Anticipating Gettier

Few philosophical articles have generated as much scholarship as Edmund Gettier’s 1963 piece in Analysis “Is Justified True Belief Knowledge.” In the article, Gettier provides two examples of beliefs that are both justified and true, but which he correctly assumes most philosophers would not accept as knowledge. What appears to have been overlooked in the flurry of responses to Gettier is the fact that the insight he presents into our understanding of what it means to know something was not new. Bertrand Russell had articulated the very same insight fifty years earlier in his The Problems of Philosophy.

Gettier’s article includes two counterexamples to the definition of knowledge as justified true belief. Both counterexamples work in the same way. That is, both involve an individual’s holding a belief the truth of which is logically entailed by another belief they hold which is both justified and assumed, wrongly as it turns out, to be true, and both count on justification being conferred via the relation of entailment. Since both examples work in the same way, for the sake of brevity I’ll consider only the first.

In this example, two men, Smith and Jones, have applied for the same job. Smith is described as having “strong evidence” that Jones will get the job, and that Jones has ten coins in his pocket. Smith’s evidence is described by Gettier as being sufficiently strong to justify his beliefs that Jones will get the job and that Jones has ten coins in his pocket. The conjunctive proposition “Jones is the man who will get the job, and Jones has ten coins in his pocket,” Gettier explains, entails a third proposition that “The man who will get the job has ten coins in his pocket.” Since Smith can see that belief in the truth of this proposition is entailed from the proposition whose truth he is justified in believing, he believes this proposition as well.

Included in the justification of Smith’s belief that Jones would get the job was that “the president of the company assured him that Jones would in the end be selected.” But, of course, the president could change his mind. And that is precisely what happens in Gettier’s example. Smith, not Jones, gets the job. It doesn’t take a great deal of imagination to envision this scenario. No amount of justification for beliefs relating to matters of empirical fact, at least no amount of the sort of justification Gettier presents (i.e., the kind available to introspection) can guarantee the truth of such beliefs.

What is more surprising, but certainly not implausible, is that Smith just happens to have ten coins in his pocket. So Smith’s belief that “The man who will get the job has ten coins in his pocket” is, as it turns out, both justified and true. That is, the belief is justified because is was entailed by Smith’s belief that Jones would get the job and that Jones had ten coins in his pocket. And it is true, because he, Smith, happens to have ten coins in his pocket as well.

Gettier correctly assumed, however, that despite the fact that Smith’s belief that “The man who will get the job has ten coins in his pocket” turned out to be both justified and true, that virtually no one would be willing to say that it constituted knowledge.

Gettier’s point wasn’t really all that original though. The problem is that Smith inferred a true belief from a false one. That is, he inferred the truth of the proposition “The man who will get the job has ten coins in his pocket,” from his belief in the truth of the proposition that “Jones is the man who will get the job, and Jones has ten coins in his pocket.” Jones didn’t get the job, however, he did, and “a true belief,” Russell explained back in 1912, “is not knowledge when it is deduced from a false belief.”

Russell’s example of a true belief that is inferred from a false belief is much simpler than Gettier’s in that it makes no reference to justification. His example involves someone believing truly that “the late Prime Minister’s last name began with a B” based on his false belief Balfour was the late prime minister. In fact, the belief that the late Prime Minister’s last name began with a B was true at the time Russell wrote The Problems of Philosophy because the late Prime Minister, at that time, was Henry Campbell Bannerman.

The entailment condition is implicit in Russell’s example because the proposition that “Balfour was the late Prime Minister” entails the proposition “The late Prime Minister’s last name began with a B.” Again, Russell’s example makes no reference to justification, but this omission in no way weakens the strength of his insight that true beliefs inferred from false ones can’t constitute knowledge. This is precisely the insight put forward by Gettier, though it’s dressed up in his paper with the trappings of justification and even more baroquely with the contrivance of conveying justification via the relation of entailment.

So Gettier’s paper, despite the splash it made, and indeed, to some extent continues to make, conveyed no philosophical insight into our intuitions about what it means to know something that had not already be conveyed by Russell fifty years earlier.

The Paradox of Personal Identity

What makes us who we are? Our bodies? Our minds? Our moral character? “The most natural theory of personal identity,” writes the philosopher Richard Swinburne, is that it is “constituted by bodily identity” (Personal Identity, Oxford, 1984). There are obvious problems with this theory, though. The brain, seems more important than the rest of the body. A person can lose an arm or a leg and still be obviously the same person. Damage to the brain is different. A person who suffers severe memory loss, or a dramatic personality change, as a result of a brain injury or degenerative brain disease is not so obviously the same person they were before.

“The traditional alternative to a bodily theory of personal identity,” continues Swinburne, “is the memory-character theory.” According to this theory, it’s the continuity of our memories and our character that constitutes our selves. But which is more important, memory or character?

Our sense of self is closely connected with our memories, or more specifically, with the personal narratives we construct based on those memories. If the memories go, then so do the narratives. If we reach a point where we can no longer remember who we were, then the person we were is gone. There’s still a person there, of course, but it is, in an important sense, a different person, a new person, a person with no past.

