Epistemic Neutrality

A while ago, I wrote a post in which I attempted to show why it is not possible for a thinker to suspend judgment about each and every proposition. He will take for granted that certain words mean certain things, that he doesn't need to act a certain way, etc. This is just another way of saying that everyone makes knowledge claims - implicitly or explicitly. When this is acknowledged, the ordeal of worldview comparisons can begin.

This much is clear. But there are people who seem to think - for whatever reason - that it's good to be epistemically neutral... as if just by their say-so it is true and somehow makes them appear objective? I'm not quite sure what the underlying reasoning is. At any rate, the persons I have in mind don't deny that knowledge is ideal but also continually give the impression that true epistemic humility consists in always learning yet never being able to come to a knowledge of the truth (2 Timothy 3:7).

The idea that a sort of wiped-clean, blank epistemic slate - "doubt everything," "build from nothing," or whatever other form this idea may take - is a necessary step in order to achieve knowledge is bizarre. In spite of the above considerations, suppose one could suspend judgment on all matters. But if one were ever truly thoroughly epistemically neutral, there would be by definition no criterial basis on which he could ever move to non-neutrality. The first step toward knowledge is not neutrality, it's committing to some criteria by which one can allegedly distinguish what is and isn't knowable.

On Clark's system, for example, I need to know another's first principle in order to show it to be self-defeating on its own grounds. But apagogic argumentation itself presupposes I hold to a worldview from which I am able to operate. To know how to undermine skepticism, for instance, I must first be able to know what skepticism is. But how? Certainly not on skeptical grounds, for skepticism is self-defeating. You can't really criticize another's worldview until you have - or, at least, think you have - one of your own. But then is it is clear that one can't criticize from a self-pronounced position of epistemic neutrality, for one cannot operate from a position whose sole requirement just is to abstain from all positions, including criteria for knowledge. That itself is self-defeating. 

More likely than not, the problem is that people don't mean what they say. They don't actually advocate suspending judgment on absolutely everything. But then so much for their arm-chair objectivity.

Truth, Knowledge, and Justification

I've recently been spending time working through all of Clark's writings (all of which I am aware, anyway) and transcribing the statements he's made which seem to me to be remotely related to meta-epistemology: relevant definitions, necessary and sufficient conditions for knowledge, and so on. I'm currently sitting on 110+ pages of material. I still have a few books to works through, but so far, the process has at least helped me to focus my own views. In this post, I just want to consider the meanings of a few key terms, terms whose meanings are generally too often taken for granted for my liking. 

I've written posts on the philosophic definition of knowledge that I prefer here and here. To state it briefly, knowledge is propositional belief in which the possibility of error is precluded. That is, to say I know a proposition is true is to say I cannot be mistaken in my claim that the proposition is true. This position is called infallibilism. 

[Note that I don't deny that "knowledge" can refer to other, less stringently conditioned belief states or whatnot. I just do not find these as interesting.] 

When I speak of [epistemic] justification, I have in mind what it is that allows us to know. Knowledge is different than opinion. How we can determine into what category a given belief falls depends on whether or not we are able to justify our belief, to provide the reason or reasons which rule out the possibility of error. If we are able to appeal to reasons according to which we can show a given knowledge-claim is true, then we say our knowledge-claim is justified. We do, in fact, know the proposition, for we have justification for our belief in it.

Finally - and I would consider this to be subject to polishing more so than the above - when I speak of truth, I intend to refer to the - not merely "a," as if there were more than one - entire set of consistent propositions, the concretes - propositions whose subjects refer to individuals or particulars and therefore can't function as genera - of which correspond to some other reality (propositional or non-propositional). This is a sort of hybrid between the coherence and correspondence theories of truth. 

[Note that how truths correspond to other realities is something we do not necessarily need to know in order to know a given truth. This paragraph has to do with what truth is, not with epistemic justification.]

It goes without saying that there is a lot more to the discussion than this. The very definitions I provide use terms which probably require further explanation (belief, proposition, possibility, error, reason, set, consistent, correspond, concretes). It wouldn't hurt, anyway. And obviously, I did not even attempt to explain the entire set of what I think can be known, what suffices as justification, or what is true. I have tried to make a beginning of this elsewhere on this blog. Although I think every proposition in this post can be known, is justifiable, and is true, it would require me to write a book to show why. I am still young enough and foolish enough to feel the temptation, but this will do for now. As Clark was so fond of saying, everyone has to start somewhere.

Unknowables

If we reject propositional monism, then given that propositions alone can be true – and, hence, propositions alone can be known – we must assert there are unknowables. Does this not seem paradoxical, perhaps contradictory? How can we know [of] something that is not knowable?

