Problem of Epistemological Foundationalism (Článek)
Keywords: foundationalism, epistemology, noetics, knowing, coherence, truth
Klíčová slova: fundacionalismus, epistemologie, noetika, vědění, koherence, pravda
PDF version (146 KB)
The term “foundationalism” is today used with both deprecating and complimentary connotations depending on which philosophical or religious circles are speaking. On the one hand we have one camp of philosophers who are virtually unanimous in their opinion that foundationalism, which characterized modernism, has to be abandoned. Thus we read announcements of the demise of foundationalism like those of D. R. Street, J. R. Franke or Stanley Grenz who say that “…foundationalism is in dramatic retreat”. (Grenz, Franke 2001: 12)  In the same line speak number of other authors. Wentzel van Huyssteen, for instance, says: “Whatever notion of postmodernity we eventually opt for, all postmodern thinkers see the modernist quest for certainty, and the accompanying program of laying foundations for our knowledge, as a dream for the impossible, a contemporary version of the quest for the Holy Grail.” (Huyssteen 1998: 216) Nicholas Wolterstorff uses similar rhetoric: “On all fronts foundationalism is in bad shape. It seems to me there is nothing to do but give it up for mortally ill and learn to live in its absence.” (Wolterstorff 1976: 52)
On the other hand there are thinkers that take foundationalism as the most obvious, even inescapable part of human noetic structure. P. K. Moser for example concludes his article on this topic: “This explains why foundationalism has been very prominent historically and is still widely held in contemporary epistemology.” (Moser 1995: 323) Other thinkers speak similarly. J. L. Kvanvig, for instance, says that in spite of some attacks, “many epistemologists were not convinced that foundationalism was to be abandoned even if the criticisms were granted. According to these epistemologists, far from having shown that foundationalism itself was at fault, the critics of foundationalism had only been attacking one particular version of foundationalism…” (Kvanvig 1986: 345) In the same line speak other kindred spirits, for example William Alston (1976a; 1976b), Cornelius F. Delaney (1976), Kelly J. Clark (1990) or Mark Pastin (1975).
To avoid sloganeering, it is necessary to approach the problem with appropriate carefulness and take into account the considerable amount of discussion that goes on today in both camps of philosophical circles, for it seems that the power of preconception is effectively strong in philosophical methodology as well as in any area of human research.
The task of this essay, therefore, is threefold: 1) to define foundationalism, 2) to bring order to the terminological confusion and map the typology, and 3) to assess the competing epistemological theories (namely coherentism and pragmatism). These three steps should help the reader to make some preliminary conclusions concerning the in/adequacy of foundationalism as an epistemological theory.
2 Definition of foundationalism
The whole idea of foundationalism is hidden behind a simple question that is often legitimately asked in our everyday conversations: How do you know? The normal answer in normal situation attempts to provide some reasonable basis for the claim. For example, if a friend tells me that the dean of our college is a thief, I might be not willing to believe such a claim without any support, indeed, I might be even offended by such an accusation and require some good explanation without delay. If the answer I get is that Peter said so, it is legitimate for me to ask, how does Peter know? Usually we do not accept an argument such as that Peter heard it from Paul, so the conversation continues till the claim is based on some good acceptable foundation or refuted as unwarranted (malicious gossip – in this case).
The above example indicates the traditional motivation for foundationalism that has been formulated as the so called regress argument. The argument has a long history, it goes back to Aristotle. In the simplest form, it says that there must be a basis for all truth claims (even for one such as that of my example above) and that infinite regress never provides a foundation, it only delays providing one forever. Hence, there must be some foundation/s for our claims and knowledge in general, if they are to be taken seriously (comp. Aristotle’s Posterior Analytics I.3:5–23).
Among the epistemologists – whether foundationalists or non-foundationalists – there is consensus in definition of foundationalism, though the actual wording may differ. Let me quote two examples of such definitions from two different philosophical dictionaries:
“Foundationalism [is] the view that knowledge and epistemic justification have a two-tier structure: some instances of knowledge are non-inferential; and all other instances thereof are inferential, or non-foundational, in that they derive ultimately from foundational knowledge or justification.” (Moser 1995: 321)
“Foundationalism is a view about the structure of justification or knowledge. The foundationalist’s thesis in short is that all knowledge and justified belief rest ultimately on a foundation of noninferential knowledge or justified belief.” 
