Been writing a bit for my paper with Manuel on Holistic Knowledge. We are trying to defend the idea that knowledge ought to be holistic, and maybe any claim of knowledge about anything specific is bound to be no more than contrastive. Here is the bit on our take on Davidson's argument:
We believe Davidson has pointed towards an account of intelligibility according to which it is a necessary condition for a sufficiently large class of thoughts to be intelligible that it responds to the world.1 His argument arises from considerations concerning the holistic character of thoughts: they are intelligible only in critical masses. This amounts to say that only within a critical mass of thoughts we can say that a thought becomes interpretable and testable – it is in a critical mass that it acquires meaning and turns into something that can be true or false. Thoughts are only not self-standing units like atoms, but rather they depend on the whole where they belong – that is, they depend on the critical mass of thoughts where they are placed. Thoughts, and therefore also thinking, deal in critical masses. Davidson's holism – which finds its point of departure in Quine's rejection of the two dogmas that asserted an independence between meaning and empirical import – ties together thoughts that make other thoughts intelligible and those that result from an empirical engagement with the world: they are all part of a critical mass where no thought responds to the world in isolation.
We can begin to come to grip with Davidson's holistic account of thought and its connection to the world by considering some skeptical challenges and some usual ways to overcome them. The usual structure for a skeptical argument (but by no means the only one) is to sever the connection between thoughts and the rest of world. So, for instance, the skeptic suspects each of my beliefs and concludes that all (or most of) my beliefs could be false. The skeptical strategy is to assume that each of my beliefs are self-standing and can be doubted on independent grounds. Such skeptical challenge can be summarized in terms of an argument of illusion (AI)2 that can be presented as follows:
(P) (It is intelligible that) I can be wrong about each of my beliefs.
(C) Therefore, (it is intelligible that) I can be wrong about all of my beliefs.3
Now, on the face of it, there are two ways to counter AI. One can accept the inference and attempt to show that premise P is false. And one can question the validity of the inference. The first option has been the effort of most epistemological endeavors and often this is done by arguing that not anything can be wrong – which often amounts to seeking foundations that cannot be wrong. Most (if not all) foundationalist projects rely on the bottleneck picture; even though bottlenecks are not always foundations. In any case, it is clear that once accepted the inference, the conditions are present for taking intelligibility and responsiveness to the world to be independent and all is set for the bottleneck picture to look compulsory. If we can understand our beliefs while considering that all of them can be false, we will feel compelled to either accept a skeptical conclusion or to look for a way to ensure us that, for some reason, not all of our beliefs can be false and therefore P is false.
The second option to counter AI is to accept P and resist conclusion C by showing that the inference is not a valid one. If each of my beliefs can be wrong but not all of them, then not every belief can stand (or fall) on its own. At least some beliefs require other beliefs both to saddle them with content and to promote a contrast between them and the world. The line of argument to the effect that AI is invalid has to take beliefs as items that cannot be understood or individuated away from the critical mass of thoughts where they belong. Beliefs cannot considered in an isolation because each belief depends on others – no belief is an island. Each belief stands in the shoulders of their fellow beliefs. We shall refer to this rejection of AI as an invalid argument as the master claim of holism: the claim that my doubting each of my beliefs in isolation doesn't affect the critical mass of my beliefs because my doubts cannot get off the ground without the support of other beliefs within the critical mass. In general, we can formulate the master claim of holism as follows: severing the contact between each thought in isolation and the world doesn't affect the critical mass of thoughts because one cannot sever the contact between a thought and the world but by relying on the support of other thoughts within the critical mass.