Ironically, even on the assumption of propositional monism there would be unknowables: “One cannot know what is false” (Gordon Clark, Clark and His Critics, pg. 399).  Sure, we can say something about false propositions – we can say “x is false” where x means “A is not A” – but x itself is not knowable. The same principle applies to meaningless statements.

One may object that this is not a relevant counter-example because falsities and meaningless statements can still function as subjects of true propositions. Maybe the objection has a point. Maybe not. The objector will have to specify why it is relevant that “their” unknowables can function as subjects.

But either way, that propositional monism is false is demonstrated – or, at least, is purported to be demonstrated (link, cf. here) – by elenctic or reductio ad absurdum argumentation. In such cases we assume a position for the sake of argument in order to show why it can’t be true – it’s self-defeating. The method is not to begin with the contradictory of propositional monism. As such, we don’t have to act as if we can really communicate or refer to the unknowable [non-propositions] qua the unknowable [non-propositions]. And surely along these lines the correspondence theory of truth is worth some attention (though not to the exclusion of, say, the coherence theory of truth).

Objections to the Need for an Omniscient Communicator

When I've explained why I think communication from one who is omniscient is an epistemic necessity (link), there are a few objections I’ve received (and some I haven’t) which I want to explore in this post:

Objection #1: Your argument implies you need to be omniscient in order to know that the revelation in question is from one who is omniscient.

This objection is given by rationalists or those who pose as rationalists for the sake of argument. It tries to take what reasons I give for the need for an omniscient person and throw them back at me. In essence, it seems to deny that discursive knowledge is possible.

The problem with this objection is that usually no explanation is given as to why it can’t be the case that the communication from an omniscient person is a or the self-authenticating sufficient condition for knowledge (link). That is, there is no higher standard, but there doesn’t need to be any higher standard. This is not to suggest that the communication itself can’t be the subject of other tests for truth – I’ve argued for several necessary preconditions for knowledge (link) – but the point is that no first principle or axiom can or need be subject to external validation. Scripture is what is says it is, and we can know it for that reason. There are indeed tests for truth without which it could not be the case it is what it says it is, but these confirmatory evidences do not on that account become premises according to which we base our belief in its divine origin. For in order for Scripture to truly be the sufficient condition for knowledge, it must be the case that these criteria too are prescribed by Scripture.

Objection #2: If Scripture is self-authenticating, why isn’t there more agreement that it is God’s word? And what about other people who claim they have self-authenticating revelations from a different omniscient person? 

These are usually the follow-up responses to Objection #1. Really, the first question is little different than the question posed so often by Roman Catholics: if Scripture is perspicuous, why don’t people agree? In both cases, the answer is the same: people’s misuse of a sufficient[ly perspicuous] source is no indication that the source is inherently faulty. The noetic effects of sin, for example, is provided by Scripture as an explanation for these occurrences. See also divine determinism in Romans 9:1-6ff. as another or further explanation.

And as for competing claims of divine revelation or whatnot, this is where the tests for truth or necessary conditions for knowledge alluded to above – not to mention internal critiques – are really useful. As mere necessary conditions, they may not be able to function as the basis upon which one can justifiably infer what worldview is true, but given a true worldview via a sufficient condition for knowledge, they can function as bases upon which one can demonstrate what is false. 

Obviously, there are any number of competing worldviews to Christianity. Infinitely many possible worldviews, in fact. Not only would it be impossible for any Christian to justify any knowledge if he first had to reduce to absurdity all other worldviews first, it would be impossible for any individual to do so with respect to any worldview. This is just to say that one must begin with truth before he can refute error. This may seem clear enough, but when a person proceeds to fault me for having accepted a particular worldview even though I haven’t examined all others – as if such were even possible – you have to wonder if they understand the implications of their [self-refuting] statements.

Objection #3: If truth can be self-authenticating, then an omniscient communicator is unnecessary. 

I actually haven’t encountered this objection, but it’s one I wanted to mention so that anyone who does will have some idea of how to respond. The idea is that an omniscient communicator is just an unnecessary middleman. If the proposition that "not all truths are related such that to know one thing entails omniscience" is said by the Christian to be self-authenticating as communicated by an omniscient God, why not just take the proposition itself as self-authenticating and leave off it’s having been revealed by God?

The problem with this is that the one who claims partial knowledge is in no position to know of himself that, say, the doctrine of internal relations is false. In other words, he is begging the question. It may indeed be the case that the doctrine of internal relations is false, but only one who knows all truths would be able to examine the propositions He knows and determine such without having to rely on anyone else. Only such a person could then tell this to others who can’t rely on themselves. Or, as I essentially point out in one of the above links, the proposition that "it is self-authenticating that not all truths are related such that to know one thing entails omniscience" is itself in need of scrutiny in the context of all [other] propositions known to be true. But this requires omniscience. If true – and it is – we have not cut out the middleman but rather established Him as an epistemic necessity for us.

