Foundationalism
The term “foundationalism” is now used–often with derogatory connotations and without any clear definition–both in literary and religious circles to refer to various positions that stand in contrast to relativism, such as the belief that there is absolute truth or a real world that we do not construct or the belief that it is possible to know anything rationally. These uses are not helpful and distract from the clear meaning the term has been given in epistemology, the branch of philosophy that deals with knowledge and rational belief.
Foundationalism as understood by philosophers is first and foremost a position regarding the structure of justified belief or of knowledge. A foundationalist holds that all inferred beliefs must, to meet the requirements of rationality, be supported by a finite chain or tree of supporting beliefs, rather than by loops or circles of inference or by an infinite regress of reasons. On this view, beliefs cannot justify or support themselves, even in the presence of other supporting premises. Foundationalism thus divides justified beliefs into those that are justified by inference from others and those, the foundations or terminating nodes of a structure of reasons, that are justified without the necessity for inference–those that are, in some sense, intrinsically reasonable for a subject to believe.
A chief structural rival to foundationalism is coherentism, a position according to which loops of justification are permitted, sometimes with the proviso that a belief must also have support that comes from beliefs other than itself. On a coherentist model, inferential justification resembles an interwoven mat or a thicket (without “roots”) rather than a chain or a tree.
It is important to bear in mind that foundationalism and its rivals, such as coherentism, are about the structure of a subject’s reasons at a particular time, not about the development of his belief system over time. Foundationalism therefore does not rule out the possibility that, at one time, a subject’s belief A could legitimately be used to support his belief B, while at another time B could support A.
Foundationalists defend their position by two types of arguments. On the one hand, they argue that non-foundational inference structures (those that have no terminating endpoints) are rationally defective. These arguments include explanations of the vitiating nature of both circular reasoning and infinite regresses of reasons.
On the other hand, foundationalists argue that there are foundational beliefs–beliefs that it is rational to hold without inferring them from anything else–and that these differ clearly from beliefs that do require support. Throughout the history of philosophy, foundationalists have usually restricted the set of foundational beliefs or propositions to a fairly small group of types which are deemed to have a special epistemic status. For many foundationalists (such as C. I. Lewis), these have included only beliefs about experiences to which a subject has immediate access, such as “I am having this kind of sensation,” and beliefs about mathematical or logical truths, such as “1 + 1=2,” that can be grasped by an immediate act of infallible cognitive intuition. Others have held that one can directly perceive at least some physical objects and that beliefs (like “that’s a dog”) about objects one is currently observing are also correctly included in the foundations, though beliefs about the past, objects not currently present or not readily observable (such as electrons), high-level scientific theories, and so forth are not foundational. In either case, the motivation for restricting the foundations has been to show that the division of beliefs into those that must be supported by reasons and those that are in one sense or another “self-evident” or “immediately justified” is itself a compelling and obvious component of the structure of rational thought.
More recently, some of the philosophers known as externalists have adopted the requirement for a foundational structure–no loops or infinite regresses–but have held that beliefs can be properly included in the foundations if they are the result of a “belief-forming mechanism” that is in fact reliable or “well-designed and properly functioning in its intended environment” but that does not as a matter of fact take other beliefs as its inputs. (It is not necessary to be a foundationalist to be an externalist, but some externalists are structural foundationalists.) Candidates for such processes have included sensory perception, memory, knowledge of other persons, and even (for the Christian philosopher Alvin Plantinga) the internal instigation of the Holy Spirit.
This version of foundationalism departs from older versions in the sense that, the broader the set of belief-forming processes that are taken to generate foundational beliefs, the less obvious is the distinction between foundational and non-foundational beliefs. This broadening of the foundations removes part of the rationale for a foundational structure–namely, the idea just discussed that foundational beliefs, which require no support, are strikingly and obviously different from all the rest, which do. Whether a belief is foundational or non-foundational will, on an externalist model, often depend only upon the facts of the matter in our world regarding what sorts of processes do reliably produce true beliefs without taking other beliefs as inputs, or what sorts of processes were given to us by design and reliably induce us to form beliefs non-inferentially. Many beliefs (such as “God exists” or “Einstein’s theory of relativity is true”) might on this view turn out to be foundational in virtue of the process that produces them, even though they seem prima facie to require argumentative and even empirical defense.
Thus in recent years the debate between foundationalists and coherentists has to some degree been overshadowed by that between internalists and externalists.
Lydia McGrew
Kalamazoo, MI
June, 2003
Book Resources On FoundationalismThe Foundations of Knowledge by Timothy McGrew
Warrant and Proper Function by Alvin Plantinga
Epistemic Justification by William Alston
|
|
|
