Not all of our beliefs are justified, nor are all of our justified beliefs to the same degree. That our beliefs are justified is something that concerns us and should concern us as epistemic agents, since it is assumed that if a belief is (well) justified, if we have good reasons for it or have formed it in an appropriate way, this will make its probability of being true increases. In this way, justification is considered a fundamental epistemological good. Now, what is the fact that a belief is justified? What kinds of facts, properties, states, or conditions can confer justification on a belief? What is the nature of justification?
If the distinction between coherentism and foundationalism —or infinitism— (see Epistemic Justification) referred to the structure of justification, the distinction between internalism and externalism refers to its nature, that is, to what type of factors can act as justifiers of a belief. This distinction is relatively recent, it appears for the first time in (Armstrong, 1973) and, although there it refers to the different types of definitions of knowledge, it has subsequently been imposed as a way to distinguish between two types of theories about the (nature of knowledge). the justification. According to the most generally accepted conception, the internalist conception of epistemic justification is one that defends that only what the subject has or can have cognitive access to (through introspection or reflection) can justify a belief; in general, then, only mental states (experiences, memories, other beliefs) can be justifiers. Externalism is the negation of internalism: it is not true that only what is cognitively accessible can be a justifier. For example, if A forms a certain belief from reading a news item in a serious newspaper and forms another belief by reading a tabloid newspaper, his first belief will be justified and the second will not, although the subject does not know that one of the newspapers deserves trust and the other does not. The objective reliability of the source, whether or not it is known to the subject, is relevant to the justification of the belief.
We can say that our appeal to justification responds to two different intuitions or presents a double aspect that, unfortunately, is not always easy to match. It is this double aspect that favors the different positions (internal and external) that we are dealing with here. If what centers our interest with respect to justification is what makes it reasonable for the subject to acquire or maintain a belief, our position will tend to be internalist. If, on the contrary, we emphasize what increases the probability that the belief is true, externalism will appear to us a more attractive position.
As we have characterized this distinction, we can see that internalism is the most demanding position, only what is internal to the subject's mind can justify a belief. To the extent that externalism is the negation of internalism, it could seem that it is more permissive: there would be phenomena that, without being cognitively accessible to the subject, nor being mental states of the latter, could be justifiers. However, there are two types of externalism: 1) that which considers that the internalist condition is not sufficient for justification; that is, in addition to what is internal to the subject's mind, some external condition is needed, something that does not have to be accessible to the subject, (which would be a more permissive externalism than internalism) and 2) which holds that internalist conditions are not only not sufficient for justification, but they are not necessary: it is only some condition that does not have to be accessible to the subject (such as the reliability of the process by which they have been acquired) that determines the justification of the beliefs. In this second case, externalism would not be more permissive than internalism, but would simply introduce demands of a different nature.
Internalist theories of justification
Within internism we can also distinguish two types, according to their degree of demand: 1) what we can call status internism and 2) simple internism. Both types, as interns, agree on:
a) Only what is internal to the subject's mind can be a justifier.
But while simple internism requires only this condition, status internism further maintains that:
b) Nothing can be a justifier unless the subject is (or can be) aware that it is (that is, of his justifying status).
That is, for simple internalism it is not necessary that what justifies the belief is recognized as such by the subject, it is enough that it be cognitively accessible by mere reflection. Thus, according to this position, a perceptual experience, for example, can serve as a justifier of a belief. My perceptual experience of a tree before me justifies my belief that there is a tree before me. On the other hand, for status internism (BonJour, 1985, ch. 2; BonJour & Sosa, 2003), this same experience cannot justify said belief if, in addition to having it, I am not aware (currently or potentially) of the justifying relationship between one and the other; that is, if I do not believe in turn that such an experience makes it probable that there is a tree before me. In this way, an important difference between one type of internism and another is that status involves a doxastic conception of justification: it maintains that our beliefs can only be justified by other beliefs. On the other hand, simple internalism admits that also experiences, memories ..., can be justifiers. Simple internalisms are dogmatism (Pryor, 2000, 2001) and evidentialism (Feldman and Conee, 1985; Conee and Feldman, 2001).
Pryor's dogmatism rightly holds that such things as experiences and memories are immediate justifiers and that when a belief is based on them it is immediately justified. For its part, evidentialism maintains that (in the formulation of Bergmann, 2004, which corrects that of Fedman and Conee, 1985): “The belief C of S is justified if and only if C is an adequate doxastic response to the evidence of S ”(Bergmann, 2004, p. 35). That is, the subject's belief will be justified if it occurs in response to the reasons, experiences, memories, etc., that he possesses.