Thoughts on Contemporary Epistemology

I recently received Contemporary Debates in Epistemology as a birthday present. I was curious was the hot topics are among the epistemologists of today and figured that sooner or later I should branch out my reading if I seriously intend to pursue a degree in philosophy. Anyway, there are eleven topics debated in this book. I won’t go through all of them, not the least reason for which is that I don’t feel I have an adequate grasp of all of them. But among these topics are closure, skepticism [with respect to the external world], contextualism, and infinitism. With so many opinions, it’s hard not to find someone to [dis]agree with. It’s a mixing and matching. You have to read critically, i.e. you can’t just cross someone off for making one point with which you disagree. I disagreed with Dretske regarding closure, yet I think he makes a good argument against contextualism. The same goes for my readers. I’m sure I made bad points. Hopefully, I made some good ones too, and I hope further these points are not overridden by my bad ones. Now then, with respect to this book, you’ll have to get it to read specifics one way or the other, as I’m just going to comment on a few points that were of interest to me:

Closure

Closure is the idea that “If S knows p and knows that p implies q, then S is in a position to know that q” (pg. 2). On the face of it, this seems obviously true. Another author, Richard Fumerton – whose helpful distinction between necessary and sufficient conditions for knowledge I have quoted on this blog before (link) and whose books Epistemology and Metaepistemology and Skepticism would seem to be worth a look – remarked in a footnote of an essay on a different subject that “the bottom line is that if a philosopher advances a view that forces us to reject closure, that should be taken as a reduction of that philosopher’s view” (pg. 96). A very well made and succinct point.

The reasons why one would reject closure, however, make it a worthwhile discussion. One reason to reject closure is that it allegedly makes possible “ordinary knowledge.” If we reject closure, we could in principle know, for example, that “John ate all the cookies” without having to know “There are physical objects,” “There are other minds,” or “The past is real” (pg. 3). “John ate all the cookies” may implies these propositions, but we don’t have to know these propositions to know “John ate all the cookies” if closure is rejected. Likewise, we wouldn’t have to worry so much about Descartes’ demon, solipsism, and recently spontaneously generated universes.

Of course, we could accept closure, these “heavyweight implications” of so-called ordinary knowledge, and on these accounts reject ordinary knowledge as infallibilistic. In fact, as a Scripturalist, that seems highly preferable. Much of what we say or do are not based on beliefs which can’t be mistaken. Clark was right to throw the objection that he couldn’t know his wife back in the faces of his critics. So what if our knowledge of such isn’t infallible? The more interesting question is what knowledge is infallible.

Skepticism

This brings me to the issue of skepticism with respect to the – or perhaps “an” would be less question-begging - external world. Surprisingly – pleasantly surprisingly – the two authors seemed to agree that “we cannot avoid knowledge skepticism with respect to the physical world if we understand knowledge as requiring justification so strong that it eliminates the possibility of error” (pg. 86). [This too was written by Richard Fumerton.] His reference to a sort of justification which would preclude the possibility of error is another allusion to infallibilism, which I wrote about here a few months ago. Scripturalists will do well, I think, to focus on metaepistemology – the definitions of knowledge, epistemic justification, truth, and the like. It may turn out there is more agreement to be had with the scholastic paragons of contemporary epistemology than one could hope for.

Contextualism

If I understand contextualism correctly, it rests either on an equivocation of the meaning of “knowledge” or is a flat contradiction, neither of which is particularly useful. It basically says (pg. 45):

…whether S knows something – that she has two hands, for instance – depends on the context of the person who is saying S knows it. If I, a philosopher, worried about brains in vats and Cartesian demons, say it, then S doesn’t know she has two hands… But if S, an ordinary person on the street, someone without the least tincture of philosophy, says she knows that she has two hands, what she says is true. She attributes knowledge to herself in an ordinary, practical, context, in which demons and handless-brains-in-vats are not relevant possibilities… 

So who, according to contextualism, is right? Am I, a philosopher, right when I (given my context) say that nobody knows they have hands. Or is S right? We are, I’m afraid, both right. And that is where my low opinion of contextualism comes from.

If there is an equivocation – if what S means by “knowing” she has two hands is different from what the philosopher means when he says S [does not] “know” she has two hands – then rather than preserving knowledge of having two hands, it just highlights that S is probably not squarely facing the skeptic’s challenges (cf. the “heavyweight implications” in the section on Closure above).