Graphically foundationalism could be expressed like this:
An important issue in foundationalist epistemological structure is principle called basing relation. It deals with the relationship between basic and non-basic beliefs. For example the belief that the leaves are rustling stands in the basing relation to the belief that there is a wind blowing. The basing relation is characterized by two features: irreflexiveness and asymmetry. J. P. Moreland and W. L. Craig explain:
“A relation is irreflexive if something cannot stand to that relationship to itself. For example, ‘larger than’ is irreflexive since nothing is larger than itself. Applied to beliefs, this means that no belief is based on itself. A self-evident belief is not based on itself, even though it is justified immediately; rather, it is grounded in its experienced luster or obviousness, in the felt unavoidable inclination to believe it. A relation is asymmetrical in that, given two things A and B, if A stands in that relation to B, then B does not stand in that relation to A. ‘Larger than’ is asymmetrical. If A is larger than B, B cannot be larger than A.” (Moreland, Craig 2003: 114)
When defining foundationalism, it is also important to note one essential issue – the foundationalist approach to knowledge presupposes certain epistemological givens, which is of course a subject of great criticism.
All foundationalists, whether they reflect on it or not, clearly assume that the reality is cognizable by our (human) cognitive faculties, thus they presuppose the so called correspondence theory of knowledge (or of truth). Some philosophers acknowledge this and reflect on the correspondence theory in their writings,  other foundationalists simply assume it without any deeper reflection.  There are, however, philosophers-foundationalists that do not just acknowledge correspondence theory in their noetic system, but also ask how is correspondence possible in the first place. I will examine these issues in concluding chapter of this essay.
The study of definitions of foundationalism leads to the important preliminary observation that the contemporary foundationalists agree on two things: 1) their general presupposition of correspondence between reality and cognitive faculties; and 2) their understanding of the two-tier structure of epistemic justification. What is a matter of discussion, however, is the precise understanding of the nature of the basic or foundational beliefs.  It is here where foundationalists differ and form various types of foundationalism – as we shall see.
3 Typology of foundationalism
To classify different types of foundationalism is not an easy task, for different authors employ not only different criteria of classification, but also different terminology. Most foundationalists differ over the issue of the nature of basic or foundational beliefs. There are at least three issues concerning the basic beliefs over which foundationalists differ.
1. The first issue is the extent of basic beliefs, that is, which beliefs should constitute the foundations. According to classical foundationalists there are very strict rules about what legitimately counts as a basic belief. They argue that only self-evident truths of reason and sensory beliefs should be allowed in the foundation (comp. Moreland, Craig 2003: 112). This type of foundationalism is sometimes also called narrow or strict foundationalism (Nash 1999: 272ff).
Other foundationalists – typically adherents of Alvin Plantinga – argue that the foundations should include a much wider range of beliefs. K. J. Clark enumerates for instance: perceptual beliefs, memory beliefs, beliefs accepted on the basis of testimony, beliefs in other minds, beliefs in an external world, mathematical beliefs, elementary truth of logic, etc. This kind of foundationalism is usually called broad or modest foundationalism. It is almost needless to say that the latter version is much more common today.
2. Another criteria of discernment amongst foundationalists is the issue of conditions necessary for basic beliefs to be counted as basic (or sometimes called properly basic). Some foundationalists are internalists, some externalists. The internalists claim that the conditions that ground the properly basic beliefs are internal to the knower. For example the belief that it is raining outside at the moment is grounded in sensory experience (my seeing and hearing) which is internal to the knowing subject. Likewise it is, according to the internalists, with all basic beliefs – they are all grounded in something within the knower.
Externalists, on the other hand, believe that the factors grounding the basic beliefs are not those to which the subject must have internal access. Some authors use adjectives internal/external interchangeably with deep/shallow.
3. The foundationalists also differ about the question of how strong the justification for foundational beliefs is. Strong foundationalists claim that the foundational beliefs are certain, infallible, indubitable or incorrigible. The meaning of each term differs slightly, but the point of this view is that if a belief is to be qualified as foundational it must exhibit certain “epistemic immunities” – it must be immune to correction, incapable of being doubted reasonably, incapable of being mistakenly believed and the like.
