But the more important distinction for us is between direct and indirect knowledge. Direct knowledge is knowledge that is known immediately by us and on our own authority. Indirect knowledge is knowledge that we are related to only through someone else; it is mediated by those others and therefore always involves the issue of authority, for it is on the basis of authority that we determine whom to listen to or not.
Examples of direct knowledge include things like the fact that you can't be in two places at the same time, that you are younger than your parents, and that dogs are produced by nature but automobiles are only products of human artifice. Some (but not all) of mathematics is direct knowledge. You don't need an authority to tell you that 2+2=4. And if you can follow Euclid's proof that there are an infinite number of primes. then the fact that there are an infinite number of primes is direct knowledge for you.
Suppose you can't follow the proof. Then you can still be related to that fact as knowledge, but only indirectly through the authority of someone else who can follow the proof. A consequence of this is that the same piece of knowledge can be known directly by some and indirectly by others. Everyone knows 2+2=4 on his own authority; but very few people know that Fermat's Last Theorem is true on his own authority, for its proof is so sophisticated that only the most educated mathematicians can follow it.
It can be seen that indirect knowledge depends on direct knowledge. If I'm taking something on the authority of another, it is not unreasonable for him to be taking it on the authority of another as well, but somewhere the chain has to end with someone who simply knows it directly. Otherwise we have a train with nothing but freight cars and no engine. (An example is a child who believes in the Big Bang on the authority of his teacher, who in turn believes it on the authority of cosmologists. But the cosmologists know it directly because they have gone through and understand the scientific case for the Big Bang.)
What about science? Jerry Coyne tells us on page 187 of Faith vs Fact that "I see science as a method not a profession... Any discipline that studies the universe using the methods of 'broad' science is capable in principle of finding truth and producing knowledge. If it doesn't, no knowledge is possible." So to have "science" in the strict sense we must produce it through the method that defines science. Unfortunately, very few of us - actually no one - has the time or resources to develop his entire base of knowledge through the application of scientific methods. We must, to a great degree, rely on the application of the scientific method that others have performed and take their results as a given; or, rather, we can only be related to their scientific knowledge indirectly through appeal to their authority as scientists.
The irony of the advance of science is that the more it advances, the less it becomes directly available to any individual man. Back in the early days of modern science, an intelligent amateur could keep abreast of, and perhaps reproduce, most of the crucial scientific results. It's not hard to reproduce Galileo's experiments with rolling balls and, if he can get his hands on a telescope, verify the existence and movements of Jupiter's satellites for himself. And he can easily reproduce Franklin's experiments with electricity or Pascal's with atmospheric pressure. But as science advances, it requires increasingly expensive and elaborate apparatus to construct experiments; and those experiments themselves require a much larger base of knowledge to understand. A high school student can be brought to an understanding of Galileo's experiments in acceleration in the course of one day's class. He'll need another four or so years of intensive education, at least, to understand how and why recent experiments have demonstrated the existence of the Higgs boson, assuming he is capable of mastering the relevant material at all. And that student, while mastering physics, will not be spending his time mastering biology and genetic science, so that, however much he might end up directly related to knowledge in physics, he will still be indirectly related to all that genetic science produces, and all that the other sciences produce. So the more science advances, the more all of us are indirectly related to scientific knowledge, including scientists themselves.
It thus becomes crucial for us to understand the distinction between direct and indirect knowledge, how they are related, and how to handle each type of knowledge appropriately. I've already discussed the distinction between the two types of knowledge. How are they related? As pointed out above, indirect knowledge is dependent on direct knowledge, since indirect knowledge is really just direct knowledge removed some number of times from the original source.
But indirect knowledge is dependent on direct knowledge in another way, and that is subjectively. By that I mean the only means we have available to evaluate indirect knowledge is through direct knowledge. When a scientist says that the Big Bang is true, how do I know whether to believe him or not? I could appeal to some other instance of indirect knowledge, for instance that other scientists agree with him, but this only pushes the problem back a step, since I now have to think about how to evaluate that piece of indirect knowledge. Again, at some point I must have recourse to something I simply know directly, through which I can evaluate competing claims of indirect knowledge.
The process of analyzing and appropriating direct knowledge is philosophy. The crucial distinction with direct knowledge is that it is not mediated; that is, it must ultimately be known without reliance on anyone else. Kierkegaard discusses this in his analysis of Socrates in Philosophical Fragments. A true teacher - that is, in my terms, a teacher of direct knowledge - is only the occasion by which someone comes to know, and the process has only completed when the teacher has become dispensable. It is for this reason that philosophy does not "progress" or produce "results" - one of the perennial charges against it. A "result" is knowledge that can be appropriated without reproducing the process by which it came to be known - for example, when an engineer uses the facts about electronic devices to design a system without first proving all those facts scientifically for himself. "Results" are therefore by nature indirect knowledge. Philosophy cannot produce "results" without falsifying itself; and everyone who would make progress in philosophy must reproduce for himself the process by which philosophers have come to know - and in the process, make those philosophers dispensable. There are no "results" that can be handed on from Plato's Republic. But someone who reads it may come to know things for himself that he might otherwise not know.
The fact that the teacher becomes dispensable is one characteristic of philosophy; another is that it appeals to direct experience as its evidential basis, on the eminently reasonable principle that it is the only possible basis. For my own, immediate experience is the only direct contact I have with reality (if in fact I have contact with reality at all); anything else is mediated and therefore a subject of indirect knowledge. This too, like the fact that philosophy doesn't produce "results", sometimes puts people off philosophy, for it makes philosophy seem a matter of purely "subjective" preference. And it is subjective, in the sense that it is only I that have access to my own experience. This is true, necessary and unavoidable, nearly tautological, yet is frequently overlooked. From p. 195 of Faith vs Fact:
"I'm hungry," my friend tells me, and that too is seen as extrascientific knowledge. And indeed, any feeling that you have, any notion or revelation, can be seen as subjective truth or knowledge. What that means is that it's true that you feel that way. What that doesn't mean is that the epistemic content of your feeling is true. That requires independent verification by others. Often someone claiming hunger actually eats very little, giving rise to the bromide "Your eyes are bigger than your stomach."
The fact you feel hungry is a fact concerning reality as much as any other. Whether you really need to eat or not is irrelevant to the truth that you in fact have the feeling . Ultimately, science itself depends on subjective knowledge, because scientists must read meters and look through microscopes - "I am seeing an amoeba through this lens" or "The voltmeter says 5 volts." There is really no way to escape the subjective nature of these experiences. For instance, trying to "independently verify" them as Coyne suggests, for instance, by asking someone else whether they see 5 volts as well, may be a reasonable procedure, but it only works because we take our subjective experience of what someone else tells us - "I am hearing Joe say the voltmeter reads 5 volts" - as itself not in need of independent verification. Otherwise, we are back to the familiar infinite regress that comes up so often in this context.
The philosopher faces the fact that all our knowledge - direct, indirect or otherwise - can ultimately be evaluated only in light of our own personal experience. The philosopher serves as an ultimately dispensable aid in analyzing and discovering the significance and meaning of that experience. The scientist simply takes the meaning of personal experience for granted so he can get on with his science. And he is perfectly justified in this, but he is in danger, like Coyne, of misunderstanding the real relationship between science and philosophy - which is really a misunderstanding of the basic human condition.
Coming next: How direct knowledge is used to evaluate indirect knowledge. Hint: Read Plato's Apology.