Peter Unger maintains that all knowledge requires certainty. Moreover, since he insists that nothing can be known for certain, Unger concludes that “nobody ever knows anything to be so” (Unger, 42). This is Unger’s argument:
1. “If someone knows something to be so, then it is all right for the person to be absolutely certain that it is so” (42).
2. “It is never all right for anyone to be absolutely certain that anything is so” (43).
3. Therefore, “[n]obody ever knows that anything is so” (43).
Succinctly, nobody can know anything. As “anything” makes explicit, Unger suggests that we cannot have knowledge of our own existence, external objects, past or present experiences or even that 1+1=2.
He also insists that knowing anything with certainty is inherently dogmatic. Being certain involves a negative attitude; it implies that nothing (new information, evidence or experience) “will be seriously considered to be at all relevant to any possible change in one’s thinking in the matter” (44). Unger defines this as the attitude of certainty. This is why it is wrong “for anyone to be absolutely certain”. I agree with Unger and concede that (2) is correct.
However, Unger’s rejection to the attitude of certainty leads to the rejection of all knowledge. This is where Unger is an error. I intend to argue that premise (1) is dubious and that knowledge requires justified true belief but never certainty itself.
Before continuing, I must clarify Unger’s notion of certainty. Hitherto, I have used Unger’s notion of certainty (denoted in italics) without explanation. Certain is an absolute adjective analogous to the concept of flat. An absolute adjective is or is not. A board, for example, is flat iff it lacks any changes in gradient. Similarly, to be certain of something we must be devoid of all doubt. Any doubt implies some un-certainty. According to Unger, knowing entails absolute certainty. But, since nothing can be known without some doubt, he concludes that nothing can be known.
I cannot accept Unger’s conclusion. Conceding that (2) is correct, I will focus on (1). Unger aims to defend (1) within the second section of his paper. Hoping to support (1), Unger provides an example where a person, who is supposedly certain of something, is asked how they can be certain of it. By asking the question, Unger believes we imply that it is not all right for that person to be certain and further imply that they cannot be certain because they do not really know the thing. However, these are only implied because Unger assumes that everyone accepts his notion that if someone knows something then it is all right be certain of it – but if they do not then it is not all right . After the example, Unger ends his direct argumentation for (1) and concludes that the example demonstrates an analytic relationship between knowledge and certainty. However, Unger fails to consider alternatives to what is implied by the question in his example. Contrary...