Reply to the response to Entry 249

Definition of term in the entry (as requested by the Inexpressible Committee):

‘Epistemic justification’ can be defined in numerous ways, and has been - the entire topic is an industry of controversy. Alston has an excellent paper - ‘Concepts of Epistemic Justification’ - in which he attempts to exhaustively define epistemic justification to find which definition is ultimately the best. Papers by Goldman, Bonjour, etc., are also excellent sources for the topic.

In other words, generically, ‘epistemic justification’ is the notion of the justification of one's belief so that the subject be considered to ‘know’ (or not) that belief. Briefly, it is:

S is justified in believing that p iff S has reasonable/sufficient grounds/believes he possesses adequate evidence/etc for believing that p.

The part after ‘S is justified in believing that P iff S...’ above is basically the definition and heart of the controversy over defining epistemic justification. I.e. there are problems with defining epistemic justification as: S is justified in believing that p iff S has sufficient grounds for believing that p. Again, there are problems with pretty much all definitions of epistemic justification, however, some are more tenable for reasons of their own."

Entry:

"Also, we are confused by your distinction between internalism and externalism, in which you appear to imply that the latter is not subject to infinite regress and can be epistemological justified. Since we apparently cannot get outside of our minds and know that we are, so that all we know is what we know, how can there be a distinction, in terms of knowing, between internalism and externalism?" (Excerpt from Response)

"This is a clear confusion between, what I assume you've been calling ‘absolutism’ (i.e. ‘objective’/a priori truth) and the notion of externalism. Externalism is only an account of epistemic justification, not the objectivity of a proposition's truth-value. There are serious concerns with attributing truth-values to propositions in the first place. One is that some propositions are neither true nor false (though some argue the contrary), or the problems caused by indexicals/token-reflexivity.

So when I make clear the distinction between internalism and externalism, a distinction between an account of epistemic justification is being made, not one between whether or not we can ‘objectively’ know something (i.e. by ‘meta-analysis’ of some sort, i.e. your example: ‘we cannot apparently get outside of our minds and know that we are’). The reason externalism is rightly so-called externalism is because epistemic justification is outside of the subject's awareness/analysis, while internalism isn't.

I'm not following your distinction between a proposition as ‘absolute’ and ‘non-absolute’. In fact, I don't see how there could be a distinction. Moreover, how would it affect epistemic justification in any way since propositions themselves don't have any relevancy to epistemic justification directly (the beliefs of those propositions do however)?

The problem of infinite regress is still inherent to your challenge unless you can remedy by explication of the aforementioned distinction (between absolutism vs. non-absolutism) and its relevancy to epistemic justification, and more specifically, the internalism."

Michael De July 22 2001

Response:

You appear to distinguish externalism from internalism on grounds that the epistemic justification for externalism is "outside of the subject’s awareness/analysis", whereas the epistemic justification for internalism is not. Also, you define epistemic justification as the "reasonable/sufficient grounds" an individual believes he or she possesses for his/her "evidence" for his/her belief, so that in terms of externalism, the "reasonable/sufficient grounds" for his/her evidence is outside of his/her awareness, and therefore, in terms of externalism and epistemic justification, (and in contrast to internalism), all the individual can know is the evidence for his/her belief. The problem with this distinction between externalism and internalism, is that an individual by being aware of his/her evidence of his/her belief must have some awareness of the grounds, no matter how insignificant, for attaining the evidence. For example, a scientist in a laboratory has evidence from an experiment that when water is mixed with oil, the two substances remain separate, and that the reasonable/sufficient grounds for his evidence is that he has visually and chemically identified the two substances prior to the experiment, and visually and chemically identified them after the experiment. However, to claim that the scientist has no reasonable/sufficient grounds for his evidence or experiment does not make sense, because a reasonable/sufficient ground, at some level, is necessary for there to be evidence. Therefore, we think your distinction between externalism and internalism based on the account of epistemic justification does not hold.