Weak foundationalists of course reject such strong claims. They argue that foundational beliefs need to be merely prima facie justified. That means that a person is justified in believing something just in case he or she has no reason to think there are some defeaters sufficient to remove his or her justification. This version is also called by some authors fallibilist foundationalism (comp. Erickson 2002: 257).
To be sure, many authors do not go into such typological details in their treatment of the topic. Such typologies are then usually incomplete and partially overlap with other typologies. 
B. Tuchanska  for example, distinguishes only three types of foundationalism: 1) Empiricist foundationalism, where the basis of knowledge is constituted by the “immediately given”, i.e., sense data. 2) Rationalist foundationalism searches, according to Tuchanska, for “a sourceful cognitive act, i.e., Cartesian Cogito” that would reveal the ultimate foundations of knowledge and certainty. 3) Transcendentalist foundationalism aspires, unlike to the two previous types, to solve the question of how cognition as a relation between a subject and object is possible, or – in other words – how synthetic representations and their objects can establish connection. The main representative of this type of foundationalism is, according to Tuchanska, obviously Immanuel Kant.
D. K. Clark (not to be confused with K. J. Clark) approaches the problem of typology from a completely different point. He speaks about two main types of foundationalism, namely source-foundationalism and belief-foundationalism, and two subtypes of the latter one – classical and soft foundationalism. Clark explains that source-foundationalism is a “typically modernist notion that knowledge is the reflection of truth and that we can discover a stable foundation for it in God, History or Reason”. (Clark 2003: 153) Belief-foundationalism, on the other hand, refers to a class of theories about individual items of knowledge: it “focuses on the relationship between individual beliefs within a human noetic structure”. (Clark 2003: 154) Clark’s classical foundationalism, then, corresponds to the classical (narrow) foundationalism in the above given typology, but his soft foundationalism somewhat overlaps with broad and weak foundationalism.
It is needless to say there are many other authors that approach the problem in their own way and thus there are many more overlapping typologies and sub-typologies,  but according to our knowledge, we have covered all the main varieties.
4 Criticism of foundationalism: Alternatives
Besides the traditional theories of epistemic justification, there have recently appeared other theories, particularly the deflationary, redundancy, minimalist, disquotation and semantic theories. But these theories are quite minor and recent. Let them be a subject of the test of time to see whether they are worthy of any further attention. In this essay, I will examine only two theories that have historically been the main competitors to foundationalism which (more or less consciously) presupposes epistemological realism, that is the view of correspondence. 
Historically, the greatest critic and also the most significant alternative to foundationalism has been coherentism. The contrast between coherentism and foundationalism is best evident in the regress argument problem. Let me recall the basic notion of the argument. If we are asked what justifies one of our beliefs, we typically answer by providing some other belief that holds it up. If we are asked about this second belief, we usually come up with a third belief, and so on. There are only three possible ends for such a chain: 1) it could go on forever, 2) it could eventually result in some foundational belief, or 3) it could come back to itself, i.e., contain a belief that has already occurred earlier in the chain. Since the infinite chains are not in normal argumentation possible, we are left with a choice between chains that end and circular chains. Although traditionally the circular reasoning is understood as logical fallacy, the coherentists believe that when the circle is large and rich enough, it is acceptable as a good ground for a belief.
Thus coherentists reject the traditional foundationalistic building (or pyramid) metaphor where the more basic beliefs provide a foundation for the non-basic ones, and replace it with a net metaphor. According to coherentists “all beliefs representing knowledge are known or justified in virtue of their relations to other beliefs, specifically, in virtue of belonging to a coherent system of beliefs”. (Moser 1995: 154) Thus from the coherentist point of view, “there is no basic, privileged class of beliefs that serve as foundation for justifying other beliefs but which need no justification from other beliefs”. (Moreland, Craig 2003: 122)
Most coherentists today usually prefer a holistic form to the linear one. According to linear coherentism, beliefs are justified by other individual beliefs in a linear, circular, inferential chain. Holistic form, on the other hand, means that individual beliefs are related in a highly complex and mutually integrated network where the so called independence principle plays an important role. This principle says “that the larger the number of independent mutually consistent factors one believes to support (or constitute evidence for) the truth of a proposition, the better one’s justification for believing it”. (Audi 2003: 210) In such a network the connection between individual independent beliefs does not merely transfer warrant as in foundationalism, it guarantees warrant. Amongst coherentists there has been a discussion about what coherence itself amounts to. This discussion caused some divisions,  but the basic principle remains: logical consistency is all that is necessary and sufficient for warranting a belief or a set of beliefs.
