, conceivability does not entail ontological possibility.
Comments