Regarding the problem of infinite regress, we contend that infinite regress only exists if an individual perceives his/her belief as certain, so that when the belief is broken down in terms of its meaning, it leads to a repetition of a fundamental idea to the belief, like the example you previously described, of "knowing that we know [how?] that we know [how?] that we know [how?], ad infinitum..." However, if an individual perceives his belief as a non-absolute, he/she will not face infinite regress, because when his/her belief is broken down, it will reach a stage of not knowing. So in terms of your example, it will go like this: we know {how?} that we know {how?} that we know {how?} that I do not know. Therefore in terms of epistemic justification or reasonable/sufficient grounds for evidence, infinite regress does not necessarily apply. (For you to maintain that there is no distinction between absolute (certain) and non-absolute (uncertain) belief in the perception of meaning/knowledge, then we would like to know how.)

251. Entry:

Reply to the response to Entry 250

"Also, you define epistemic justification as the "reasonable/sufficient grounds" an individual believes he or she possesses for his/her "evidence" for his/her belief, so that in terms of externalism, the "reasonable/sufficient grounds" for his/her evidence is outside of his/her awareness, and therefore, in terms of externalism and epistemic justification, (and in contrast to internalism), all the individual can know is the evidence for his/her belief" (Excerpt from Response)

There is some confusion here. I didn't say that epistemic justification is defined as "reasonable/sufficient grounds an individual believes he/she possesses for his/her "evidence" for his/her belief". I said that is one possible definition. It's also one possible definition that's considerably problematic (i.e. S believe that p, and S believes he possesses reasonable/sufficient grounds for believing that p. However, if S mistakenly believes his grounds to be sufficient (when they're not), then he also mistakenly believes he is justified in believing that p (when he is not). Thus, this definition of justification is frivolous.).

"However, to claim that the scientist has no reasonable/sufficient grounds for his evidence or experiment does not make sense" (Excerpt from Response)

Of course. If you define epistemic justification as above, then the scientist will not be epistemically justified since he lacks reasonable/sufficient grounds. However, here is where you're confused again. Externalism doesn't claim that epistemic justification, according to the above definition, entails that the subject not have reasonable/sufficient grounds for his evidence (as that wouldn't make sense, because nobody could ever be justified since the definition require that the subject have reasonable/sufficient grounds), but that the subject not know the grounds for the evidence for the belief that p, even though he still possess reasonable/sufficient grounds for the evidence for the believe that p (he just doesn't know them). This is necessary to avoid the infinite regress problem. One such example of such a theory of justification is reliablism (a proponent, for example, for such a theory is Alvin Goldman).

"Regarding the problem of infinite regress, we contend that infinite regress only exists if an individual perceives his/her belief as certain" (Excerpt from Response)

This is certainly unusual. Wouldn't any belief whatsoever be considered to be "certain" by the possessor of the belief? Is what you mean is something like the following example?

1. "I believe in God, but I'm not entirely sure that God exists."

This, and like beliefs, are actually reducible to multiple beliefs (as emphasized by use of the logical operator equivalent to 'but', which shows that it's truth-functional). The two beliefs would be as follows:

1.1 "I believe in God" or "I believe that God exists"
1.2 "I believe that it is possible that God does not exist" or "I believe
that it is not logically necessary that God exists"

Those two beliefs are certain at the time of their possession by the possessor. If a defeater of the belief causes the possessor to change his belief to "I do not believe in God", then that belief is certain at the time of possession. A belief at the time of possession is always held to be certain by the possessor. The truth of the proposition that is believed is something entirely different, and may in fact be what you're referring to as "absolute" or "non-absolute" (i.e. whether or not the truth-values of propositions are absolute).

"However, if an individual perceives his belief as a non-absolute, he or she will not face infinite regress, because when his/her belief is broken down, it will reach a stage of not knowing" (Excerpt from Response)

This is certainly a problem with epistemic justification yet again. Is it enough that S know the grounds for the belief that p, or must S know that he knows the grounds for belief that p, and so on? This is similar to a causal theory of knowing, where sufficient grounds for a belief must cut off at some causal link in the chain, otherwise you either end up infinitely regressing or reaching a first cause. Of course, the notion of a "first cause" is as problematic for a theory of justification just as much as infinite regression is.

