“The Structure of Justification
Finally, let us consider briefly another problem in epistemology, the problem about the “structure” of knowledge. There are at least three main views.
Foundationalism says that certain kinds of knowledge are “basic” and need no further justification. Other knowledge is built up as a superstructure on the basis of the foundation. For example, empiricism is a form of foundationalism, because it says that the knowledge of sense experience is basic, and that everything else derives from it. Other forms of foundationalism may argue that other kinds of knowledge are basic.
How does foundationalism approach a specific knowledge claim, like the claim that theft is wrong? It depends on which kinds of knowledge are basic. The idea that theft is wrong could be treated ei- ther as a basic form of knowledge (an intuitive dictate of conscience) or as a result built on the basis of a lot of reasoning about social benefits. This reasoning about social benefits would in turn be based on a foundation of previous knowledge about human beings and societies, which would eventually go back to sense data. At least for empiricism, the steps in tracing back to foundations would stop at that point, because empiricism thinks that sense data are basic.
A second view, coherentism, says that no beliefs are absolutely “basic” or foundational. All beliefs are justified by coherence with other beliefs. We have already met a similar approach in dealing with the nature of truth. How does this work out in practice? Juliet believes that theft is wrong because that belief harmonizes with other moral beliefs, all of which affirm the importance of respect- ing other people, and because it harmonizes with her observations about social benefits of not stealing. The difficulty here is the obvi- ous one: how do we avoid subjectivism, in which Juliet’s beliefs are internally coherent within her mind, but may not necessarily have any relation to the outside world?
A third view, contextualism, says that we seek justifications for belief only within relatively specific contexts. We take for granted most of what we believe and test a particular idea only within the context in which testing is appropriate. For example, Juliet tests her belief that theft is wrong within the context of other moral principles. She takes for granted many of her beliefs about other persons and their motivations and the ways that society functions. These beliefs offer a broader context in which she can draw conclu- sions about how theft impacts society.
Evaluating Theories of the Structure of Knowledge
Now how do we evaluate these three approaches to the structure of knowledge?
From a Christian point of view, humans are dependent and are not the ultimate standard for knowledge. So a foundational- ism that locates the foundation in something in the world or in the human mind is idolatrous—it replaces God with some created thing or some aspect of creation. For example, empiricism idolizes sense experience.
Yet a Christian does have a “foundation” in a sense. God is the ultimate source of knowledge and also the standard for knowledge. But it is also important to say that our knowledge of him is medi- ated through revelation. So we are always in a position of dependence, in which we depend not only on God himself, but also on tacit knowledge that God has given us. Our knowledge includes ac- quaintance with other persons, and that acquaintance, especially with parents, brothers, sisters, teachers, and school classmates, has been instrumental in bringing us to the state of knowledge in which we currently live as adults.
What about coherentism? A Christian should of course reject any coherentism that never acknowledges a more ultimate stan- dard than the ideas in one’s own mind. Yet we can also see a grain of truth in coherentism, because the process of growing up from childhood involves coherent interaction with parents, teachers, and the world. This interaction takes place according to God’s de- sign and God’s providence, and in the midst of God’s presence. Our beliefs change and grow as they interact with one another and with the beliefs of those around us. (So we should take into account the social dimension of knowledge.) In this process, we are interacting with divinely given norms. We are not imprisoned within our own “house” of belief, as a secular form of coherentism might suggest.
Finally, contextualism makes some sense against the back- ground of ordinary experience. Most of the time we take our beliefs for granted. Our critical inspection of a particular idea or belief usually takes place within some kind of limited context. Contex- tualism could be seen as simply a kind of observation about typi- cal human experience in ordinary situations. But contextualism is wrong if it pretends that we never ask more ultimate questions, such as how we justify knowledge as a whole, as opposed to how we justify a particular belief that theft is wrong. People do, after all, have the experience of asking more and more ultimate ques- tions. If theft is wrong, it must be because there are moral stan- dards that can be known. So what are moral standards, and how can they be known? When Juliet thinks that theft is wrong, is she just listening to her own preferences? Contextualism ignores such larger questions, or gives up on answering them, or becomes a form of coherentism.
If we like, we can see here a hint that several perspectives are at work. Foundationalism is like a particle perspective, at least with respect to basic knowledge. Each bit of basic knowledge is like a particle, distinct and not in need of further support. Co- herentism, by contrast, is akin to a field perspective. Each belief makes sense only when tested by means of its relations—coherent or incoherent?—with respect to an ever-widening circle of other beliefs. Contextualism is field-like as well, inasmuch as it appeals to contexts. But the contexts are limited; they are the contexts relevant to “local” problem solving. In this respect contextualism is suggestive of problem solving and has an affinity with the wave perspective, which asks about progress in time toward answering a question about knowledge.
These three approaches actually belong together. They are each useful perspectival starting points for considering the structure of knowledge. But each is inadequate when it is used to ignore the others.”
(Poythress, Redeeming Philosophy)