Should Externalism Be Preferred To Internalism?

In this essay I will attempt to argue that internalism should be preferred to externalism. I will be arguing this on the basis of the two theories' different attitudes towards forgotten evidence. I will try to argue that the common intuition in both internalism and externalism, that forgotten evidence can still be counted as justification for a belief is false. I will argue that forgotten evidence cannot be used to justify a belief now, though it did once justify it, and that because of this, if the believer cannot access at least some justification for her/his belief, the belief is no longer subjectively justified. That is not to say that the belief is not objectively justified, but I do not believe that that automatically makes it subjectively justified, which I believe is an important part of knowledge. I will then go on to attempt to argue that the arguments that internalism can justify forgotten evidence are wrong, but that this is not a short-coming, but a virtue of the theory. Then I will try to argue that since externalism does allow that forgotten evidence can justify a belief, it cannot be preferred to internalism.

Internalism is the thesis that knowledge or justification is gained by having good reasons for one's true beliefs. Some examples of processes that one can use to form one's current beliefs are perceptual experience, memory, and previously formed beliefs. It is important to note that a subject S's reasons for believing a proposition p are not facts about p or p itself. Rather they are that p, or facts about p, are perceived by S in certain ways. For example, S does not form the belief that the tulips in the garden are red because they are red. Rather she/he forms that belief because it appears to her/him that the tulips in the garden are red. This is an internal factor in the knowledge requirement. For internalists, knowledge requires that one has a true belief with good supporting reasons or evidence. The good reasons/evidence requirement here becomes the justification requirement in the classical model of knowledge.

There are two branches of internalism, and they are known as mentalism and access internalism. The most common form of internalism is access internalism, which will be the focus of this essay. Within accessibility there are two branches: actual access and accessibility. Actual access is the idea that for every proposition p that one knows, one is also aware of the knowledge basis, or roots of p. Accessibility is the idea that for every proposition p that one knows, one can become aware of the knowledge basis, or roots of p. The actual access requirement seems to be too strong. It is implausible that one is always aware of where one learned a fact every time one uses it, especially facts learned long ago. In my opinion accessibility seems more plausible and is therefore a stronger claim. If one had to remember the basis for every piece of what we would like to call knowledge, most of our basic vocabularies would not count as knowledge, For example, I do not remember where, when or how I learned what a bus is, as I learned it a long time ago. However, it seems highly counter-intuitive to say that I do not therefore know how to recognise a bus. It would also have the absurd result that I 'know' a complicated philosophical concept that I learned about yesterday, more than I 'know' what a bus is, because of having memory of where and when I learned about the latter but not the former. Because of this, I will focus on accessibility.

Externalism is the thesis that knowledge does not require internal justification. There are different forms of externalism, but I will focus on process reliabilism, supposedly the most popular form of externalism. All externalists agree that in order to have knowledge, one must have a belief resulting from a process that reliably connects belief to truth. According to externalism, no support from any other beliefs or systems of beliefs is necessary. According to Alston, reliability requires that a process yields a high proportion of truths over a wide range of ordinarily encountered situations. This is known as process reliability. Alston admits that this definition is imprecise and that the already vague boundaries between what a typical and an atypical case is may shift over time. However, something that is intuitively pleasing about process reliabilism is that it rules out skeptical problems, by only focusing on facts that are directly relevant (or close) to the situation that one is actually in, and scepticism is assumed not to be relevant in most situations.

One criticism of access internalism is that a belief that was justifiably formed, which is later retained but the evidence forgotten, and is therefore inaccessible, is seemingly no longer justified under internalism, though one is supposed to have the intuition that that belief is still justified, and that this is therefore a failing of access internalism. Internalists respond to this case, saying that if the memory of the learned fact is vivid, the subject feels confident in the fact, and has no undermining evidence, the belief is still justified. The belief is similarly justified, even if it originally came from a disreputable source, if the believer knows themselves to usually be conscientious, because in forgetting the bad source, the believer looses their defeater. The belief therefore becomes more justified, if not completely justified. I believe that the problem with this argument is that it ignores the objective truth about the inaccuracy of the original source. In my opinion Conee and Feldman are confusing epistemic virtue with justification. The believer may be personally justified, as far as s/he is aware, i.e. because of not having any defeaters, in believing the proposition for which s/he originally had bad evidence, but as far as the evidence itself is concerned, if it came from an inaccurate source and therefore did not count as good evidence at the time it was acquired, I do not see how the evidence being forgotten makes it more justified.