"(For you to maintain that there is no distinction between absolute (certain) and non-absolute (uncertain) belief in the perception of meaning/knowledge, then we would like to know how.)" (Excerpt from Response)

Thus I have shown that there is no distinction between absolute/non-absolute beliefs (or at least the distinction of another kind - i.e. of propositions, not beliefs), and that they have no relevancy to epistemic justification directly.

Michael De July 22 2001

Response:

We do not think you are successful in getting externalism to avoid infinite regress by claiming that the subject (or individual) does not know the grounds for the evidence. The notion of ‘not knowing grounds at any level’ is false. As we have shown previously in the experiment example, the scientist cannot have any evidence without being consciously aware of a ground(s) for the evidence. Why is this so? Grounds provide a framework for evidence, without which there would be no evidence. To illustrate, we return to the scientist, and ask how he could have an experiment with water and oil, without being conscious of a ground(s)? What would be the basis for the experiment? How could he identify water or oil, or any other substance, or even what substances are? The answer is that he must be conscious of a ground(s) at some level, whether it is chemically identifying water and oil, or simply visually identifying water and oil. However, all is not bad for the externalist position, because there is a way to avoid infinite regress: concede that the grounds for evidence the subject is consciously aware of are non-absolute (uncertain), and therefore they are not subject to infinite regress due to them being cut off in the causal chain from lack of complete, absolute knowing.

How can a belief at the time of possession always be held certain by the possessor, when due to self-referencing, there is apparently nothing we can know with complete certainty? ("253. At the foundation of well-founded belief lies belief that is not founded." Ludwig Wittgenstein from "On Certainty") The possession of beliefs or the time of possession of beliefs, or a subject’s awareness of his/her possession of beliefs, do not necessarily equate with certainty regarding the existence of those beliefs from the subject’s perspective. It is possible that the subject is mistaken. Also, how can you distinguish belief from proposition, when there is no belief we are conscious of without it containing a proposition?

"Is it enough that S know the grounds for the belief that p, or must S know that he knows the grounds for belief that p, and so on?" (Excerpt from Reply)

We contend that S must know that he knows some grounds for belief that P, otherwise there is no belief that P, and that S cannot know that he knows all the grounds for belief that P due to self-referencing.

If propositions, according to you, have no relevancy to epistemic justification, how do you account for the apparent fact that there is no epistemic justification at any level without propositions? For you to claim that there are unknowable grounds that subjects possess, you are still incorporating a proposition into your epistemic justification. (i.e. the possession of unknowable grounds.)

252. Entry:

Reply to the response to Entry 251

"I am not saying that the scientist wouldn't have grounds for the belief that p, but that he wouldn't know all of the grounds that constitute justifiedness in the internalist sense. Indeed, the scientist possesses the grounds, but acquired them non-consciously (i.e. subconsciously by a reliable mechanism such as sense perception; or by subconscious meta-analysis of other beliefs that would suggest/entail the belief that p).

So, whereas internalism, say, requires that the S possess and know all the grounds which constitute adequacy for the belief that p at any time, externalism require only that S possess adequate grounds for the belief that p, but only know some of those grounds for that belief. Hence the problem of infinite regress is solved by the externalist account of epistemic justification.

If now you're asserting that for S to be epistemically justified, he must possess adequate grounds for the belief that p but not know all of the grounds which constitute adequacy [epistemically], then what you're asserting is an externalist account of justification. And this is exactly what seems to be the case.

When I referred to a belief to be certain (by the possessor at the time of belief), I meant subjectively certain. In other words, I meant that the possessor of the belief believes it to be certain at the time he possesses it. Beliefs themselves aren't objectively certain in any sense, but arguably the propositions to which they refer might be. The distinction between a belief and a proposition is that the proposition is the object of a belief. Moreover, beliefs are not necessarily true nor false, they do not necessarily possess predicative truth-values like propositions do, and thus they are treated differently (at the very least logically different).