Critics of coherentism point out several main problems. First, coherentism “seems too much to ask”, says Susan Haack (1993: 114). Is it possible for a person to think through all his or her beliefs to ascertain whether all of them are interdependent and coherent? Surely this might be an ideal, but it hardly seems possible. If it is actually impossible, then no one could ever be justified in holding any beliefs.
The second objection is sometimes called the plurality objection. It states that there could be two or more alternative coherent set of beliefs which are, nevertheless, logically incompatible with each other. Since coherentism makes coherence not just a necessary, but also a sufficient condition for justification, how shall we solve such dilemma? A mentally ill person, for example, who thinks that he is Napoleon might have an equally coherent set of beliefs as those of his therapist. According to coherentism both sets of beliefs are equally justified, but obviously this is not the case.
The third problem that is often discussed amongst the critics of coherentism might be called the “drunken sailor” or isolation problem. If beliefs are holding each other up, and thus creating a coherent system, what is holding the beliefs? The important questions here are: Where did we get our beliefs from? Why did we choose these beliefs, and how are they holding each other up? Since there is no “privileged class of beliefs that serve as foundation”, how shall we distinguish between coherent illusions that do not have justification and sets of beliefs that do have justification?
As a philosophical theory pragmatism was advanced by American thinkers Charles S. Peirce, William James and John Dewey. Recently it has experienced a certain revival through the pen of Richard Rorty, who considers himself to be a neo-pragmatist.
Pragmatism as a theory of knowledge stresses the practical aspect of knowing. In general it says that a belief A is true if and only if A works or is useful to have. In other words, a person is justified in believing any proposition or theory that produces good results. According to W. James, “truth is the expedient in the way of knowing. A statement is known to be true, if it brings the right results”. (Geisler 1999: 741) That is why for pragmatists the main task of scientific inquiry is not and should not be an accurate description of the external world, but rather to help us to cope with the world. Rorty, for example, clearly suggests that our inquiries should avoid trying to search for any truth with a capital T: “For pragmatists, true sentences are not true because they correspond to reality, and so there is no need to worry what sort of reality, if any, a given sentence corresponds to – no need to worry about what ‘makes’ it true.” (Rorty 1982: 15)
D. K. Clark comments on this: “Since there is no one actual truth, ‘truth’ is whatever your peers allow you to get away with.” (Clark 2003: 159) The practical benefit of this approach is clear – especially in the times of collapse of the modernist (often oppressive, colonial etc.) program – every individual or community can find or develop their own unique pattern for living. Contemporary Western men or women might prefer different patterns than the ones of, e.g., Chinese Thervada Buddhists, Nuwaubianists  or Cheondoists  but it is perfectly legitimate and acceptable to hold them.
The pragmatic view has been extensively criticized, for even the intuitive reflection says that something is not true simply because it works. If pragmatic utility justifies us in believing certain things, than we could be justified in believing things that are obviously false. Critics point out that if truth is not a matter of fact, but a matter of what a given community allows in its midst, it is not difficult to see the disastrous ethical implications. Consider for example the case of the German leaders who insisted that the concentration camps where Jews were massively killed never existed (and there are still today neo-Nazi groups that reject the historical existence of Holocaust). Similarly, the Chinese leaders had for a long time insisted that the murders of hundreds of demonstrators in Tiananmen Square never happened.
