Wednesday, March 14, 2007

What science cannot discover, mankind cannot know

Bertrand Russell once wrote "What science cannot discover, mankind cannot know." One has to wonder, however, how this can possibly be the case. After all, Russell didn't say "What science cannot discover, except for this sentence, mankind cannot know." And you would think Russell would have known a thing or two about self-referential paradoxes. Has science discovered that we cannot know anythiung over and above what science can discover? What discoveries support that conclusion? I keep saying over and over again until I'm blue in the face that something cannot be a discovery of science unless science could have discovered that the opposite is true. So what that means is that science could have discovered that science is not the only source of knowledge, but instead discovered that it was. When did that happen?

Here's the Russell passage:

I conclude that, while it is true that science cannot decide questions of values, that is because they cannot be intellectually decided at all, and lie outside the realm of truth and falsehood. Whatever knowledge is attainable, must be attained by scientific methods; and what science cannot discover, mankind cannot know.

This post on Thinking Christian also deals with the difference between science and scientistic philosophy.

How do you like the picture of philosophy's most notorious dirty old man?


Anonymous said...

Do you suppose Russell ever said, "Sometimes a pipe is just a pipe!"?

Good point on his rather dubious statement there. I'm surprised he would say something like that. For all I've disagreed with him, he was extremely bright; it seems he would know enough not to make such a basic error.

Anyway, thanks for the link, Victor.

Anonymous said...

This article is quite intriguing. I have never thought about science in this point of view, but I can clearly see where the writing is coming from. This is a true philosophy that requires much thought. There is no evidence to prove that if science doesn’t prove it, it doesn’t exist, yet on the contrary we believe there are unexplainable things throughout the world. This one really gets you thinking. Russell did have a great point when coming up with his philosophy. Yet I do agree. Now that we believe there is more than the eye can see, we never take anything and just look at it for what it is. There has always got to be something deeper, or something different than just it being plain and simple. This is a great quote on science, probably one to remember for a lifetime!

amandalaine said...

Interesting post.

I don't know much about Russell, but, why did you call him "philosophy's most notorious dirty old man"? Due to his philosophy or behaviors? Just wondering.

Thanks! I enjoy your blog a lot.

Victor Reppert said...

Russell not only advocated free sex, unlike Ted Haggard, he practiced what he preached. No hypocrisy for him! I guess it would be disappointing if Russell had said what he said in "Marriage and Morals" and lived like a monk, but in fact he was committing adultery on a regular basis into his eighties.

Anonymous said...

Interesting that you equate having sex outside of marriage with being dirty.

In any case, his statement seems to be referring to the scientific method of testing hypotheses and rejecting conclusions that don't fit the data. Humans have been doing that long before they called it the scientific method.

Anonymous said...

Russell was clearly wrong, as he cannot rule out knowledge gained by ESP

Anonymous said...

to "es"
Victor says "he was committing adultery on a regular basis into his eighties." Adultery is "voluntary sexual intercourse between a married person and someone other than his or her lawful spouse."

If another man/woman regularly has sex with your wife/husband, I think you might think, "that is dirty."

Victor Reppert said...

So, is the statement "What science cannot discover, mankind cannot know" testable? If so how?

Anonymous said...


As a general rule, if you suspect that a heavyweight like Bertrand Russell has made an obvious and elementary error, you should first seriously consider the possibility that you have misunderstood him. And in this case I think you have.

If Russell had said "What logic cannot discover, mankind cannot know", you might have caught him out in a self-referential paradox. But scientific "discoveries" are not analytic propositions, they are inductive generalizations founded on a series of observations. And Russell's statement would be consistent (although still possibly wrong) if he believed that all the relevant observations did support his claim.

A more formal justification of Russell's statement might look like this:

*Let S be the set of all possible synthetic propositions.

*Let D be the subset of S containing all synthetic propositions that we know have been discussed by philosophers, theologians or scientists at some point in the last 2500 years.

*Write SMx to mean "the proposition x can be studied using the scientific method".

*Write Cx to mean "either there is a consensus that the proposition x is true, or there is a consensus that the proposition x is false".

Then Russell's argument (in my opinion) reads:

(1) For all x in D, if ~SMx then ~Cx. (Observation)
(2) There are some x in D with SMx and Cx. (Observation)
(3) For all x in S, if ~SMx then ~Cx. (Abductive hypothesis from (1) and (2))

A paradox is avoided if steps (1) to (3) represent a legitimate instantiation of the scientific method. Russell presumably believed that they do.

And in answer to your question whether Russell's statement is "testable", I should point out that the conclusion (3) is easily falsified: just find a proposition x in S with ~SMx and Cx.

Victor Reppert said...

Some people who are otherwise heavyweights are not good at avoiding elementary mistakes. Russell attributed to Kant the view that one should never borrow money, because if everyone borrowed money, there would be no money to borrow. He also attributed to Aquinas the view that there had to be a first cause because everything has a cause, and then of course that pointed out that the first cause would then have to have a cause, failing of course to consider that Aquinas constructed his argument so as not to suffer from that kind of difficulty.

Some strong chess players never drop pieces, but others sometimes do.

Anonymous said...

Victor, think more on Anonymous No. 1's point. Russell has, in fact, made a testable assertion. Ascribe any reasonable scientific definition to "know," and then try to list the things mankind "knows" independent of what it has discovered other than through science. You end up with a list of bits of "knowledge," with the truth of each bit going in and out of fashion over time, inconstant and seen differently through various cultural and intellectual prisms. The stuff we know independent of science really doesn't look much like knowledge. Quite different from the apparently inexorable progress of science, where facts once ascertained are equally true for everyone and in every context, one discovery builds upon another. What passed for "knowledge" before science gradually retreats further and further toward the margins. Science never seems to retreat; it just gradually expands its utility and ability to explain the way things are, and leaves us less of the unknown to fear. This leaves less scope for religion, which has become much tamer and less threatening to free enquiry since science has been on the scene. -- Anonymous No. 2

The Brazilian Librarian said...

A good point to make is that if Russel might be saying "what science cannot discover, mankind cannot know" as a statement of rigid logic. By being so, it cannot be true if it has paradoxes. Furthermore, scientific theories which were created after an exaustive use of the scientific method cannot be concidered proofs in the mathematical sense. and as the statement is of rigid logic, this again imples teh claim to be logically wrong. However, if it is not a statement of rigid logic, if 'know' means scientific knowledge, then you should reconcider your conclusions. However, if know means scientific knowledge, Russel's conclusion is quite obviuous (if it is not of rigid logic), but perhaps not important. I agree with you, just pointing it out.