STRENGTHENED JUSTIFICATORY HOLISM
The strengthened view needs to be (a) justificatory, (b) holistic, and (c) abstract in the sense that justification is not tied to particular contexts in which particular challenges are made.
The idea is that while a particularist claim of the form “e is evidence that h” may be true even if neither e nor h is “holistic,” a justification for that claim must satisfy (a)-(c). Can there be such a thing?I will discuss a very strong proposal, with the idea that if this doesn't yield justificatory holism, then perhaps nothing will. The proposal is that for a system of beliefs ES to adequately justify an evidential claim of the form
(1) e is (veridical) evidence that h,
one who is in ES needs to be able to meet not just some particular challenge to (1) but all possible ones. And for this purpose, since possible challenges to (1) are so numerous and varied, all parts of ES will need to be invoked to meet them all. True, some particular challenges to (1) may be met by appeal to particular parts of ES. But all possible challenges require the entire system ES. And anyone in ES who cannot meet all possible challenges to (1) by appealing to ES is not justified in believing (1).
Even if this were to yield justificatory holism (which, I will argue, it doesn't), it is an impossible standard of justification! To be sure, there are possible challenges to (1) that can be met by appeal to the set of beliefs in a particular ES. But there will always be possible challenges that cannot be met—challenges based on information not available to those in ES. For example, on the present proposal, the epistemic situation ES of wave theorists of light in the mid-nineteenth century would not justify the evidential claim that
(2) Diffraction and interference phenomena constitute veridical evidence that the wave theory is correct.
This is because, although it was not available to midnineteenth-century wave theorists in their ES, one of the “possible challenges” (which became actual at the end of the nineteenth century) derives from the results of the Michaelson-Morley experiments, demonstrating the absence of an ether drag.
Of course, had mid-nineteenthcentury ether theorists known about and accepted the results of these experiments, their epistemic situation would have changed. But, given the holist claim I am considering, in the mid-nineteenth century, they could not have known about these experiments, and if they had, they would not have been able to defend claim (2) by reference to their original epi- stemic situation.The present holistic proposal leads to an interesting form of skepticism. Assume, as in the case above, that someone in a given epistemic situation ES at a time t cannot know about future experiments and observations that will be made well after t that may yield results that refute or at least challenge beliefs in ES. 'lhere will be possible challenges to an evidential claim such as (2) that one in ES cannot meet by appeal to ES. If so, then on the present version of holism, no matter what one's epistemic situation, since one will be unable to meet all such challenges, one will not be justified in believing such evidential claims. I say that this is an interesting form of skepticism, because even though it may be true that e is veridical evidence that h, and even though one may come to know that e is true, one can never be justified in believing that e is veridical evidence that h. I take it that on an evidential view of “knowing that h,” it is required that one be justified in believing that there is some e that is veridical evidence that h. But this is precluded by the present version of justificatory holism.
Most important, for present purposes, even if we were to grant that justification requires a belief set ES that can meet all challenges, we still would not get justificatory holism. All we get is that an evidential claim of form (1) is justified, relative to an ES, only if all possible challenges to (1) can be met by reference to the beliefs in ES. But this doesn't require holism rather than particularism. It would be satisfied if each possible challenge to the evidential claim can be met by reference to one or more beliefs in ES.
Perhaps some challenges will require most of the beliefs in ES, others fewer. Just because all the beliefs in ES may be needed to justify all the possible challenges, it doesn't follow that all of them are necessary to justify each challenge. It doesn't follow that each possible challenge needs to be met by appeal to all the beliefs in ES. If not, then justification is not holistic.What I have just said applies also to weaker versions of justificatory holism that say not that one who is in ES needs to be able to meet all possible challenges to an evidential claim by appeal to the beliefs in ES but only some possible challenges of certain designated types. As with the stronger version, even if all the beliefs in ES were required to justify all the challenges in this restricted set of challenges, it doesn't follow that all the beliefs in ES are needed to justify each challenge. Again, we do not get justificatory holism.
Finally, as noted earlier, the justificatory holist wants: (a) justification, (b) holism, and (c) abstraction from particular contexts and challenges. With justificatory particularism, using my concepts of evidence, one can get (a) and (c). As far as (a) is concerned, if e is ES-evidence that h, then anyone in ES is justified in believing that e is veridical evidence that h, and hence that h is true. As far as (c) is concerned, if e is potential or veridical evidence that h, then e is a good reason to believe h, independent of any particular contexts or challenges. What the particularist will not give the holist is what the latter most wants—viz. (b), holism. He will not say that something can be evidence only for an entire system of hypotheses, and he will not say that an evidential claim is justified only by reference to an entire system of beliefs but not to particular beliefs in that system.
11.