Epistemology naturalized
We have been discussing the relationship between justification and knowledge on the assumption that we can decide the issue by thought experiments.
Each time a proposal has been made, we have followed Socrates' example in the Theaetetus, testing the proposal against cases, like Goldman's Henry and the barns, or Gettier's Smith, Jones, and the coins in the pockets. This suggests that what we are doing is exploring the nature of our concepts of knowledge, belief and justification, on the assumption that we can always judge correctly whether these terms apply to particular cases. That is not an unreasonable assumption: anyone who knows English knows how to use the words “know,” “believe” and “justify”—knows, that is, what those words mean. And surely someone who knows what those words mean knows when they can and cannot properly be applied. But if we know what these words mean, why can't we just say what they mean? Why, that is, has it been so hard to find an answer to Socrates' definitional question, “What is knowledge?” It looks as though, on one hand, we can tell when the word “know” applies in a case (provided we are told enough about it) but, on the other, we are not very good at uncovering and explaining how we tell whether it applies. If we could tell, then we would surely have agreed on an answer to the definitional question long ago.I shall return to questions about the relationship between our knowledge of the meanings of the words in our language and our ability to spell out what we know in the next chapter; see 3.13. For now, however, I want to observe that we could have proceeded in a different way. We could have drawn not just on our intuitive understanding of the concepts of knowledge and justification but also on scientific study of the processes by which people come to believe things, on cognitive psychology, for example, or the sociology of knowledge.
We could, that is, have taken up the study of knowledge not as a purely conceptual inquiry but alongside work done in the sciences. To take that approach to epistemology would be to follow the recommendation of the American philosopher W. V. O. Quine, who proposed in 1969 that we should “naturalize” epistemology. In a famous article, entitled “Epistemology Naturalized,” Quine suggested that epistemology should be “a chapter of psychology and hence of natural science.”This is a slightly surprising proposal, because, as we have seen, inquiring into the nature of knowledge involves thinking about when and how our beliefs are justified. To claim that a belief is justified is not just to say when it will be believed but also to say when it ought to be believed. And we don't normally think of natural science as telling us what we ought to do. Science, surely, is about describing and explaining the world, not about what we should do?
One way to reconcile these two ideas would be to build on the central idea of reliabilism and say that what psychology can teach us is which belief-forming processes are in fact reliable. So here epistemology and psychology would go hand in hand. Epistemology would tell us that we ought to form our beliefs in ways that are reliable, while psychology examines which ways these are: so the “ought” comes from epistemology, not from psychology, leaving us able to continue to think of natural science as free of “oughts.” Claims about what people ought to do, say, or believe are prescriptive: they don't just describe what people do, they prescribe what they ought to do. So this way of dividing up the job between psychology and epistemology leaves epistemology the job of prescription and retains the view that psychology describes our mental processes. Quine suggested later that the “oughts” of epistemology are like the “oughts” of engineering: when Emma the engineer says that you ought to use steel of a certain strength in making a bridge, she means only that you should use that steel if you want the bridge to hold up under the load it is going to have to bear.
The “ought” is conditional: it assumes a certain aim, in this case to build a bridge that will take a certain load. We shall see, later, when we come to discuss morality, that the German philosopher Immanuel Kant argued that moral “oughts” were not conditional—his term was “hypothetical”—in this way. Rather, they were what he called “categorical.” (See 5.3.)So what is the aim upon which the “oughts” of epistemology are conditional?
The obvious answer, as Quine proposed, is that epistemology says you ought to believe what you are justified in believing if you want to have true beliefs. And that suggests a way of formulating an understanding of what knowledge is: it is true belief produced by processes that normally produce true beliefs. Understood that way, we can see the tradition of phenomenological approaches to justification as a series of hypotheses about what processes are most likely to produce true belief. In the empiricist tradition, it was assumed that we are so constructed that we will usually get true beliefs if we believe our senses. Simply coming to believe what we are naturally disposed to believe on the basis of our senses is therefore justified, and so we do not need to study our own sensory systems in order to get closer to the truth. But the existence of hallucinations and illusions—both of which Descartes discussed—shows, of course, that our senses are not, in fact, so reliable that we cannot learn from studying them about better ways of forming beliefs. And once we see that, we can see that a foundationalist empiricism, which treats what our senses tell us as a secure foundation for all our other beliefs, is not warranted.
Similarly, when rationalists say that reason is the major source of our knowledge, they are assuming that we are so constructed that we will usually get true beliefs if we follow what Descartes called the “natural light” of reason.
But experience has taught us that our reasoning capacities are in fact quite limited: people regularly make elementary logical mistakes, for example. Furthermore (as Descartes, who was something of a scientist, knew very well), reason by itself cannot lead us to the truth about the world around us. So here too there are grounds for doubt that relying on this method will get us to the truth.As a result, then, of the development of naturalized epistemology, there has been increasing interest in using the insights gained from scientific study of the ways in which we acquire our beliefs to enhance our grasp of the nature of knowledge. This approach has led to the development of evolutionary epistemology, which draws on Darwin's ideas about evolution in two important—and importantly distinct—respects. First, evolutionary epistemology examines the consequences of the fact that our cognitive capacities are themselves the product of an evolutionary process. And second, it explores how ideas and theories compete with each other and are selected, in a way that is somewhat analogous to the process of the natural selection of biological traits. Here, then, the philosopher's interest in questions about knowledge comes into close interaction with the work of biologists and psychologists.
2.10