Infinitism

Of all the debates I skimmed, this and perhaps the debate on a priori knowledge seemed to arouse the most incredulity from any of the debaters. The ideas that infinitism could be true or that there is no a priori knowledge are, thankfully, seemingly minority positions. With respect to infinitism in particular, the closing paragraphs from the detractor will close out this post (pg. 155):

In my essay I said that the (to my mind) most severe difficulty with Klein’s infinitism is that it is committed to the thesis that inference alone can create justification. Here Klein seems to embrace this commitment wholeheartedly, holding that the longer the chain of inferential justification for a given belief the greater the justification created, and that, if the chain is long enough (but still finite), the justification can “increase to the degree required for knowledge.” This seems to give us the result that knowledge does not require infinitely long chains of inferential justification after all: infinitism gives way to inferentialism. Worse yet, given Klein’s thesis that inferential justification is the only sort of justification there can be, we seem to get the result that one could start with a belief (or set of beliefs) that is totally unjustified, because it lacks any inferential justification, and by spinning out a long enough chain of inference from it reach a belief that has the degree of justification required for knowledge. 

These results are so counterintuitive that I hesitate to attribute them to Klein. But how else are we to interpret the quoted remarks?

Fuzzy Language

From pgs. 205-206 of A Christian View of Men and Things:

Now, any given word must signify one thing, or a finite number of things, or an infinite number of things. If the word has a finite number of meanings, then it would be possible to invent a name for each meaning, so that all words would have a single meaning. But if each word has an infinite number of meanings, reasoning and conversation have become impossible, because not to have one meaning is to have no meaning. But if a word has a meaning, the object cannot be both man and not-man. If the skeptic attempt to avoid the arguments, he might do so by saying nothing. In this case, however, there is no skeptical theory awaiting refutation. Or he might accuse Aristotle of begging the question by using the law of contradiction. But, then, if he says this, he has said something, and has himself admitted the force of logic.

While true, I don’t think the above is meant to suggest fuzziness in language is entirely useless. In some cases, ambiguity seems preferable. In particular, I’m thinking about instances of justified deception. A Nazi shows up on my door asking if I am hiding Jews. I respond, “there are no Jews here.” Now, through no fault of my own, the Nazi might interpret this to mean I am not hiding any Jews “at all,” whereas I actually meant I am not hiding Jews “in the visible vicinity.” The fuzziness lies in the meaning of “here.” Obviously, I didn’t answer the question the Nazi asked, but then again, I am under no obligation to answer him in the first place.

Now, if there were no ambiguity in language, I obviously could not deceive a Nazi. At best, I could say “search for yourself.” But if he persisted in questioning me, he would come to realize I was intentionally avoiding his question. What would follow I would consider to be less than desirable. So there is good reason for language to be fuzzy, though in an ideal world – a world without Nazis – clarity would be a worthy goal.

Classification and Trinitarianism

If we were to construct a Porphyrian tree, we would begin with the denotative list of all possible subjects and then, by naming in descending order of scope various genera in which we are interested, narrow this list to find individuals or species which fit our criteria. In the case of the Trinity, this process might look as follows:

Being: {personal, impersonal}

Personal: {divine, angelic, demonic, human}

Divine: {Father, Son, Spirit}

Insofar as the Father, Son, and Spirit are acknowledged by all Trinitarians to be three distinct persons, it is remarkable that some resist the inference that when we speak of the Trinity, we also speak of three distinct beings or divinities. Why is this the case? The Trinity are as much concrete members of these classes as they are of the class of persons. 

I suppose that in the case of "being," there may be an equivocation - for what some think "being" means might be something other than "the denotative list of all possible subjects," in which case they would have to specify just what they think it does refer to - but clearly there are three distinct members of the genus "divine." I can only think that the rejection of three divinities would be because so many [monotheistic] Trinitarians equate the concept of being "divine" with being "God" without any means of avoiding the implication of tritheism. But this is just special pleading.

Obedience, Faith, and [Good] Works

Given that God commands men to repent and believe (e.g. Mark 1:15), I've had a somewhat frustrating time understanding in what sense faith is not a good work. Doesn't faith "in the gospel" require mental activity and an act of will, viz. in understanding and assenting to certain propositions? And isn't obedience synonymous with good works (cf. Romans 1:5, 16:26)? Didn't Paul and Silas tell the Philippian jailer he had something to "do" in order to be saved (Acts 16:31-32)? 

I'm not sure why I made this question out to be more difficult than it really is, but in any case, I was struck with a sudden and obvious thought the other day. Faith is what consummates union with Christ, and it is Christ's work which merits for those united to Him a crown of righteousness or [eternal] life. Faith is obedience to a command, but it is not on account of this obedience that we merit anything (link). Works, on the other hand, are the grounds upon which merit is accrued to an individual - even good works (link). But no one can merit eternal life for Himself apart from perfect obedience to the law. We can merit other things: death by our sin (Romans 6:23), wages by our labor (Romans 4:4), rewards (25:14-30), but faith has always been held by Reformed Christians to be the instrumental, not meritorious, cause of justification. So obedience would seem to be a genus of which faith and good works are species.