'If propositions, according to you, have no relevancy to epistemic justification...'
Actually, I said propositions have no relevancy to epistemic justification directly - beliefs do. Indirectly they play a role in justification since the justifiedness of a belief is related to the proposition(s) to which it refers. However, this relation does not constitute the entirety of the justifiedness of a belief. Consider, for example:

S believes that p

For S to be epistemically justified in believing p, S must:
1. possess adequate grounds for the belief that p
2. not possess defeaters that would render the belief unjustified
3. possess a coherent relation between the belief and proposition(s) to which it refers (i.e. to avoid incoherence, or the independent satisfaction of conditions).
4. etc.

This is the reason that propositions and beliefs are treated differently, especially for epistemological purposes. A proposition is either true or false, a belief is either justified or unjustified.
So, if we take your now externalist account of justification when considering the proposition 'We cannot know who we are', we can certainly answer/overcome this. Briefly, it isn't the proposition at all that needs overcoming (as that doesn't make much sense in the first place), but the belief of the proposition.

If you wanted to overcome the belief we could further apply your condition that it not be necessary that the believer know all of the grounds for the belief that 'We cannot know who we are' and thereby avoid infinite regress. Then, we know who we are, viz., animate, corporeal/material biotic entities comprised of such and such matter (quarks, leptons, gluons, atoms, molecules, and so on). This is justified since the evidence/grounds for the belief are justified by an agreeable definition of epistemic justification. I.e. the evidence satisfies the condition of , say, adequacy, coherence, etc., and thus the belief is justified."

Michael De July 27 2001

Response:

Now that you have conceded that externalism requires that "S [subject] possess, [and know], adequate grounds for [beliefs]", but not know all the adequate grounds for beliefs, the distinction between externalism and internalism rests on whether or not internalism requires knowledge of all adequate grounds for belief. Due to the self-referential nature of knowledge, how can a subject know that he/she knows "all" the adequate grounds for his/her beliefs? We do not think it is possible within the limit of self-reference. Therefore, we think the distinction between externalism and internalism is illusory--they both require that the subject possess some adequate grounds for belief, but due to our apparent conscious limit, not all the adequate grounds. In regard to the competition, the challenge of infinite regress falls short, because due to the apparent limit on what we can know, the causal chain of our knowing, whether in terms of externalism or internalism if we were to accept the distinction, cuts off at some link in the chain, thus avoiding the infinite regress problem.

We think that on the surface your distinction between belief and proposition stands, but if we cut into the grounds for your distinction, we think the distinction loses all form except as differing labels.

Your argument for the distinction rests on the following grounds:

1. proposition is the object of a belief.

2. beliefs do not necessarily predicate truth-values, unlike propositions.

To test your grounds, we use:

1. the belief, "I believe the earth is round."

2. the proposition, "The earth is round."

Clearly, your first ground holds, because "earth is round" is the object of the belief. However, it is from evaluating the second ground, the first ground and argument as a whole are defeated or shown to be less reasonable than the position that belief and proposition are fundamentally the same. On what grounds? The proposition "The earth is round" is stated by a subject(s); and since our knowledge is not absolute as in truth-value, the proposition is an expression of belief, though the subject is not known. Therefore, both proposition and belief are forms of belief, with knowledge of the identity of the subject, or the lack of, being the main difference.

We agree that due to the limitation of us knowing all the grounds for a belief, it is possible to avoid infinite regress in knowing something. However, by knowing some of the grounds for a belief, does not necessarily mean that the belief is justified or valid in relation to antagonistic beliefs. So your argument that we can know who we are based on the evidence of partially knowing that we are "animate, corporeal/material biotic entities comprised of such and such matter (quarks, leptons, gluons, atoms, molecules, and so on)", and that the evidence is adequate and coherent, does not necessarily hold. For instance, it is questionable that your evidence directly pertains to who we are, because it could indirectly pertain to who we are viz., limited representation. To answer this question of limited and unlimited representation, we must look at the knowledge and its origin. It is our contention that knowledge is created at the point of sensory reception, whereby sensory receptors respond to external stimulus, thereby create sensory information which is then potentially converted into conscious knowledge, so that what we know is not what really is, but limited representation of what really is. Hence, we contend that the argument that we can know who we are through partial knowledge of who we are is defeated or less reasonable than the contention that we can only know who we are in a limited sense, without actually knowing who we are partially or entirely.

Entries 246-249,  Entries 253-255