If the pragmatic view of knowledge is correct, than a person is justified in holding a claim as true on the basis that it is useful for his or her group. Within the community of the political leaders the belief that those things never happened does have utility. Peers of these leaders let their peers get away with such claim, for it produced certain political benefit. Is such claim justified then? Yes, according to pragmatism. It is Rorty’s theory in practice. Alvin Plantinga comments on the pragmatic implications in his own style: “…if you have done something wrong, it is not too late: lie about it, thus bringing it about that your peers will let you get away with saying you did not do it; then it will be true both that you did not do it, and, as an added bonus, that you did not even lie about it. One hopes Rorty is just joshing the rest of us. But he isn’t.” (Plantiga 1994: 14)
Another objection against pragmatism is that it neglects the fact that it needs non-pragmatic justification for its own beliefs. In relation to science Rorty once claimed: “The only sense in which science is exemplary is that it is a model of human solidarity.” (Rorty 1991: 39) But here is the problem: What warrants the belief that solidarity is something desirable or good? Rorty asserts that the ideal free democratic community “serves no higher end than its own preservation and self-improvement, the preservation and enhancement of civilization. It would identify rationality with that effort, rather than with the desire for objectivity. So it would feel no need for foundation more solid than reciprocal loyalty”. (Rorty 1991: 45)
But such an argument of course rests on certain (hidden or forgotten) assumptions about objective reality. It assumes that certain values of this world, specifically self-improvement, freedom, loyalty, civilization are in some trans-systemic, objective way good. Since Rorty announced the death of epistemology, how does he know?
So the main argument against pragmatism is that in justifying beliefs about the goals pragmatists pursue, they must either depend on non-pragmatic strategy or else baldly (or fideisticly) assert their preferred goals.
5.1 Concluding observations
Three major approaches to the warranting of individual beliefs were discussed: foundationalism, coherentism and pragmatism. As for foundationalism, I consider the strict, narrow, strong or classical (depending on what terminology we employ) versions as hardly defendable, for it is much too restrictive. If only self-evident or incorrigible beliefs were rationally acceptable as properly basic, then most of our ordinary everyday beliefs would be excluded as irrational. That is one of the main objections formulated by A. Plantinga: “Consider all those propositions that entail, say, that there are enduring physical objects (this is the problem of external world), or that there are persons distinct from myself (this is the problem of other minds), or that the world has existed for more than five minutes; none of these propositions, I think, is more probable than not with respect to what is self evident or incorrigible for me.” (Plantiga 1983: 59–60)
My observation, therefore, is that the knowing processes by which humans do actually come to know reality are much more open and complex than the rigid type foundationalism allows for.
The debate between foundationalism on one side and coherentism and pragmatism on the other has definitely been beneficial and illuminating for both sides, for the argumentation revealed that the knowing process is more complex than the past epistemologists thought. On the one hand, the coherentists correctly stressed that coherence is an important, even necessary condition of any true set of beliefs. Coherentism also rightly emphasizes the independence principle which says that the larger the number of independent, but mutually coherent beliefs there are, the more adequate is the warrant of a claim.
Similarly the pragmatists correctly stressed the practical aspect of truth. What is true should definitely work, at least in the long run.
On the other hand the advocates of coherentism and pragmatism could gain a lot from the debate too. R. Audi summarizes the essence of the argument like this: “Correspondence theorists have replied that points made by proponents of coherence and pragmatic theories of truth confuse the criteria of truth, roughly, the standards for determining whether a proposition is true, with the nature of truth, what it is. Turning blue litmus paper red is a criterion of acidity, but it is not what acidity is…” (Audi 2003: 249–250)
In other words, what is true will definitely be coherent, and likewise, what is true will work, but not the other way round – the fact that something coheres or works, does not make it true. 
5.2 Concluding proposals
My “modest” proposal to the epistemological discussion argues for a modest foundationalism. In the epistemological literature the reader may come across other adjectives qualifying the kind of foundationalism I argue for – for example moderate, soft or fallible – but I consider the word modest as most appropriate from the language-aesthetic-emotional point of view.
After studying all the varied approaches, what then should characterize or constitute the theory of epistemic justification, which is called modest foundationalism?
I. Modest foundationalism holds that the two-tier structure of epistemic justification is the most adequate one – there are many non-basic beliefs that are in a variety of ways grounded in basic beliefs.
II. Modest foundationalism suggests that the inferential connections that transmit justification can take a variety of forms. They need not be only deductive, but can also be inductive or abductive (retro-ductive).