Recent work in “personal identity theory” has shifted its focus from memory to character, or more specifically, to moral character. Research done by Nina Strohminger, a psychologist at the Wharton School of the University of Pennsylvania, and Shaun Nichols, a philosopher at the Sage School of Philosophy at Cornell showed that the extent to which someone suffering from Alzheimers seemed different to those close to them was determined almost entirely by changes in their moral character rather than by memory loss (“Neurodegeneration and Identity,” Psychological Science, Vol. 26 [9] 2015 pp. 1469-1479).

The primary aim of Strohminger’s and Nichols’ study, they explain, “was to determine the influence of neurodegenerative symptomatology on third-person judgments of personal identity” (p. 1470). But a theory of personal identity that locates that identity outside a person’s own sense of self is deeply unsatisfying.

Here’s the puzzle. It makes sense to suppose that character is constitutive of who we are. On closer examination, however, it seems neither necessary nor sufficient for personal identity. Or more correctly, whether it’s either necessary or sufficient would appear to depend on one’s perspective. So long as people retain a critical mass of the memories that make up their personal narratives, their characters can change and even change dramatically, without threatening their sense of self. That is, character can change without any corresponding change of personal identity from the first-person perspective. If people’s characters change enough, however, they will seem like different people to others, which is to say they will seem like different people, from the third-person perspective.

Conversely, people can experience such profound memory loss that they no longer have a coherent personal narrative and hence completely lose their sense of self. Memory loss does not necessarily affect character, however. So these same people who are no longer the person they were from the first-person perspective, may remain recognizable as the same person to others, which is to say, from the third-person perspective.

So it appears the judgment of whether a given person is “the same person” after significant memory loss as they were before depends on the perspective from which one makes the judgment. Someone whose personality remains recognizably the same will appear to be the same person from the third-person perspective, which is to say from the perspective of people who knew them before, even if they have entirely lost their own sense of who they used to be. On the other hand, a person can retain their subjective sense of self even after personality changes that make them effectively unrecognizable to others.

So it seems that just like electrons will appear to be either waves or particles depending on how one observes them, so can we appear to be either the same person we have always felt ourselves to be, or an entirely different person depending on who’s doing the observation, ourselves or someone else.

But which perspective are we to take as definitive? The first person or the third person? Philosophers, as Thomas Nagel observed so compellingly in The View From Nowhere (Oxford, 1989), are biased in favor of objectivity, which is to say, they are biased in favor of the third-person perspective.

Do other people really have the final say, though, on who we are?

Getting Gettier Wrong

Edmund Gettier showed in “Is justified true belief knowledge?” that it’s possible to have what most people would consider a justified belief that is true by accident, or a belief where the “justification” is not related to the truth in the way we intuitively feel it ought to be in order for the belief to amount to knowledge. This insight gave rise to a variety of attempts to develop theories of belief justification that would avoid what came to be known as “the Gettier problem.” Alvin Plantinga argues he has developed such a theory, though he refers to his theory as one of “warrant” rather than “justification.“ Plantinga argues that Gettier problems relate to anomalies in the cognitive environments in which beliefs are formed and that his theory of warrant avoids such problems by taking the cognitive environment into account in a way that standard theories of justification do not. It’s clear, however, that Plantinga’s ”warrant” does not avoid the Gettier problem. I argue that Platinga’s purported solution to this problem is based on a misunderstanding of the problem. The problem does not, in fact, concern the cognitive environment in which beliefs are formed in the way Plantinga argues it does.

Plantinga argues that knowledge is possible only if one assumes some kind of pre-established fit between the way our minds work and the nature of objective reality. This fit is part of what Plantinga calls God’s “design plan” (2011, piii), a plan that involves “a match between our cognitive powers and the world” (2011, xiv). A belief has warrant, according to Platingua, “if it is produced by cognitive faculties functioning properly (subject to no malfunctioning) in a cognitive environment congenial for those faculties, according to a design plan successfully aimed at truth” (1993, piii).

“Gettier problems,” observes Plantinga,

come in several forms. There is Gettier’s original Smith owns a Ford or Brown is in Barcelona version: Smith comes into your office, bragging about his new Ford, shows you the bill of sale and title, takes you for a ride in it, and in general supplies you with a great deal of evidence for the proposition that he owns a Ford. Naturally enough you believe the proposition Smith owns a Ford. Acting on the maxim that it never hurts to believe and extra truth or two, you infer from that proposition its disjunction with Brown is in Barcelona (Brown is an acquaintance of your about whose whereabouts you have no information). As luck would have it, Smith is lying (he does not own a Ford) but Brown, by happy coincidence, is indeed in Barcelona. So your belief Smith owns a Ford or Brown is in Barcelona is indeed both true and justified; but surely you can’t properly be said to know it. (1993, 32).

