Reliabilism’s causal component fails

By Aristophanes

Many epistemologists seek to determine the nature of justification, an integral part in a substantial number of popular theories of knowledge. Among them, Alvin Goldman, a professor of philosophy at the University of Arizona, attempts to explain justification through a “causal requirement.”

Goldman, who set out with the goal of defining justification in non-epistemic and explanatory terms, concocted a theory he termed “reliabilism,” in which a belief is justified if and only if it is formed through a reliable belief-forming process. Under reliabilism, an externalist theory, belief requires a causal connection to the evidence available for that belief. The theory is externalist because, unlike an internalist theory, it does not require that the subject know that he knows a proposition — the external world plays a key role in the nature of justification by providing possible support or potential defeat of a knowledge claim that can be independent of a subject’s own mental state.

“A religious fanatic may be psychologically incapable of doubting the tenets of his faith, but that doesn’t make his belief in them justified,” Goldman says. What does make them justified would be their specific relation to facts about the external world.

Keith Lehrer, an opponent of all externalist arguments for the nature of justification, naturally opposes Goldman’s theory. Reliabilism commits a grave error, which Lehrer calls the “causal fallacy.” In this, he claims the theory confuses a subject’s reason for believing a proposition with the cause of a subject’s belief of a proposition. Ignorance of the reliability of a reliable process could defeat knowledge, Lehrer says.

“Accounts of knowledge in terms of causation or the receipt of information fail to provide an account of our knowledge of general and theoretical truths,” Lehrer says. “Causal or information-receiving analyses of knowledge have the virtue of explicating knowledge in a way that explains the connection between truth and belief, between reality and thought, and provides an answer to skepticism. We may, however, maintain the connection between truth and belief without committing ourselves to a restrictive causal connection.”

To illustrate, Lehrer presents the case of racist Mr. Raco, who believes, due to his prejudice, that all members of his race are immune to a certain disease while members of another race are not. As it turns out, Mr. Raco is also a doctor who goes on to study the same disease in depth, a decision resulting from his obsessive prejudice. The medical facts clearly state that Mr. Raco was right; members of his own race are in fact immune to the disease while members of the other race are not. Mr. Raco thus has reason to believe this proposition because of established medical fact, although the causal chain that led him to his belief originated with an unwarranted prejudice. Surely, however, our intuition tells us that Mr. Raco still knows the proposition.

Lehrer would say that, as in the case of Mr. Raco, reliabilism fails because we can sometimes have a justified belief that is causally independent of the evidence justifying it. The distinction is between why someone would believe something and how they would know it. For Mr. Raco, he does know the truth of his race’s immunity to a disease because of a causal chain that began with his prejudice. That belief is confirmed through the reliability of external medical fact, not caused by it, as it was Raco’s racist tendencies that did the causing.

The reliabilist, who only focuses on the why aspect of justification, will still draw a correct conclusion in most cases, Lehrer says. However, in the few where why and how contradict, the reliabilist’s refusal to consider the distinction between former and latter will lead him to a result that opposes our intuition. Knowledge requires something other than the mere possession of correct information obtained reliably, Lehrer says.

“All externalist theories share a common defect, to wit, that they provide accounts of the possession of information rather than of the attainment of knowledge,” Lehrer says.

In response to this objection to reliabilism, Goldman would likely clarify that his theory does encompass the correct intuition in the Mr. Raco case, and that Lehrer has merely constructed a straw man version of his argument. Mr. Raco’s belief was caused by unwarranted prejudice, which would be an unreliable belief-forming process. Because of this, reliabilism would, at first, lead us to think that the belief is not justified. And yet it surely is, as it is backed by conclusive medical evidence. To amend this, the reliabilist can alter his theory to not only encompass what originally caused a belief, but also what sustains it. In this instance, the theory is sustained by the conclusive medical evidence. The reliabilist may even try to argue, in order to preserve the causal requirement, that for something to sustain a belief is in some manner to still cause, if not the original forming of the belief, then the belief’s continuation. What we have here may be two “causes,” one that forms the belief and one that sustains and protects it into the future.

Goldman’s theory is historical, meaning it makes justification of a belief reliant upon its prior history. This leads directly into the theory’s causal requirement — the relationship between cause and effect naturally entails that prior, historical events must be taken into account when discussing whether something is adequately justified. This is in opposition to what Goldman would call “current time-slice” theories (a phrase he borrows from Robert Nozick). In a current time-slice theory, the existence of sufficient justification for a belief is merely a matter of what is true of the subject at the time of belief, not how the justification came about.