III. Modest foundationalism recognizes that relatively large numbers of beliefs might be considered as basic or foundational. In contrast with strict classical foundationalism, the modest version includes in the set of basic beliefs not only the self-evident or incorrigible beliefs, but also sensory beliefs, memorial beliefs, testimonial beliefs, beliefs about external world, beliefs in other minds, beliefs about the past, etc.
IV. Modest foundationalism reflects on its cognitive-metaphysical presuppositions. This characteristic logically arises from the previous one – since my view allows beliefs produced immediately by my cognitive faculties to form the body of basic beliefs, how can I trust my cognitive faculties? The answer is, because my view presupposes the existence of Creator who created both the human cognitive faculties and the external reality that is cognizable by these faculties; there is a correspondence. I admit that it is a metaphysical presupposition, it is a starting point of one stream of epistemology.  Such a presupposition encourages a philosopher to use and rely on the cognitive faculties in the ultimate search for truth.
V. Modest foundationalism reflects on the nature of the reliability of the cognitive faculties. A modest foundationalist recognizes that he or she is a “a man of unclean cognitive faculties, and dwells in the midst of a people of unclean cognitive faculties”, that is, he or she is aware of his/her own cognitive limitations.  That is why modest foundationalists agree that both foundational and super-structural justification is characterized by defeasibility – our beliefs may be defeated. Modest foundationalists are aware of their capability for making mistakes and so are reflectively open to subject their beliefs to further inspection that may refine them. 
VI. The next characteristic of modest foundationalism is again logically tied with the previous one. In modest foundationalism coherence is extremely relevant. Although coherence alone is never sufficient to ground an entire web of belief,  it is very important in warranting particular beliefs. Incoherence may serve to defeat justified, even directly justified, and hence foundational, beliefs.
After consideration of the non-foundationalist alternative accounts of the structure of the epistemic justification my conclusion is that the foundationalist metaphor requires certain critical qualifications, but need not be discarded completely – as some postmodern critics believe. Such a criticism of foundationalism is usually directed at the Cartesian version of foundationalism and as such is irrelevant to the contemporary discussion, which is much more subtle and elaborate. Since about 1975 quite a large number of foundationalisms – that differ significantly from the classical version – have appeared on the scene. It is not possible to lump together all the types of foundationalism as the version of Descartes’ rationalism and then dismiss it. The significant distinctions must be taken into account if the scientific discussion is to be accurate.
The account of foundationalism given above implies that a very large variety of epistemological views (if not all) can be classified as foundationalism. For every view or conviction has some basis on which it rests – whether the philosopher reflects on it or not. Even a non-foundationalist has a basis for his or her non-foundationalism. A philosopher that embraces an extreme form of relativism does so because he or she thinks there are good reasons for accepting it. Belief that there are no noetic absolutes accessible to human beings inferentially supports – and so functions as a foundation for – methodological non-foundationalism.
So the careful study of the different epistemological approaches helps us to understand that the mere absence (or even rejection) of the word ‘foundation’ in one’s epistemic rhetoric does not mean that the philosopher has managed to elude the foundationalistic structure of his or her argument.
 See also D. R. Street’s article Faith Without Foundations: Christian Epistemology and Apologetics after Modernity, URL:< http://www.geocities.com/danstreett/index.html >.
 Fumerton, R. Foundationalist Theories of Epistemic Justification. The Stanford Encyclopedia of Philosophy (2010 Edition), Edward N. Zalta (ed.),
 The concept of abductive (or retroductive) reasoning is not new – already S. C. Pierce’s spoke about it. But amongst Christian foundationalists it is an issue only in recent literature. See e.g. McGrath (2001–2003) or Clark (2003). J. Wood also speaks of an explanatory inference, mutual support and cognitive spontaneity, see Wood (1998).
 See for example Groothuis (2000) or Moreland, Craig (2003).
 Even the two definitions foundationalism given above are a representative examples that we have chosen deliberately. The first definition deals just with the structure aspect of knowledge; the latter definition, however, recognises and states what makes the basicality of basic beliefs – self-evidence and self-justification.
 A. Plantinga for example dropped an “atomic bomb” to the epistemological circles recently by introducing the idea that religious beliefs can be “properly basic”. He argues that religious beliefs may be directly grounded in religious experience without the mediation of other beliefs just as perceptual beliefs are directly grounded in perceptual experience. Many epistemologists found this idea controversial. For more details see Plantinga (2000).
