Indirect knowledge is knowledge that we are unable to evaluate in the terms by which it is directly known. For example, it is only the cosmologist who has the time, resources and education to draw scientific conclusions about the physical history of the universe on a cosmic scale. The rest of us, to the extent that we can be related to that knowledge at all, are only related to it through the cosmologist and to the extent that we believe what he tells us about cosmology. The key characteristic of indirect knowledge is, therefore, that it is mediated by another.
Naturally we want to only believe things that are true and avoid believing things that are false. In the case of indirect knowledge, then, this must involve an evaluation of the mediator through which we are related to the knowledge. In direct knowledge, we evaluate the evidential and logical basis for the knowledge ourselves; in indirect knowledge, we evaluate the reliability of the mediator who is, presumably, himself directly related to the evidential and logical basis for the knowledge. (It should be remembered that there might be a chain of mediators; the significant point is that the chain must eventually end in someone directly related to the knowledge. For the purposes of this post, however, there is no significant difference between a chain of one or many so the chain will taken to have only one link for the sake of clarity.)
It would be defeating the purpose, of course, if we tried to evaluate the mediator in terms of a direct relationship to the knowledge itself. For instance, there is no point in me trying to evaluate whether a cosmologist is reliable by reviewing his work in light of an application of cosmological science itself. For if I could do that, I could relate myself directly to cosmological knowledge and wouldn't need the cosmologist in the first place. We only avail ourselves of indirect knowledge when direct knowledge is unavailable to us.
While we can't evaluate a mediator directly in terms of the science he mediates, we can evaluate him in terms of his general human nature as a knower. For the canonical example of how to do this, we turn to Socrates in Plato's Apology. It will be recalled that Socrates was told by the oracle at Delphi that he was the wisest of men. Incredulous at this, Socrates attempted to prove the oracle wrong by finding a man wiser than himself, which he thought would not be difficult to do. Among the individuals he interviewed in this quest were the skilled craftsmen. This is the result:
Last of all, I turned to the skilled craftsmen. I knew quite well that I had practically no technical qualifications myself, and I was sure that I should find them full of impressive knowledge. In this I was not disappointed. They understood things which I did not, and to that extent they were wiser than I was. But, gentlemen, these professional experts seemed to share the same failing which I had noticed in the poets. I mean that on the strength of their technical proficiency they claimed a perfect understanding of every other subject, however important, and I felt that this error more than outweighed their positive wisdom. So I made myself spokesmen for the oracle, and asked myself whether I would rather be as I was - neither wise with their wisdom nor stupid with their stupidity - or possess both qualities as they did. I replied through myself to the oracle that it was best for me to be as I was. (from the Apology, in the Collected Dialogues of Plato edited by Hamilton and Cairns)What Socrates has noticed is that being in expert in one thing does not make one an expert in everything, which is of course common sense. But he has noticed something else which is more significant, and even paradoxical, which is that being an expert in a field has a tendency to make people think they have a competence in other areas that is undeserved. I say it is paradoxical because one would think that through the process of becoming an expert in one field a man would realize how difficult it is to become an expert in any field, and so would tend to a natural humility concerning knowledge outside his own specialized field. Yet the opposite seems to happen; becoming an expert in one field tends to make one think he is an expert everywhere.
This observation is even more relevant today than it was in Socrates's time. For as I pointed out in the original post, science becomes more specialized the further it advances. That is, to become a scientific expert today means spending an increasing amount of time on an increasingly narrow domain. Scientists are subject to opportunity cost as much as anyone else; a scientist can only become an expert today on early universe cosmology by spending his time studying that and not other things - for instance genetics, chemistry, electrical engineering or botany, not to mention law, economics, history or philosophy. But, just as in ancient Greece, the expert of today will pretend to a competence outside his narrow area of expertise.
How can we use this principle in our evaluation of indirect knowledge? We should not take for granted that an expert is able to distinguish that which he knows through his expertise and that which holds merely by his opinion or ordinary reason. In other words, he may have no clear self-understanding of what he knows and what he doesn't know and why. Thus what we get from him may be a mix of his expert opinion on the subject on which he is competent - what we want - and his opinions on other subjects on which he has no particular competence better than our own - what we don't want. It is up to us to sort out the one from the other. But beyond that, we should be more likely to rely on the expert testimony of an expert who has the self-awareness to distinguish his expert opinion from his merely ordinary opinion. Such self-awareness indicates that the expert is aware of what it means to know, and we can have more confidence that what he is giving us is in fact only that which is justified by his expert opinion.