The problem with these theories is that they do not require as a condition for something to be considered evidence that it has to be a reliable indicator of the truth of the belief. For both dogmatism and evidentialism (and, in general, for internalism) an experience justifies whether it is a true perceptual experience or a delusional one. In other words, if two subjects are in the same mental state, it cannot be that the belief of one is justified and that of the other is not. For example, if the "perceptual" experience that a subject would obtain in the presence of a cat justifies him to believe that there is a cat in front of him, then if he has a qualitatively identical experience, this will also justify him to believe that there is a cat in front of him. , even if it is a case of delusion and there is really no cat before him. Even if we were brains in a vat or were systematically deceived by an evil Cartesian genius, our beliefs, insofar as they correspond to our evidence, would be justified, even if all or most of them were false. At the end of the day, what has caused the belief is not something that (at least not always) is within the cognitive reach of the subject, nor is it internal to his mind, with which he cannot influence the justification according to the internalist positions. But if evidence, and therefore justification, does not increase at all the probability that our beliefs are true, then why do we care? What epistemic value would they have?
On the other hand, with regard to evidentialism, as in this conception of justification only the evidences available to the subject are taken into account and not their responsibility and efficiency when acquiring them, it may happen that there are many counter-evidences of which the subject does not have due to having been negligent and, despite everything, his belief would be justified. Epistemic irresponsibility could favor justification.
As we can see, internalism places the emphasis on that aspect of justification that has to do with the subject's relationship with his belief and with the reasonableness of acquiring or maintaining a belief from such a perspective, and largely forgets the relationship of justification. with the truth, her role as conductor of the truth.
2. Externalist theories of justification
In general, when speaking of externalist theories, only those of the second type are usually considered as such, that is, those that defend that the justifiers of a belief are external to the subject's mind, that they do not have to be cognitively accessible. .
These include proper functionalism and, above all, reliabilism. According to the first of these theories, a belief is justified if it has been formed by properly functioning cognitive faculties (Plantinga, 1993; Bergman, 2006). For its part, reliabilism maintains that the justification of a belief does not depend at all on the reasons available to the subject, nor on the logical connections that the content of the belief maintains with the other beliefs of the subject, but on the fact of that the process that generated it is reliable, that is, that it provides (in the right circumstances) a high percentage of true beliefs. According to its highest representative, Alvin Goldman, (Goldman, 1976, 1986) this is all that is needed for a belief to be justified. Reliable processes are, in general, perception, memory, reasoning and introspection. Thus, for example, if under normal observation circumstances, I acquire the belief that there is a book before me as a consequence of my perceptual experience of the book, my belief will be immediately justified, for under such circumstances perception is a reliable process of understanding. obtaining beliefs.
A theory that straddles both types of externalism is Sosa's epistemology of virtues (Sosa, 1991). According to this author, a belief is justified if it has its origin in an intellectual virtue, that is, in a faculty (perception, memory, reasoning) that generates, in the right environment, a high index of true beliefs. Now, Sosa maintains that this type of justification is sufficient for "animal knowledge" (one in which beliefs are direct responses to the impact of one's own experience), but not for "reflective knowledge", which requires a " epistemic perspective ”, which is constituted by the beliefs that the subject has about the reliability of their faculties, etc. And, he argues, "reflective knowledge is better justified than animal knowledge."
We can see, then, that the fundamental interest of externalism is the connection of belief with truth. The justification of the belief must be an indication, a symptom of the truth: justified beliefs are much more likely to be true than those that are not justified. Justification has to do with the relationship between belief (the process that generated it) and the world. The difference between one type of externalism and another is whether only such a connection is required or it is also considered that the subject must have reasons or evidence in favor of his belief.
If internalism prioritized the relationship of the subject with the belief and the reasons that it had in its favor, centered its interest in the fact of whether it was reasonable to believe given the reasons that were possessed, externalism focuses on the relationship of belief with the environment and the connection of justification with the truth. But the externalist theories that we are seeing also have a clear drawback. If the subject's belief is justified from the external point of view, but he has no reasons that he can offer (not even to himself) to justify it, his epistemic position seems weak. Having a justification that is completely ignored does not seem like an adequate justification. Therefore, we have that, as Comesaña (2010, p. 571) says, “reliability without evidence is blind, evidence without reliability is empty”.
It seems, then, that an adequate theory of epistemic justification must incorporate both internalist elements (having evidence) and some externalist element. That is, it seems necessary to include some external element that helps to specify what constitutes evidence and what does not. If, for example, I say that there is a cat in front of me and I am asked how I know it or why I believe it, the fact that I say that I know it because I see it indicates something in this regard. I have said that I know because I see it and not anything else because I consider my experience as evidence in favor of the belief that there is a cat before me. I am assuming that my visual experience makes it likely that there is a cat before me. Well, we could demand that for something to constitute evidence in favor of a belief it must make the truth of the belief really probable. The intuition that the brain's perceptual beliefs in the vat would also be justified remains to be resolved, although their experiences did not make it probable that they were true. Indexical reliabilism aims to solve this problem . Ultimately, according to this theory, it is a matter of saying that the justification of beliefs depends on the reliability of the process in our world. Thus the beliefs of the brain in the vat will be justified because perception is a reliable process in our world (although it is not in yours).