During the course of my summer semester independent study, I studied the main arguments of various positions in epistemology, as reflected in the writings of several philosophers. My readings were organized into four units, or four topics: (1) skepticism and the definition of knowledge, (2) foundationalism, coherentism, and externalism, (3) naturalized epistemology and virtue epistemology, and (4) contextualism and relativism. Despite learning quite a bit, I remain convinced that all of the philosophers to date have only partially succeeded in their efforts – all have failed to completely address skepticism, or to create a successful epistemology. I believe, therefore, that skepticism is unavoidable, except in the case that we implement axiomatic systems of knowledge, and that epistemology is in the end in need of revision. I will start first with a brief summary of my objections to the conclusions of each reading, and then offer my skeptical conclusion.
The problem originally posed by Descartes, (summarized by Barry Stroud) has been addressed from several different angles, each philosopher attempting to solve it in his or her own way. Though some philosophers might claim that the problem has been solved, (or that there is no problem at all, or that we ought to imagine the problem in a different context) the problem remains. My explanations as to why this is the case follows in my objections to the conclusions of my readings, as outlined below:
G.E. Moore’s claim that he can know propositions to certainty, such as the fact that he is not a brain in a vat, as pointed out by Peter Unger, is an arbitrary claim, and a presupposition. It entirely avoids the most important questions the skeptic asks.
Nozick’s truth-tracking technique fails to seriously address the skeptic, because it denies the closure principle (more on that in another essay, perhaps) and because of its massive concession to skepticism and reliance on externalism – a theory of knowledge that becomes less useful the more we focus on the serious questions that the skeptic poses, such as, “how do you know you are not a brain in a vat”?
Chisholm’s attempt to recast the premises of the justification argument, in an attempt to argue that the senses are neither justified nor unjustified, fails as well, because it avoids the question of the senses entirely. Simply because Chisholm has made n the senses, and claimed that they are neither justified nor unjustified, does not answer the question: are the senses reliable? Sure, Chisholm’s argument looks great on paper, and from a philosophical perspective, it certainly is an achievement, but simply assuming that the senses are reliable because the argument fits nicely with foundationalism, and because it is a skill defense against coherentism, is no reason to claim that it has answered any of the real questions with which we are concerned.
Sosa’s virtue epistemology (and really, all virtue epistemology) is also insufficient. Although Sosa points out quite skillfully the weaknesses of traditional foundationalism as well as the weaknesses of traditional coherentism, and though his concept of a bi-level epistemology seems to be on the right track, his attempt to ground his third option in intellectual virtues is a mistake, since it does not address the relativity of virtue, and again, rests on an arbitrary foundation. A more useful approach for Sosa might have been to get to the heart of what we mean by “stable and reliable dispositions” in a more biological sense, as Stephen Stich attempts to do in his essay.
Davidson is right to claim that all forms of foundationalism are simply an attempt to ground belief one way or another on the testimony of the sense. He is also right to claim that grounding belief in any objective, ultimate, or absolute evidence is folly. Davidson’s coherence theory is so close to the mark that it nearly captures the essence of my epistemology, but ultimately fails, because it is not the sort of bi-level epistemology needed, and it does not consider axioms, nor does it answer (satisfactorily) the objection of the foundationalist, namely, that coherent webs of beliefs can include falsities. Still, Davidson refutes foundationalism well, (in most respects) and he is very near the truth, though his theories are not sufficient.
Goldman’s attempts to avoid the Gettier problem, though ingenious, not only fail to avoid the problem, but also fail to address skepticism. The main source of Goldman’s failure is the theory itself, for reliabilism is a failed theory, if nothing else for the simple fact that reliabilism entails externalism, which, as aforementioned, is an unsatisfactory theory of knowledge since it assumes that we can know things or be justified in a belief without having any access to the evidence that justifies said belief (this is partially true, as my theory points out, but not entirely true, and it is cast in the wrong context). Other objections show that Goldman’s position is untenable, such as the new evil demon problem (Lehrer), and Stich’s own objection that normative epistemic terms are culturally biased. For these and other reasons, Goldman fails to seriously address the skeptic’s questions.
Keith Lehrer, the creator of the new evil demon problem, although genius in his responses to Goldman, fails to refute skepticism, since he argues forinternalism. Unfortunately for Goldman, internalism, like externalism, is a failed epistemological concept (internalism, like externalism, is only partially true. The truth is somewhere in between the two theories, since we, as humans implement a bi-level epistemology that is both externalist and internalist at the same time, though those terms may not apply well to my own theory). Though Lehrer is right to reject reliabilism, he is incorrect to argue for traditional internalism, and in his so doing, he avoids the serious questions of the skeptic. This may seem confusing, as it seems that there could only be two options: either internalism, or externalism, and if you view epistemology traditionally, then this is where the confusion lies. Nevertheless, Lehrer’s arguments fall short, despite how one might view epistemology, since even by a traditional account, strict internalism is faulty.
