Russell On Induction

Induction, as Russell describes on page 66, is "when a thing of a certain sort A has been found to be associated with a thing of another certain sort B, and has never been found disassociated from a thing of sort B, the greater the number of cases in which A and B have been associated, the greater is the possibility that they will be associated in a fresh case in which one of them is known to be present." Also, Russell states that the more this association presents itself, the more likely it is to become nearer to a certainty. Basically, what Russell is saying is that when there are two different events that have only been witnessed as a tandem, that it is highly unlikely for these two events to be witnesses singularly. For example, when lightning strikes one would expect to then hear a clap of thunder following the bolt of lightning. But, the principle of induction applies only to the fresh case mentioned before. Trying to discover the probability of the thing of sort A and the thing of sort B always being associated would be considered trying to find a "general law." Finding the probability of this general law would be accomplished using the same associations but the probability of the two events occurring together would be significantly lower than that of the fresh case alone, even though more repetitions of the association would increase the probability in favor of the general law.

Russell states that all of our assumptions about the future are based on events in the past. He says that people expect the sun to rise everyday because that is how it has always been. People expect the laws of motion to remain in operation until tomorrow because that is how it has been up until that point in time. Russell states on page 61 that doubting the laws of motion will remain in operation until the following day is an interesting doubt because there are no known forces that can act on the Earth that will cause the law to cease functioning. He states that if a large object were to come and knock the Earth out of rotation then the sun would no longer rise the next morning, but the laws of motion would remain unaffected from this event. That is what makes this doubt so interesting. As far as human knowledge is concerned there is no known event that could cause an exception to the laws of motion, thus changing the way in which the law functioned. Russell then asks the question "Have we any reason, assuming that they have always held in the past, to suppose that they will not hold in the future?" What he means is that there have never been any events prior to now that have changed these general laws, what reason would we have to doubt that they will not perform the same tomorrow and the days following that. He believes that even though we may believe we have found a law that has no exceptions, it can never be proven to be without exception because all of the data collected is from previous experiences which still may have the slightest possibility of changing.

Further on in Russell's explanation of induction he states that the inductive principle is "not capable of being disproved by an appeal to experience." For example, an argument could be made stating that it is probable for all squirrels to be brown. Just because