 The same is probably true with our attempt to systematize the problem of typology, as our knowledge is lamentably limited.
 See her essay: Is a Non-Foundationalist Epistemology Possible? Available on
 See for example: Triplett (1990), Audi (2003), McGrath (2002, Reality).
 I am leaving out completely the view of noetic scepticism, for I do not consider it a meaningful alternative – how could we meaningfully know that we cannot know anything?
 Different coherentists have argued for different conditions on what coherence must include. Some argue for entailment coherence, that means that a set of beliefs is coherent only if each member of the set is entailed by all the other members of the set. Other coherentists stressed so called explanatory coherence: each member of a set of beliefs helps to explain and is explained by the other members of the set. Yet another coherentists argued for probability coherence which affirms that a set of beliefs is coherent only if it does not include beliefs that are improbable. For more detailed treatment of the problem see Moreland, Craig (2003).
 Nuwaubianism is an umbrella term used to refer to the doctrines and teachings of the followers of Dwight York. The Nuwaubians originated as a Black Muslim group in
 Cheondoism is a 20th-century Korean religious movement that has its origins in a peasant rebellion in 1812. It is rooted in Korean shamanism and Korean Buddhism, with elements drawn from Christianity. Its theology is a mixture of monotheistic, pantheistic, panentheistic principles.
 N. Geisler provides an example: “A set of false statements also can be internally consistent. If several witnesses conspire to misinterpret the facts, their story may cohere better if they were honestly trying to reconstruct the truth. But still it is a lie. At best, coherence is a negative test of truth. Statements are wrong if they are inconsistent, but not necessarily true if they are.” Quoted from Geisler (1999: 741).
 I am aware that such position might be charged as fideism. However, it is obvious that every stream of epistemology (or philosophy) has its meta-epistemological presupposition(s), its starting point – whether it is reflected upon or not. The question is what presupposition are we to adopt? We adopt the Creator presupposition because we are convinced that the retro-ductive reflections seem to confirm that this presupposition best explains the reality. Starting with such a presupposition encourages a philosopher to explore reality with a constant coming back to examination of the consistency and adequacy of the starting presupposition. Let us consider a negative example. (Pre)suppose there is no Creator who would have a good and deliberate intention to order things in a cognisable way. How do we account, then, for the everyday experience of an orderly cognition of the universe? (If the cows were flying, we would be surprised.) Or let us suppose our cognitive faculties are not reliable. How could we know that? If our cognitive faculties were not reliable how could we ever learn about it, since the only means of knowing it are the cognitive faculties. Is it possible to know reliably that our knowing faculties are unreliable? Would not that be a clear example of the self-referential incoherence?
 In other words, the modest foundationalist recognises that his/her cognition is always theory laden. However, that does not mean that the laden-ness cannot be reflected and overcome.
 An example may help here. Suppose my wife is a professional piano player who trains every afternoon from three to six p.m. One day I am sitting in my office and at three p.m. I hear the piano from her room. The conviction that arises through my normally functioning belief-forming faculties is that my wife is practising again. But later I find out that that it was a CD player, for my wife wanted to take a rest that day. Note that my original belief was perfectly rational. Anyone with normally functioning cognitive faculties would form such a belief in that circumstance. But with the new information I learn the reality is different.
 As well as the pragmatic criteria alone is not sufficient.
Alston, William. Two Types of Foundationalism. The Journal of Philosophy 73, 1976a, 165–185. ISSN 0022-362X.
Alston, William. Has foundationalism been refuted? Philosophical Studies 29, 1976b, 287–305. ISSN 0031-8116.
Audi, Robert. Epistemology. A Contemporary Introduction to the Theory of Knowledge. New York: Routledge, 2003. ISBN 978-0-415-28108-9.
Audi, Robert (ed.). The Cambridge Dictionary of Philosophy. New York: Cambridge University Press, 1995. ISBN 978-0-521-48328-5.
Clark, David K. To Know and Love God, Method for Theology. Wheaton: Crossway Books, 2003. ISBN 978-1-4335-1605-4.
Clark, Kelly J. Return to Reason. Grand Rapids: Eerdmans, 1990.