Alston’s attempts fail, because he is merely revising reliability in an attempt to avoid the generality problem, but as I have already noted, reliability demands externalism, and externalism is a failed attempt to achieve an objective connection to external processes – a objective that is impossible to achieve. A version of internalism is what is needed, though the formulations presented in epistemological writings thus far (such as Lehrer’s) are insufficient. In any case, Alston’s reliability remains unconvincing.
Quine, the only philosopher besides Davidson (and Geach) who is very near the mark, attempts to naturalize epistemology by merging it with psychology and semantics. Here Quine may have been correct, if he had considered a bi-level epistemology in his naturalization, and if he had included evolutionary and biological factors, such as universal cognitive principles (rather than culturally divergent cognitive principles, as Stich may have suggested). Kim’s objection to Quine, though true relative to a linear account of epistemology and certainly applicable to the majority of Quine’s theory, is no objection to a theory that includes axioms, or bi-level epistemology. Kim is correct in arguing that naturalized ethics would make little sense since ethical relativism (just what we see) would be impossible, but this says nothing about why Quine’s epistemology is false. Having distaste for a world where we cannot be justified in our beliefs is not a valid objection to Quine’s theory. In essence, Kim says, “without justification we couldn’t have any real, absolute, or certain knowledge,” to which Quine and I say, “Tough.” On Kim’s rebuttal to Quine’s psychological approach: I do concede that Quine was a bit astray, but he was very near the mark. I would like to think that Quine realized that epistemology is in need of revision, and that philosophers are asking all the wrong questions. This single fact alone makes his theory invaluable to philosophers.
Goldman’s “Folkways” essay fails on account of its reliance on externalism, but the “virtue” of Goldman’s argument is the concept of a cognitive virtue or vice. Here Goldman seems to be on the right track, though he makes a mistake and attempts to use cognitive principles to support an erroneous theory, namely, virtue theory. Though his paper is a skillful rebuttal to reliabilism, it remains insufficient in addressing the skeptic. Other than the aforementioned, there is not much to be said, concerning “Folkways”.
Plantinga likewise offers little to the discussion. He argues for reliabilism and externalism, and rejects coherentism – for little rational reason. Plantinga’s objections to coherentism amount to something like: “I don’t like the fact that coherent webs might be necessarily circular, and I don’t like the fact that coherentism is not foundationalism”. To my mind, Plantinga’s arguments are entirely created according to the presupposition with which he began: absolute and ultimate knowledge must be true, and coherentism must be false. His work with “warrant” does nothing to mask this fact, and we are left with an insufficient (albeit refined) reliabilist theory. Interestingly enough, Hillary Putnam and Plantinga both object irrationally; Plantinga’s objection to coherentism amounts to incredulity, and so does Putnam’s objection to Quine’s naturalization.
Zagsebski’s virtue theory is as arbitrary as they come, though I respect her attempt to ground epistemology in virtue ethics. In the end, like all virtue theory, her epistemology is insufficient. Like “Folkways,” the theory is so insufficient (for the same reasons as all other virtue theories) that there is not much to write, concerning her work.
Lewis’s “Elusive Knowledge” was enlightening; not because it contains a sufficient theory, but because it touches on why epistemology needs to be refined. Though Lewis fails to refute skepticism since he relies on contextualism, (which does not seriously address the skeptic’s questions, but simply seeks to avoid them, since the skeptic’s questions are not about context disagreements, but are concerning the quality of our evidence) he does help to illuminate bi-level epistemology, or, in other words, he illuminates to us there are two levels of cognition or “knowing” to discuss. By mentioning that the skeptical argument switches “context” mid-way, he hasn’t pointed out any flaw in the argument itself, but has, rather, pointed out the boundary between our axiomatic assumptions, from which our propositions arise, and the beyond – the realm that is hopelessly divided from our minds, yet tied to it – the realm that creates and crafts our universal cognitive principles (that realm that we assume exists). All in all, although enlightening, Lewis’s arguments are insufficient, and contextualism fails to answer the skeptic.
Stich fails to address the skeptic simply because of his unsupported theory of diversion among cognitive principles. The rest of his arguments are near the mark, however, since he focuses his gaze right where we should all be looking: at the functioning of the mind itself, and the principles – crafted by natural selection – that drive the mind. Stich’s arguments are an excellent rebuttal to analytical philosophy, but they are not the right rebuttal, for the existence of universal cognitive principles is highly supported and corroborated by science and empirical evidence (as outlined in one of my previous essays). Without his erroneous assumption of cognitive diversity, Stich is left with no real argument. In the end, Stich too is “left holding an empty sack”.