Internalism states that justification requires that one not only has good reasons for ones true beliefs, but also that one believes based upon those good reasons. In my opinion, this is one way in which internalism is the better of the two theses. Because externalism only requires that one has a belief formed by a reliable process and which is causally connected to truth in an appropriate way, one could be justified without being aware that one is justified. However, I believe that this also means that the case of forgotten evidence cannot be justified according to internalism. This is because in my opinion, once evidence for a proposition has been forgotten, though one once believed that proposition because of the evidence, one can no longer be said to do so, and the proposition is no longer justified, unless it is backed up by different evidence that is connected to it in the appropriate way. Either way, the proposition is no longer justified by the original evidence.

Externalism is not a theory that assumes that justification needs to be known to the believer. There must be a process or some properties of the belief that make it justified, but the believer does not need to be aware of this. Therefore in the forgotten evidence case, the externalist can answer that the belief is justified because of the intrinsic merits of the external source of the belief. Externalism requires that in order for knowledge to be obtained there must be the appropriate causal connection between belief and truth. The causation is external because all that is required is that belief actually be related to truth in the appropriate way, not that we have any conception of this causal connection. Using a combination of these two aspects then, externalists can say that one has knowledge if one has a belief that is causally connected to truth by a reliable process, a reliable process being one yielding a higher amount of true than false beliefs over time. As Goldman put it, a belief is justified if and only if it is well-formed, and whether or not it is well-formed can be assessed by looking at its historical causal ancestry. This approach is known as Historical Reliabilism. This could be the externalist answer to the forgotten evidence case, and I believe that it is consistent with the doctrine of externalism. One can still have a reliable way of forming one's beliefs, even if one has forgotten the original evidence for those beliefs. After all, the belief is objectively justified regardless of whether or not it is subjectively justified, and if this really is what is important in externalism, then whether the evidence was forgotten or not, it is still there, and probably it did once connect truth to belief. A process that reliably connects truth to belief is still reliable, and can still be known by the subject to be reliable, even if the evidence for the belief has been forgotten by the subject.

However, I have a different intuition about the forgotten evidence case. In my opinion, if one forms an opinion based upon good evidence but then forgets the evidence, one is no longer epistemically justified in retaining that opinion. I agree that justification is not simply a subjective matter, and that a belief can be objectively justified, no matter what the subject's reasons are to believe it. But that does not mean that one is subjectively justified for holding an opinion, the evidence for which one has no access too. I agree with J.S Mill that if one does not question one's beliefs then they loose their strength, and can become 'dead dogma.' Part of this questioning is questioning the evidence, but that cannot be done if the evidence is inaccessible to the believer. In other words it could become a belief held only out of habit, and in my opinion that is not a justified belief. Also, since internalism requires that the evidence and the belief are appropriately connected, and that this is acknowledged by the subject, I do not see how forgotten evidence can justify a belief according to internalism, as the belief is surely no longer held for that evidence, as it is forgotten, but for some other reason. That is not to say that the other reason can not be a good justifier, only that it has become the justifier, and the forgotten evidence no longer is for the subject. I am therefore of the opinion that internalism is the better of the two theses, because it rules out the idea that forgotten evidence can count as justification for a belief.

In conclusion, I believe that internalism can be preferred to externalism on the basis that it rules out forgotten evidence as justification. There are arguments that forgotten evidence still justifies a belief, but I believe that this is only true from an objective basis, not a subjective basis. This is because I believe that a belief held without at least access to its evidence is not justified for the subject. I believe that because externalism seemingly treats justification as a purely objective phenomena, it fails to pick out what is important for human knowledge, which is, in my opinion, that truth be connected with belief not just because the world happens to be like that, but because the subject is aware and has evidence that the world is as it is. I believe that internalism is better able to do this.