Finding a New Playing Field: The Failure of Foundationalism, the Benefits of Holism, and the Implications for Christian Scholarship

Religious belief has found itself under fire—and not just recently. Since the project of the Enlightenment began in the mid 17th century, questions about the rationality of religious belief has been steadily creeping into the territory that religion once dominated during the Middle Ages: a totalizing explanation of not just why we are here, but how we came to be here and how the world operates. It has only been in the last one hundred years or so, however, that Christian scholars have really felt a significant burden to defend religious belief as not only rational but absolutely true over against the strong oppositional growth of scientific naturalists and logical positivists who have demanded empirical, testable proof to determine the viability of religious belief. It is in this climate that Christian scholars have turned to methods in biblical hermeneutics such as redaction and historical criticism in order to try and provide an irrefutable, historical foundation for religious belief or they have attempted to use science itself, whether they are engaging in actual science (Hugh Ross, for example), or some form of pseudoscience.

The problem is that it has become increasingly obvious, particularly since the rise of the New Atheists such as Daniel Dennett, Richard Dawkins, and Sam Harris, that religious belief cannot meet the criteria that “science” demands. But does it have to? I place “science” in scare quotes because recent developments in both philosophy and science have demonstrated that the criteria for knowledge set forth in the logical positivist or scientific naturalist worldview of New Atheism is actually based upon flawed assumptions that can be traced all the way back to Descartes and the seeds of the Enlightenment—in other words, their criteria do not necessarily reflect the way that science is actually done today (or has ever been done for that matter) nor the way that our epistemology is actually structured. If this is the case, then why are Christian scholars attempting to beat atheistic dogma at its own game? If the rules are flawed, then why are we playing?

The thrust of this paper will be to explain the foundationalist epistemological project and its failure in the 20th century, to explain holistic epistemology as a viable alternative to foundationalism, and to illuminate the important implications for Christian scholarship. Therefore, my thesis is twofold:  Foundationalist epistemology, though still prevalent in popular thinking, is flawed and should be abandoned for a viable alternative, which I argue is holism; as a result of this shift, Christian scholarship is free to take a more balanced approach to hermeneutics—one which can accommodate the inclusion of literary theory and other methods currently seen as too relativistic—and is no longer required to “prove itself” to science on the basis of rules which science itself has never actually followed.

Descartes’ Case for Foundationalism

The majority of scholars trace the beginnings of the Enlightenment and the beginnings of foundationalism to René Descartes and his [in]famous Meditations on First Philosophy published in 1641. It is here that Descartes sets for himself a project of radical doubt in which he attempts to doubt everything that he previously had held to be true since he recognizes that, over time, his beliefs have changed—things he had held to be true in his youth may have turned out to be false once he had gained new evidence or further developed his reasoning ability. In his course of radical doubt, he sought a foundation, at least one thing he could know for certain was indubitable and universally known by all. Descartes was inspired, as Nancey Murphy points out, by the ways in which buildings were constructed. He created a mental picture that has stuck with modern epistemology ever since.[1] Descartes saw an infinite regress occuring when one attempts to justify a particular belief. The knowledge that justifies it must itself require justification, and on, and on, ad infinitum. He believed that there had to be a rock bottom, a point where it was not necessary to travel any further.

The historical reasons behind his motivations are important to take into account. Murphy writes that since Descartes lived through the horrors produced by the Thirty Years’ War, he recognized that “the epistemologist could render a service to humanity by finding a way to produce [universal] agreement.”[2] Furthermore, there was a great fear on weighing on Descartes: he had come to see the difficultly in reconciling his traditional religious belief with the new data and hypotheses he was encountering in his work as a scientist and mathematician. He was aware of Galileo’s house arrest, and did not want the same fate for himself. We can see his desire to reconcile science and religious belief in his letter “To the Very Sage and Illustrious, the Dean and Doctors of the Sacred Faculty of Theology of Paris” found as a preface to his Meditations. He wanted to make absolutely sure the Church knew his program of doubt was not going to doubt the existence of God—it was, in fact, going to prove His existence. In reality, the radical doubt he used to seek out an indubitable foundation only worked to serve the Enlightenment project of science and had the disastrous consequences of casting religious belief as irrational appeal to authority and tradition.

Foundationalism, Scripture, and Science

Part of these consequences has to do with foundationalism’s deep history with science. Foundationalism, to some scientists, but mostly to the science enthusiast, just seems so obvious that many take it completely for granted. The work of the scientist is simple: He or she observes the world; he or she develops hypotheses; he or she draws conclusions that either support or refute those hypotheses based on more observation. And all this is done, as Descartes, Locke, and many others claimed and have claimed since, without the interference of any “outside” beliefs, particularly something as pervasive as religious belief and authority. Observations and claims made about the world must be contextless because in foundationalism knowledge must be built upon facts about the world that are universal, indubitable, and known a priori.

Following David Hume, there is an important split in foundationalist epistemology that will be fruitful to trace out briefly in order to talk more completely about the inherent flaws in foundationalism and to discuss the implications it has had on Christian scholarship and the interactions in the 20th century between science and religion. The split is between the common sense realism of Thomas Reid and the German idealism of Immanuel Kant. