KI mini-Independent Study
Examining the forgotten option:
A closer look at infinitism in relation to the regress problem
The problem of infinite regress threatens to show that none of our beliefs are justified and that none of them qualify as knowledge. The two leading theories that have been proposed to solve this problem, foundationalism and coherentism, each face considerable challenges in explaining how knowledge comes about within the respective frameworks. Amidst the debate of the validity of either theories, Professor Peter Klein has emerged to defend an unpopular position known as infinitism, the view that it is acceptable for knowledge to have a chain of justification that is “infinite and non-repeating” (Klein). If Klein is right, the regress problem would cease to be a problem, and our knowledge would prove to be secure despite there having no end to our justifications. In my essay, I will outline and evaluate Klein’s justifications for his theory, and, by raising further objections to infinitism, show how his theory unfortunately faces problems as well. Ultimately, I aim to show that infinitism, as it stands, is not a tenable solution to the regress problem.
Why should we believe Klein that knowledge claims can have a chain of justification extending infinitely, without there being a set of ultimate, foundational reasons? Klein offers two principles of good reasoning that, he claims, when taken together, entails that infinitism is right.
His first principle, Principle of Avoiding Circularity (PAC), states that
for all x, if a person, S, has a justification for x, then for all y, if y is in the evidential ancestry1 of x for S, then x is not in the evidential ancestry of y for S. (Klein)
Here, Klein is formalising a principle that prohibits circular reasoning. For example, I would not be able to prove validly that I am the tallest student in class by saying that everyone else in class is shorter than me, when the proof for this is that I am the tallest student in class. By prohibiting the conclusion to be presupposed by the premises, Klein is obviously justified in formulating this principle of good reasoning.
His second principle, Principle of Avoiding Arbitrariness (PAA), states that
for all x, if a person, S, has a justification for x, then there is some reason, r1, available to S for x, and there is some reason, r2, available to S for r1; etc. (Klein)
Here, Klein is formalising a principle that requires that for a person to have a justification for a claim, a reason must be available to him to support that claim, and another reason must be available to him to support that reason, and so on. If this principle is right, there should not be any reasons in this chain for which justification is not needed. This seems like a sound principle, for if we accept that justification is not needed for a reason in this chain, it seems that we can be arbitrary in our reasoning, and prove any contingent proposition to be true. For example, we can “prove” that all bachelors are blond by saying that all bachelors are left-handed and all left-handed men are blond, and accept the latter proposition without a reason, in violation of PAA.
Klein then moves on to claim, “the combination of PAC and PAA entails that the evidential ancestry of a justified belief be infinite and non-repeating. Thus, someone wishing to avoid infinitism must reject either PAC or PAA (or both)”. This seems to follow from his argument: save for the sceptical conclusion that no one ever has any justification in believing anything, PAA necessitates that the chain of justification be infinite or finite but circular. However, PAC eliminates the latter option, leaving the only option of an infinite and non-repeating chain of justification, that is, infinitism.
In sum, this is how Klein argued for why we necessarily ought to embrace infinitism. However, two major objections threaten the tenability of infinitism.
Firstly, it appears that through infinitism, any contingent proposition can be proven. For us to see why, Audi suggests we
take the obviously false proposition that I weigh at least 500 pounds. I could back up a belief of this by claiming that if I weigh at least 500.1 pounds, then I weigh at least 500 (which is self-evident), and that I weigh at least 500.1 pounds. I could “defend” this by appeal to the propositions that I weigh at least 500.2 pounds, and that if I do, then I weigh at least 500.1. And so forth… (189–190) [emphasis by the author]
Since this chain is infinite, every claim can be “defended” by further claims in the same fashion, and nothing in this argument is unsound. However, if infinitism turns out to accept any contingent proposition as justified, then surely we cannot consider it as a tenable structure of justification.
Klein defends infinitism by claiming that setting up the structure for a claim as above is a necessary condition to establish justification, but not a sufficient condition. For there truly to be a justification for a claim, the reasons for the claim must also be objectively and subjectively available to the subject. Thus, Klein would say that the proposition that I weigh at least 500.1 pounds is not available to the subject, and thus cannot serve as a reason that I weigh at least 500 pounds. However, while subjective availability is defined as “properly hooked up with S’s own beliefs” (Klein), Klein gives seven different definitions for objective availability, each containing vague terms like “sufficiently high probability”, “deepest epistemic commitments” and “intellectually virtuous person”. Ultimately, what Klein means by “availability” is extremely unclear, and hence his arguments do not defend infinitism effectively. In other words, the argument that any contingent claim can be defended using the framework provided by infinitism still stands.
The second major objection concerns the validity of PAA when applied to certain propositions. Klein challenges foundationalists to examine the set of propositions that they consider basic and self-justifying, and raises the question, “What makes these propositions truth-conducive?” That is, why are these propositions likely to lead to the truth? Klein claims that, for this question, either there is an answer or there is no answer. If there is an answer, then the answer would serve as a reason to justify the supposedly foundational belief, and thus the infinite regress continues. If there is no answer, then Klein claims that the proposition conflicts with PAA and is arbitrary, and that it is not a good argument. However, consider the proposition P1 “If some dogs are pets, some pets are dogs” (189), which Audi suggested that no further reason explains. When hard pressed, we may suggest that the proposition P2 that this follows from our logic system to be the reason which makes P1 truth-conducive. However, beyond this, it is hard to see how P2’s truth-conduciveness can be established via reason. Klein would claim that since there is no reason for our belief that P2 is truth-conducive, we are being arbitrary. However, very few are willing to accept our logic system’s truth-conduciveness to be arbitrary. Hence, here, we have a case of a proposition without a further reason r, which we agree to be able to function as a reason for other propositions rationally. In other words, we have found what seems to be a counterexample to PAA. Since, by Klein’s own admission, infinitism requires both PAC and PAA to become a necessary option, counterexamples to PAA seriously threaten infinitism’s viability as a structure of justification.
As long as Klein is able to produce a clearer account of objective availability (to address the first objection) and defend PAA in light of any counterexamples (to address the second objection), there is still hope for infinitism. However, before these matters are resolved, infinitism does not seem to me to be a tenable solution to the regress problem, or a functioning theory of justification.
1305 words
Works Cited
Audi, Robert. Epistemology: A Contemporary Introduction to the Theory of Knowledge. 2nd. New York: Routledge, 2003.
Klein, Peter D. “Human Knowledge and the Infinite Regress of Reasons.” Philosophy and Religion Department, Montclair State University. Jan 1998. Rutgers University. 16 Oct 2007 <http://chss2.montclair.edu/prdept/HK.htm>.
Endnotes
1 For Klein, if r3 is a reason for r2 and r2 is a reason for r1, then r3 and r2 are both in the evidential ancestry of r1.