To Goldman, everything from which we rightly form a belief can be explained through a causal chain, even our processes of perception and introspection. Introspection, particularly, should be regarded as a form of retrospection, says Goldman.

“A justified belief that I am ‘now’ in pain gets its justificational status from a relevant, though brief, causal history,” Goldman says. “The psychological process of ‘seeing’ or ‘intuiting’ a simple logical truth is very fast, and we cannot introspectively dissect it into constituent parts. Nonetheless, there are mental operations going on, just as there are mental operations in idiots savants, who are unable to report the computational processes they in fact employ.”

Goldman puts forth the idea that even rapid process such as seeing and hearing are still effects of some cause. Light reflected from a nearby stove hits my eye, causing my brain to form an image that results in my seeing the stove. Each successive step in this chain is caused and relies on the step before it. The same is true of introspective beliefs. I take a mathematical fact, that two plus three equals five, for example, and believe this only because that belief was caused by my prior knowledge of myself as a trustworthy calculator of simple arithmetic.

You can take Goldman’s belief here and apply it to the Mr. Raco case. Lehrer claims that Mr. Raco’s belief is caused by his prejudice but justified by medical fact. However, if we take Goldman’s claim of introspection and perception as part of brief causal chains, then we could argue that Mr. Raco’s belief is caused by the medical facts he reads, but only in the sense that they work to sustain the belief in question. In this, Mr. Raco is first unjustified in his belief that all members of his own race are immune to a certain disease when he bases it only on his own, unwarranted prejudice. In an externalist system, it does not matter that Mr. Raco does not know that his racism is unwarranted — he still is not justified in his belief by the outside fact that his belief is not warranted. However, when Mr. Raco begins reading the medical reports, discovering that they line up perfectly with his earlier unjustified belief, that same belief becomes justified. Reading the medical studies causes Mr. Raco to reinforce his belief in a way we would determine as justifiable. If Mr. Raco is a trustworthy observer, which we assume he is, then the belief was justifiably sustained through a causal, reliable process. Reliabilism is saved.

I believe this theoretical response, while possibly adequate in the case of Mr. Raco, still fails to ultimately protect reliabilism from further criticism to its causal component. Take, for example, some beliefs that are self-intuitive, such as the idea that “I now seem to see a red car outside my window,” a belief that is true. My justification for this belief is uncaused — it just is the case. Now, if we were to amend the belief to one that was simply “I now see a red car outside my window,” the justification for that belief would in fact be caused by the external world in which a red car actually does sit outside my window, and I am a reliable observer. This is because the justification of the first belief is in a higher order state of which I cannot be in error, while the other I easily could be (perhaps I am hallucinating or merely dreaming). If my justification cannot be in error, then the justification of the belief is incorrigible. Incorrigible justification cannot be caused, as the justification exists as a component of the belief itself; they are tautologies in justification, being self-justified by their very nature. The justification is self-caused, and thus cannot take part in any meaningful, historical causal chain. Thus, reliabilism fails to explain why I am justified in my belief that I seem to see a red car outside my window. This would hold true for other theoretical or mathematical beliefs as well, as Lehrer mentioned previously.

However, an internalist view of justification does explain the justification for my red car belief. Because an internalist views justification as pertaining only to a matter of one’s mental state, independent of the outside world and any causal chain, it is compatible with incorrigible justification of certain basic beliefs. To an internalist, it does not matter whether there actually exists a red car at all — I could be dreaming, hallucinating, or under the influence of an artificial experience-making machine, and the theory would still explain my justification. Anyone else placed in the same exact mental state as I would hold an equal amount of justification. Change the external world as you please, but my level of justification would not change without an alteration of my internal mental state. This is in line with our intuition on the nature of knowledge, and thus holds as a more coherent theory than reliabilism and its causal component.

“Externalism is motivated by the doubt about whether what we accept can supply the truth connection,” Lehrer says. “The reason for the doubt is the assumption that it is psychologically unrealistic to suppose that beliefs about our beliefs are necessary for knowledge. Such higher order beliefs about beliefs are not, of course, necessary for receiving and relaying information. Even a thermometer is capable of that. Such beliefs are, however, necessary for knowledge.”

Thus, a causal requirement fails because it cannot account for the multitude of reasons one may have for justifiably forming a belief. Through some careful argumentation and alteration, the causal-supporting externalist may be able to convince us that his theory works in more cases than we would originally assume, such as in the Mr. Raco case. However, no matter how many more cases the theory adequately explains, eventually we will come upon certain cases, such as the red car case, in which it cannot apply due to the presence of uncaused, ahistorical, basic knowledge of a higher order. ■

Home → Philosophy


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s