Why There is No Problem of Knowledge
© Peter Zohrab 2013
Philosophers have long claimed that there is a "Problem of Knowledge" -- for example, see: Alan Musgrave, "The Problem of Knowledge" (ch. 1 of Musgrave 1993). Actually, there is no real problem of knowledge, and it doesn't take long to show that. After all, Musgrave's chapter is only 18 pages long, and Gettier's famous challenge to the orthodox solution to this problem (Gettier 1963) was published in a version that was only three pages long.
For a long time, the standard philosophical view was that Knowledge could be characterised as follows:
(1) I know that P iff (just in case):
(i) I believe that P,
(ii) P is true, and
(iii) I am justified in believing that P.
Musgrave (1993) gives the example: "Suppose I say I know that there is someone standing outside the door." He then goes on to ask what must be the case for this claim of mine to be correct. First of all, I must believe that there is someone standing outside the door. The second condition is that there really is someone standing outside the door. And the third condition is that "I must also be able to give reasons for my belief, or justify it, or show that it is true, or prove it."
Musgrave states, "If I say 'I know that P' and then find out that P is false, I will withdraw my claim to knowledge: I will say that I thought I knew that P but did not really know it." However, consider the following three scenarios, involving three different times (T1, T2 and T3, T1 being the earliest and T3 being the latest):
At T2, I would want to say "I know that there is no one standing outside the door," and that is why I could not simultaneously agree with my previous statement, "I know that there is someone standing outside the door". I would be making contradictory statements. However, at T1 I was using the word "know" perfectly correctly. At T2 I was also using the word "know" correctly -- all that had changed was my belief about the situation outside the door. At T3 (the present time) I am still using the word "know" correctly, but my belief has changed back to what it was at T1.
At T3, I have no idea if my belief about the situation outside the door will or will not change again at some time in the future. So is our knowledge perhaps always tentative? At any rate, it seems clear that what is involved in the second condition is not objective truth but, rather, my beliefs, just as in Musgrave's first condition for knowledge. So we could rewrite the standard conditions for knowledge as follows:
(2) I know that P iff (just in case):
(i) I believe that P,
(ii) I believe that P, and
(iii) I am justified in believing that P.
Of course, the second condition is now identical to the first one, so we could reduce the conditions to two, as follows:
(3) I know that P iff (just in case):
(i) I believe that P, and
(ii) I am justified in believing that P.
Now consider the strangeness of the following sentence:
(4) I believe that there is someone outside the door, but I don't know it.
It would be possible to imagine scenarios where this would be a possible sentence, especially with heavy emphasis on the words "believe" and "know", but, on the whole, the sentence appears self-contradictory. That throws the spotlight onto condition (ii) in version (3) (above). Who carries out the evaluation as to whether I am justified in believing that P? Surely I do myself -- I do not need to consult any outside authority. Moreover, I wouldn't even believe that P unless I thought was justified in believing that P, either. So condition (ii) is unnecessary, and we should rewrite the conditions for knowledge as follows:
(5) I know that P iff (just in case)I believe that P.
However, there is a further issue, to do with the second (the "you" forms) and third (the "he"/"she"/"it/"they" forms) persons of the verb. Philosophers such as Musgrave use the first person form of the verb ("I know"), the third person form ("S knows") and the cognate noun, ("knowledge") interchangeably, but actually these three forms are non-identical in relevant ways. Suppose, for example, we consider the sentence: "He knows that there is someone standing outside the door." What must be the case for this sentence to be correct? In this case, the three conditions cannot be reduced down to two or one. First of all, he must believe that there is someone standing outside the door. So far, so good.
The second condition is not that there really is someone standing outside the door, as we have mentioned above. The second condition is that I believe that there is someone standing outside the door, and the third condition is that I believe that he is justified in believing that P. So we can formulate the conditions for third-person knowledge as follows:
(6) He knows that P iff (just in case):
(i) He believes that P,
(ii) I believe that P, and
(iii) I believe that he is justified in believing that P.
Similarly, we can formulate the conditions for second-person knowledge as follows (although we are unlikely to state them so baldly and crassly to another person's face):
(7) You know that P iff (just in case):
(i) You believe that P,
(ii) I believe that P, and
(iii) I believe that you are justified in believing that P.
Just as a speaker may pass judgement on whether he actually knew what he thought he knew in the past, so the speaker/writer/thinker ("I") passes judgement on what other people think they know in the present, or thought they knew in the past (or even that they might think they know in the future). So the difference between belief and knowledge is that knowledge exists when the speaker/writer/thinker agrees at the relevant time with the belief concerned (although rare scenarios can be imagined where there is a more intrinsic difference between the meanings of "believe" and "know").
So there is not really any Problem of Knowledge. Using the abstract noun "knowledge" in this context misleads by reifying the issue, which is really just the issue of the correct linguistic usage of the verb "know". What people are said to "know", as opposed to merely "believe", depends on whether the speaker/writer/thinker agrees with them at the time in question.
References
Gettier, Edmund L. (1963 ) Is Justified True Belief Knowledge? Analysis 23 ( 1963): 121-123.
Musgrave, Allan (1993): Common Sense, Science and Skepticism: A Historical Introduction to the Theory of Knowledge, Cambridge, NY: Cambridge University Press, 1993.
Last Update: 10 April 2013