Williams’ arguments are not of much worth, to my mind; they avoid the problem entirely, and they make the childish claim that philosopher’s are simply “pessimistic” concerning skeptical paradoxes. I would rather like to think that philosophers are being realistic, rather than pessimistic. Williams says, “we have no reason for conceding that there could be no test for determining whether or not we are dreaming” (438) but in so saying he ignores skeptic’s real meaning, and the reality of neuroscience and human cognition. Epistemic circularity is a scientific, testable, and confirmed fact – not a guess – and Williams’ assertion that we are pessimistic for noticing this fact is, to say the least, disingenuous. He ignores science and recasts the skeptic’s arguments into absolute terms – terms that the skeptic never implemented. No skeptic anywhere ever said that it was absolutely impossible to escape epistemic circularity. The skeptic argues that all of the evidence so far (neuroscience, evolutionary biology, regress justification, etc.) leads us to believe that epistemic circularity is a scientific fact. If this is the case, then skepticism holds. Instead of succeeding in his attack on skepticism, (or revising philosophy) Williams only succeeds in weakly arguing for contextualism.
Now that I have discussed the failings of each epistemological theory, it is my task to set forth an alternative. In this, I will not write much, as my instructors and (those who have read my previous essays) are already fully aware of my arguments (though I may have been deficient in my explanation of those arguments). I will give only a brief summary of the sort of revision I think is required, if epistemology is to ever come to a place where it can address the questions of the skeptic.
Skepticism is unavoidable considering one level of epistemology while it is not, considering another. By this I mean to say that epistemic circularity is a necessary fact of cognition, not simply a sufficient fact, and that we are forever trapped within our subjectivity. Since this is the case, certainty about anything at all is impossible. Knowledge, as it is traditionally defined, is impossible to possess. Nevertheless, we “know” a few things, (in this case “know” takes on a propositional meaning, rather than an absolutist empirical meaning) such as certain logical propositions. It is no mystery how we can both “know” and “not know”, if we consider axiomatic knowledge. Starting with our basal assumptions, those assumptions that are the product of our universal cognitive principles, we form a coherent network of beliefs that is functional, and this is all we need to think and form beliefs. This modus operandi functions perfectly, without any actual (finite, absolute, ultimate, certain) truth. Truth, it seems, is irrelevant; only the coherence and the functionality of a belief system are of any importance.
For those that insist that there must be some foundational source of all things, we must go back again to our first level of epistemology – that level that concerns the external world, and we can conclude with high probability, since we have already assumed that our axioms are necessary, (assuming is the only way we can conclude anything) that we are in fact inferring things from an external world of some kind, whether it be a computer simulation or the like, and that the way we think about that world is determined and programmed by natural selection and the evolution of the human brain, (I have talked about this extensively in another essay and provided reference to the research that supports this theory) and that this programming of our mind results in universal cognitive principles that control and craft the way we think. In the case of the Humean pool table, in other words, we may just know what the cue ball would do, without any experience, since we are not a tabula rasa.
Now, all of this (further explicated in my other essays) leads us to the conclusion that epistemological questions are malformed, since they are linear, (even the philosophers who argue for a bi-level epistemology I view as linear epistemologists) and assume an ultimate, objective state of things (foundationalism) – or answer only part of the question, (coherentism) – and because they are hopelessly tied to either internalism or externalism, when the truth is somewhere in between. Coherentism comes close to the mark, but as many of the foundationalists have pointed out, it leaves us wanting some source that would help us explain how our coherent webs are created. Moreover, despite the claims of many philosophers, modern BIV paradoxes (and other epistemological paradoxes) remain unsolved. For example, the Theseus’ ship paradox has resulted in a plethora of supposed solutions, but none are sufficient in dispelling the paradox (although I believe that I have a solution to the paradox that is sufficient. See my essay on the Theseus’ paradox). In sum, epistemologists have failed to seriously address the most serious questions of epistemology.
All of the aforementioned failures are due, in my opinion, to an erroneous conception of epistemology itself – and this is exactly why our theories are in need of revision. How epistemology should be revised is a question much easier asked than answered. Nevertheless, I feel that I am at least somewhat on the right track. Others agree with most of what I believe – for example, Dr. Geache – and as I have mentioned, many philosophers, such as Quine, Davidson, and others, are very near the mark. I believe now more than ever that the key to answering epistemology’s most difficult questions is a revision of epistemology itself. We must, as Quine suggested, naturalize epistemology into semantics, and psychology – but do so whilst considering a bi-level epistemology, complete with axioms, universal cognitive principles, and modern empirical evidence from the fields of neuroscience, and evolutionary biology. My independent study has strengthened my resolve to revise epistemology in this fashion.