A MAXWELLIAN BELIEF STATE; EPISTEMIC IMPLICATIONS AND OBJECTIONS
Despite the lack of proof, Maxwell's own belief state with regard to his kinetic-molecular theory was a quite confident one, which might be characterized as follows:
1. He believed that molecules exist and that the independently warranted dynamical assumptions about them were true.
2. He believed that he was justified in so believing.
3. He believed that neither he nor anyone else had sufficient experimental evidence to demonstrate that the assumptions he was making in the theory are true.
Claims (1) and (2) about Maxwell can be supported by examining many of his published and unpublished writings in the 1870s, and not just the 1875 paper in question.[219] Claim (3) is clearly made in his 1875 paper.[220]
Let's call a belief state of the sort Maxwell was in (one satisfying (1 )-(3)) a “confident but less than perfect one” with respect to a hypothesis h (which I will abbreviate as CLP(h)). Now admittedly one can be in such a state without being justified in believing h. But my claim is that one can also be in such a state and be justified in believing that h is true. Suppose I own 85 percent of the tickets in a fair lottery, one ticket of which will be drawn at random, and I believe that I will win because I own 85 percent of the tickets. I am justified in believing this even if I haven't proved or demonstrated that I will win. Or suppose that I am a detective trying to solve a crime, and that I have a good deal of information that suspect number one is the perpetrator: the motive, means, and opportunity all fit, as do the descriptions of some witnesses. On the basis of these facts, I come to believe that this suspect is guilty— even though, let's say, not all the evidence fits exactly, and even though I need more direct and positive evidence for a court of law. In the sort of case I am imagining, I am justified in believing what I do, even if I cannot yet prove it.
In relevant respects, in 1875 Maxwell's belief state with regard to molecular theory was analogous to these. Now for some objections.Objection 1
In the lottery and detective cases, as well as in Maxwell's case, we need to distinguish what justification a person offers for his belief, on the one hand, from whether his belief is really justified, since (the opponent might say) in these cases the person in question doesn't really have sufficient evidence to be justified in his belief. One has this only if the justification is sufficient for knowledge. Although CLP(h)-states are possible, and someone in such a state may offer a justification for believing h, this is not sufficient for knowledge that h is true. Such a position has in fact been taken in epistemology by Jonathan Sutton, who distinguishes a “loose” and a “strict” sense of justification.[221] In a case such as my lottery example, he argues that although in a “loose” sense (which is used colloquially and is championed by most epistemologists) I am justified in believing that I will win, in a “strict” sense I am not, because I don't know that I will win. In a strict sense I am justified in believing only that I will probably win.
Reply
To do justice to Sutton's position one would need to carefully examine each of the epistemic arguments and advantages he offers for employing his “strict” sense of justification. Elsewhere I have drawn a distinction somewhat similar to his between “veridical” and “nonveridical” senses of expressions such as “good reason to believe,” “evidence,” and “sign or symptom of,” in which the veridical sense requires the truth of the hypothesis in question, and the nonveridical does not.[222] For purposes of this chapter it suffices to say that the claims about justified belief in the lottery, detective, and Maxwellian cases are being made in my “nonveridical” sense. They can also be made in Sutton's “loose” sense, which does not require either truth or knowledge of the truth.[223] Maxwell's general epistemic position with respect to molecular theory fits both descriptions.
It is one that other theoretical scientists are frequently in.[224]Objection 2
As has been noted, even in 1875 Maxwell recognized problems with the theory, including theoretical derivations of specific heat values that were incompatible with observed values, and the inability of the theory to explain various known properties of gases. Since there were such problems and since they caused Maxwell to have some doubts concerning the theory, how could Maxwell believe the theory, let alone have confidence in his beliefs?
Reply
Maxwell did not doubt that the theory, in its essentials, is true. He doubted that all the assumptions he was making were true, without being able to point to specific ones as being particularly dubious. And he believed that the theory had not yet been sufficiently developed to deal with the inconsistencies or with unanswered questions concerning electrical and certain other known properties of gases. But doubts of these sorts were not enough to shake his confidence that the fundamental ideas are correct and that these problems would be worked out.
Objection 3
Maxwell's 1875 paper is titled “On the Dynamical Evidence of the Molecular Constitution of Bodies.” Evidence need not provide proof. It need not even provide reasons sufficient for belief. All it has to do is supply some reasons for increasing one's degree of belief in the theory. And that is all that Maxwell was in fact doing or was justified in claiming to do.
Reply
This objection presupposes an increase-in-degree-of-belief (or probability) position on evidence, which I have criticized elsewhere[225] and will not pursue here. Suffice it to say that, on Maxwell's view, one can give evidence that is strong enough to justify belief, and goes beyond simply increasing one's degree of belief, without giving proof; and that is precisely what he was trying to do in his 1875 paper and in other writings during this period.
6.