Plantinga examines several similar examples where a person forms a true belief based on what is, in fact, some kind of deceptive practice. These examples include one, which he credits to Carl Ginet, that involves a belief that one is looking at a comparatively fine example of a barn where the comparison is based not on real barns but on barn facades that the local inhabitants of rural Wisconsin town have erected in an effort to make themselves look prosperous. Another involves a belief that horses have recently been on a particular bridle trail in Yellowstone National Park based on the sight of horse manure placed there by “a wag with a perverse sense of humor” (1993, 32). Each example functions effectively like the first in that the crucial element in the example is that it involves an attempt to deceive.

The problem, according to Plantinga is that “credulity is part of our design plan. But it does not work well when our fellows lie to us or deceive us in some other manner, as in the case of Smith, who lies about the Ford, or the Wisconsinites, who set out to deceive the city-slicker tourists” (1993, 33). That is, there’s a problem, according to Plantinga, with “the cognitive environment.” That environment is supposed to be free of deliberately deceptive elements.

If we turn to what Gettier actually says, however, we see that neither of the counterexamples he presents to the definition of knowledge as justified true belief involves any reference to deliberate deception. Plantinga describes Gettier’s “original Smith owns a Ford or Brown is in Barcelona version” as one in which Smith lies about owning a Ford.

But Smith doesn’t lie in Gettier’s example. No one lies in that example. In fact, it is not Smith, but Jones who owns a Ford in Gettier’s original example. Smith, in this, the second of Gettier’s two examples, has reason to believe Jones owns a Ford. Why does Smith believe this? “Smith’s evidence,” writes Gettier, “might be that Jones has at all times in the past within Smith’s memory owned a car, and always a Ford, and that Jones has just offered Smith a ride while driving a Ford” (1963, 122). There’s no reference to Smith lying in Gettier’s original example despite Plantinga’s claim that there is. It just happens that Jones no longer owns a Ford, but has been driving a rented Ford.

Each of the purportedly Gettier-type counterexamples to knowledge as justified true belief that Plantinga presents, involves either lying or some other form of deliberate deception.
“[A] true belief is formed in these cases,” writes Plantinga, “but not as a result of the proper function of all the cognitive modules governed by the relevant parts of the design plan. The faculties involved are functioning properly, but there is still no warrant; and the reason has to do with the local cognitive environment in which the belief is formed” (1993, 33). That is, the local cognitive environment involves deliberate deception.

Plantinga acknowledges that “credulity is modified by experience,” that “we learn to believe some people under some circumstances and disbelieve others under others” (1993, 33), but he fails to specify what sorts of circumstances provide the proper cognitive environment for credulous belief formation. In fact, as Richard Greene and N.A. Balmert point out, Plantinga fails, in general, “to provide us with a criterion by which we can identify the proper cognitive environment for a particular cognitive module” (1997, 137).

It isn’t possible avoid the Gettier problem the way Plantinga argues his theory of warrant does by pointing out that God designed us generally to believe what other people say or do because no one lies, or indeed engages in any other form of deception, in either of Gettier’s examples. The “local cognitive environment” in Gettier’s “Jones owns a Ford or Brown is in Barcelona” example is free of any element of deception. The problem Gettier points out with the traditional definition of knowledge as justified true belief does not appear to have anything to do with the “local cognitive environment,” unless contingency is considered inappropriate to that environment, since it just happens that Jones no longer owns a Ford and that Brown is in Barcelona. But contingency is a feature of almost every cognitive environment, and certainly to every cognitive environment where the beliefs formed concern the nature of objective physical reality.

References

Gettier, E. 1963. Is justified true belief knowledge? Analysis 23, 121-23.
Greene, R. and Balmert, N.A. “Two notions of warrant and Plantinga’s solution to the Gettier problem. Analysis 57, 132-139.
Plantinga, A. 1993. Warrant and Proper Function. Oxford University Press.
Plantinga, A. 2011. Where the Conflict Really Lies. Oxford University Press.

Sight and light

We normally regard seeing as intimately connected with light. But must seeing involve light? Suppose you could step into a pitch-dark room and have precisely the experiences you would have if it were fully lighted. The room would thus look to you just as it would if fully lighted, and you could find any unobscured object by looking around for it. Would this not show that you can see in the dark? If so, then the presence of light is not essential to seeing.

However, the case does not establish quite this much. For seeing is a causal relation, and for all I have said you are just vividly hallucinating precisely the right things rather than seeing them. But suppose you are not hallucinating and that if someone covered a coin you see with lead or covered your eyes, you would no longer have a visual experience of a coin. In this case, it could be that somehow the coin affects your eyes through a mechanism other than light transmission, yet requiring an unobstructed path between the object seen and your eyes. Now it begins to seem that you are seeing. You are responding visually to stimuli that causally affect your eyes. Yet their doing so does not depend on the presence of light

—Robert Audi

(This post was excerpted, with Audi’s permission, from his Epistemology: A Contemporary Introduction to the Theory of Knowledge. It’s an excellent example of how it is possible to make an interesting and even important philosophical point in very few words.)