Delaney, Cornelius F. Foundations of Empirical Knowledge – Again. New Scholasticism L, 1–19, 1976. ISSN 0028-6621.
Erickson, Millard J. Truth or Consequences. The Promise & Perils of Postmodernism. Illinois: IVP, 2002. ISBN 978-0-8308-2657-5.
Fumerton, Richard. Foundationalist Theories of Epistemic Justification. The Stanford Encyclopedia of Philosophy (Summer 2010 Edition), ZALTA, Edward N. (ed.),
Geisler, Norman L. Baker Encyclopedia of Christian Apologetics. Grand Rapids: Baker, 1999. ISBN 978-0-8010-2151-0.
Grenz, Stanley J., Franke John R. Beyond Foundationalism, Shaping Theology in a Postmodern Context. Westminster: John Knox Press, 2001. ISBN 978-0-664-25769-9.
Groothuis, Douglas. Truth Decay. Illinois: IVP, 2000. ISBN 978-0-8308-2228-7.
Haack, Susan. Evidence and Inquiry: Towards Reconstruction in Epistemology. Oxford: Blackwell Publishers, 1993. ISBN 978-0-631-11851-0.
Kvanvig, Jonathan L. The confusion over Foundationalism. Philosophia, 1986, 16, 3–4, 345–354. ISSN 0048-3893.
Meek, Esther L. Longing to Know, The Philosophy of Knowledge for Ordinary People. Chicago: Brazos Press, 2003. ISBN 978-1-58743-060-2.
McGRATH, Alister E. A Scientific Theology: Nature, Reality, Theory, 3 vols. Grand Rapids: Eerdmans, 2001–2003. ISBN 978-0-8028-6484-0, 978-0-567-08888-8, 978-0-567-08349-4.
Moreland, James P.; Craig, William L. Philosophical Foundations for a Christian Worldview. Illinois: IVP, 2003. ISBN 978-0-8308-2694-0.
Moser, Paul K. Foundationalism. In Audi, Robert (ed.). The Cambridge Dictionary of Philosophy. New York: Cambridge University Press, 1995. ISBN 978-0-521-48328-5.
Nash, Ronald H. Life’s Ultimate Questions. An Introduction to Philosophy. Grand Rapids: Zondervan, 1999. ISBN 978-0-310-22364-1.
Pastin, Mark. Modest Foundationalism and Self-Warrant. American Philosophical Quarterly 4, 1975, 141–149. ISSN 0003-0481.
Plantinga, Alvin. On Christian Scholarship. Religious and Theological Studies Bulletin, 1994, vol. 5, November/December.
Plantinga, Alvin. Reason and Belief in God. In Plantinga, Alvin, Wolterstorff, Nicholas (eds.). Faith and Rationality. Notre Dame: University Of Notre Dame Press, 1983. ISBN 978-0-268-00964-9.
Plantinga, Alvin. Warranted Christian Beliefs. New York: Oxford University Press, 2000.
Rorty, Richard. Consequences of Pragmatism, Essays 1972–1980. Minneapolis: University of Minnesota Press, 1982. ISBN 978-0-8166-1063-1.
Rorty, Richard. Science as Solidarity. Objectivity, Relativism and Truth, Philosophical Papers vol. I. Cambridge: Cambridge University Press, 1991. ISBN 978-0-521-35877-4.
Triplett, Tim. Recent Work on Foundationalism. American Philosophical Quarterly 27, 1990, 2, 93. ISSN 0003-0481.
van Huyssteen, Wentzel. Tradition and the Task of Theology. Theology Today 55/2, 1998.
Wolterstorff, Nicholas. Reason Within the Bounds of Religion. Grand Rapids: Eerdmans, 1976. ISBN 978-0-8028-1643-6.
Wood, W. Jay. Epistemology. Becoming Intellectually Virtuous (Contours of Christian Philosophy). Illionois: IVP, 1998. ISBN 978-0-87784-522-5.
(PhDr. Jan Hábl, Ph.D. teaches philosophy of education and pedagogy at the University of Hradec Králové. He studied education at J. E. Purkyně University in Ústí nad Labem, theology at the EMF School of Biblical Studies in England and philosophy at the University of Wales. The phenomenon of education has been his life theme and love ever since he first met J. A. Comenius.)