Ways of Knowing, Part 2
What is knowledge? The traditional analysis, dating back to Plato and commanding the adherence of many great minds over the years, has been that knowledge can be defined as justified true belief (henceforth JTB). In other words, if I believe something, that belief is true, and I have some justification for believing it, then I have knowledge. Stated symbolically:
S knows that p iff:
(i) S believes that p
(ii) P is true
(iii) S is justified in believing that P.
If any of these conditions were missing, it was thought that knowledge was missing as well. Then along came a little three page tornado by Edmund Gettier, entitled simply “Is Justified True Belief Knowledge?” In it he pointed out how you could have a justified true belief that did not entail knowledge. His examples entailed someone having good reasons for believing something, which then led them to a further implication of that belief, which turned out to be true while the initial justified belief turned out to be false. Yet at the end of it you had a justified true belief.
Gettier’s first example went something like this: Jones and Smith both apply for a job. Smith comes to the following beliefs:
1) Jones will get the job. (This is based off the fact that the president of the company has told Smith that Jones will get the job.)
2) Jones has ten coins in his/her pocket. (This is based off the fact that Smith saw Jones count the coins in his/her pocket and then put them back in.)
3) Therefore, a person with ten coins in their pocket will get the job.
Smith believes #3 via a logically valid argument, and based off of justified beliefs. Now as it turns out, the president of the company lied to Smith and instead gives Smith the job. Also, unbeknownst to Smith, he has ten coins in his pocket. Therefore, Smith believes #3, is justified in doing so, and his belief is true. Yet we would not wish to say that he had knowledge.
In the years since this paper, epistemologists (philosophers who study knowledge) have made many attempts to solve what is now called the “Gettier problem.” Many of these have involved trying to find a fourth condition, in addition to JTB, that would get us a proper analysis of what knowledge is. So far no one theory has gained the consensus.While many theories focus on evidence and proof as a key component for knowledge (i.e. “justification”), another line of thought seeks to study the mechanisms that produce beliefs, and to determine what it would mean for them to be able to give us knowledge. On these views, our third condition or cluster of conditions for knowledge would not be justification, but rather something else. These naturalistic accounts, also know as externalisms, seem to me to hold a lot of promise, none more so that the theory advanced by Alvin Plantinga, know widely as the theory of Proper Function, which outlines the things which provide “warrant” for a belief. According to Plantinga, if a person has a true belief that has been produced by cognitive faculties that are functioning properly, according to a design plan aimed at producing true beliefs, in a cognitive environment suited to those faculties, then they have knowledge. For example, if you believe that when you approach your car you actually see it, this will be knowledge for you if:
(i) You believe that you see your car.
(ii) You actually are seeing your car, not a mirage or a hologram.
(iii) You belief is produced by your visual cognitive faculty, that faculty is functioning properly, is designed to produce true visual experience, and is being utilized on earth, with sufficient lighting conditions being present, not during a black, cloudy, moonless night or an alternate universe where there are no light waves to produce the sensation of sight.Given these things, you have knowledge that you see your car.
While this sounds pretty common sensical (indeed, Plantinga traces his own theory’s development back to the common sense philosophy of Thomas Reid), there are some problems with it. My main concern deals with the way the major terms (proper function, a design plan aimed at producing true beliefs) in Plantinga’s theory are set up so as to presuppose theism. Of course Plantinga claims that either God or evolution could have produce these faculties, but then later spends the last several chapters of his work Warrant and Proper Function trying to demonstrate that proper function cannot be explained in purely naturalistic terms. You can’t have both.
What I propose is that, for our purposes, we replace the term proper function with sufficient function, a term that encompasses the actually reality faced by our cognitive faculties, as opposed to forcing us into either/or situations were faculties are either functioning properly or they are not, with no possibility of a gradient somewhere in between. Whereas proper function stipulates that our faculties must be working as designed, sufficient function claims they must be working well enough to do what we are asking of them. In addition to this redefinition, I would propose that instead of thinking of the design plan as having to be aimed at truth, we should think of it as regularly producing true beliefs. This takes us off theistic clouds, and back firmly on naturalistic ground. This does not mean we have to dismiss theism as a possibility, simply that we do not guarantee it by the way we define our terms.Also, besides avoiding a guarantee of theism, the above way of defining our terms gives them a far broader range, allowing us to take into account that the cognitive faculties of certain beings might work better or worse than that of another being, or that specific faculties might be stronger or weaker between species, or even within individual members of a species. For example, if your eyesight is 20/20 and mine is say 20/70, we can both gain knowledge via our faculty of eyesight as long as my eyesight is sufficiently good enough to produce the true belief in question. In other words, 20/70 vision might not be considered proper function for a human being, or even typical/average function, but it can be considered sufficient function for the forming particular bits of knowledge, though it is perhaps not good enough for others. <>In the same vein, to say a faculty must be aimed at truth would preclude the possibility that a faculty could actually be favored by evolution for a completely different purpose. For example, it is possible that an elephant’s ears have mutated as they have due to the fact that they help in releasing heat, better allowing the largest land mammal to regulate its body temperature. A side effect may well be that elephants have even better hearing than they would have otherwise. However, their adaptation was not aimed at true auditory beliefs, but rather at better body regulation. On our new revision of Plantinga’s theory, however, it would be enough for the faculty in question to be able to regularly produce true beliefs, even if it was not entirely aiming for them.
Given these adjustments, I think we have the beginnings of a good working theory of what constitutes knowledge. We will need to examine how it incorporates our cognitive faculties, and how the way we form beliefs gives them varying degrees of warrant. Then we can begin to apply it to our many beliefs, and see which of them can be considered to be knowledge, and which cannot.
To restate the position, in conclusion: a person has knowledge if her belief is true and was generated by a suffiently functioning cognitive faculty(ies) having a design plan(s) that regularly produces true belief(s), and that generated that/those belief(s) in a receptive environment.