The classic example of an expert who is the modern equivalent of the craftsmen Socrates encountered in Athens is Carl Sagan. Sagan, an expert in planetary science, wrote a number of popular books on science (e.g. Cosmos) that explored well beyond his particular competence in astronomy. One of his most popular books, The Demon Haunted World: Science as a Candle in the Dark attempts to trace science as a singular beacon of knowledge in a world haunted by superstition, religion and pseudoscience. The book is of interest here because of a passage on p. 256-257 where Sagan issues some "mea culpas" on instances where he went wrong. The instances include the following: Estimating the atmospheric pressure of Venus incorrectly; incorrectly estimating the water content of Venutian clouds; thinking there might be plate tectonics on Mars when in fact there weren't; attributing the wrong cause to the high temperatures on Titan; overestimating the effect of burning Persian Gulf oil wells on the agriculture in South Asia.
What do these instances all have in common? They are all cases of Sagan admitting error in his particular area of expertise - planetary science. Yet Sagan offered opinions on subjects far beyond planetary science; in the The Demon Haunted World itself he makes assertions about history, religion, philosophy, politics and economics among others. He gives no instances when he was wrong about politics or philosophy. Is this because, bizarrely, he's always right in areas where he's not an expert and only wrong in areas where he is an expert? More likely is that Sagan, in his area of expertise, knows when he is right and when he is wrong, but in areas outside of his expertise, he doesn't really know when he is right and when he is wrong.
And it is not hard to find instances when he is wrong in The Demon Haunted World. He claims on p. 155 that Plato "assigned a high role to demons" and quotes the following in evidence:
We do not appoint oxen to be the lords of oxen, or goats of goats, but we ourselves are a superior race and rule over them. In like manner God, in his love of mankind, placed over us the demons, who are a superior race, and they with great ease and pleasure to themselves, and no less to us, taking care of us and giving us peace and reverence and order and justice never failing, make the tribes of men happy and united.Sagan gives no attribution for this quote, but a little research shows that it is from Book IV of Plato's Laws. The context of the quote makes clear that Plato is not speaking in his own voice, but is recounting the received tradition concerning how mankind was originally ruled in the ancient, golden age of Cronus. And the continuation of the passage shows that it means pretty much the opposite of what Sagan thinks it means:
So the story teaches us today, and teaches us truly, that when a community is ruled not by God but by man, its members have no refuge from evil and misery. We should do our utmost - this is the moral - to reproduce the life of the age of Cronus, and therefore should order our private households and our public societies alike in obedience to the immortal element within us, giving the name of law to the appointment of understanding.(My translation in Hamilton and Cairns is slightly different than Sagan's, wherever he got it from.) So while the age of Cronus may have been ruled by benevolent demons, ours is not, but we can imitate that golden age by ruling ourselves through the immortal element within us - which for Plato is the soul and in particular the intellectual element of the soul - the "appointment of understanding." Plato, far from giving demons a "high role", is giving them no role at all and instead is urging us to order our affairs through reason. More deeply, Plato is wisely using the tradition of mythology to support the rule of reason; rather doing a Sagan-like move and dismissing any regard for mythology as foolish, Plato acknowledges the wisdom in mythology but turns that respect for tradition to his own purposes. In the present age, Plato argues, respect for tradition cannot take the form it once did - since the present age is manifestly not a golden age, we obviously are not being ruled by benevolent demons even if we once were - and can only take the form of ruling ourselves by the divine element within us, our reason. Sagan, rather than dismissing Plato, could probably have taken some lessons from him in how to influence people.
The point for present purposes, however, is that Sagan was clearly wrong about Plato, and in a way that a simple reading of the passage in context would have revealed to any intelligent reader. Furthermore, Sagan doesn't know he is wrong, the way he knows he was wrong about atmospheric pressure on Venus. The lesson to take away is to trust whatever Carl Sagan says about strictly scientific issues concerning planetary science, and to take anything else he says with truckload of salt.
The conclusion for now is that the first principle in evaluating indirect knowledge is to consider the mediator in terms of his character as a knower in the general sense: Is he able to distinguish what he knows from what he doesn't know? Does he know the limits of his own expertise, what he really knows through it and what he doesn't? A mediator for whom positive answers can be given is more trustworthy, both in his area of expertise and in the likelihood of not pretending to pass off as expert knowledge that which was not. In any case, it is important to sift through for ourselves what an expert tells us, sorting out what his expertise really justifies and what it does not.