Davidson makes use of the master claim to argue that we cannot be completely wrong about the world. We can get the gist of his arguments in this passage:
It does not follow, from the fact that any one of the bills in my pocket may have the highest serial number, that all the bills in my pocket may have the highest serial number, or from the fact that anyone may be elected president that everyone may be elected president. Nor could it happen that all our beliefs about the world might be false. Suppose I think I see a mouse disappear behind a chair. Clearly this belief could be mistaken. But would this belief be wrong if I did not truly believe a mouse was a small, four-footed mammal, or a chair an object made for sitting? Perhaps. There may be no saying exactly what other true beliefs I must have in order to have a particular false belief. But it seems clear that a belief of any kind, true or false, relies for its identification on a background of true beliefs; for a concept, like that of mouse or chair, cannot remain the same concept no matter what beliefs it features in. [...] Because of the holistic character of empirical belief, then, it is impossible that all our beliefs about the world are false. (1990, p. 194-5)
The interdependence of beliefs gives rise to an impossibility to hold that all beliefs within a critical mass is false. It is therefore unconceivable that all beliefs in such a critical mass were false because they are semantically tied one to the other. Those critical masses can be called semantically interdependent belief sets (sibs, for short). A sib, a convenient singular for sibs, is the unit of interpretation and empirical import. Davidson's argument can then be presented in three steps:
(1)A belief can only be interpreted within a sib.
(2)A belief can only face a verdict from the world within a sib.
(3)It is unintelligible that a sib that is not a proper subset of any other sib is false (that all beliefs in it are false)
One could interpret a single belief but this can be done only by placing it within a sib, either a sib composed by the beliefs of the interpreter of one composed by other beliefs held by who is being interpreted. Similarly, to make sense of a verdict, we need a set of accompanying beliefs. As for the conclusion (3), clearly a sib could be deemed false (a set of beliefs could be deemed false) with respect to other beliefs that make it (them) intelligible. But if the sib is large enough not to be a proper subset of any other, it cannot be taken to be false.
At least two pressing issues remain. First, the issue of whether the argument provides thought with any contact with the world at all. One can concede that a large enough sib cannot be (intelligibly) false but that yields nothing concerning its truth (or approximate truth if one manages to define an approximately true sib in terms of the truth of most of its members). The sib could be something short of truth – something like a set of beliefs that we are compelled to accept in order to doubt anything or in order to interpret other believers and make sense of language. Intesubjective, in one word, could comes short of objective. We can then reiterate the argument once again: consider the confrontation between the intersubjective beliefs (the intersubjective sib) and the objective ones – or between the intelligible beliefs and those that are true. The confrontation would only make sense if some beliefs within the intersubjective sib (or the intelligible sib) are also objective (or true). Without a critical mass of truths inside a critical mass of beliefs claims concerning local or global falsehood are nonsensical. It is only within critical masses that we can talk about interpretation and empirical testability and, as a consequence, it is only within critical masses that we can confront the beliefs that we cannot take to be false and any others. The master claim introduced critical masses and made room for sibs. Davidson's argument then made it clear that a critical mass of truths is to be present within each (large enough) sib.
A second issue is that of the size of those sibs. As conclusion (3) points at a sib that is not a proper subset of any other and therefore hints at the direction of a sib composed by all (or most) beliefs one holds. We assume sibs are consistent and therefore the beliefs one holds are not in a subset of a sib formed by the beliefs everyone holds. Davidson seems to assume that my or our beliefs about the world are suitable sibs for the argument. We shall not deal in detail with the question of what would be a suitable sib for the argument to work in order to establish a holistic contact with the world. But, leaving aside what is the most suitable sib for the purpose of Davidson's argument, it is interesting to ask whether a similar argument could work for sibs that are in fact proper subsets of other sibs.
We would like to briefly point at an interesting candidate sib that would enable contact with the future to be possible. Consider the sib composed by all beliefs concerning the future. This sib has elements like the following beliefs:
a. The sun will rise tomorrow.
b. The rooster will crow tomorrow.
c. The clock will carry on working tomorrow (as it did today).
One could then argue that each belief in this sib can only be understood in terms of the others. A generalized skeptical argument against beliefs about the future would then be blocked because one could not doubt all the beliefs in the sib at the same time (as any confrontation of any belief with the world would have to be supported by others beliefs in the sib). Analougously, a general skeptical argument against inductions would be blocked if we consider a sib with all of our beliefs acquired by induction. This is an interesting and somehow surprising holistic use of the master claim as it shows that maybe we can doubt some inductive conclusions on the ground of other inductive conclusions, but we cannot afford to doubt all inductive conclusions at once.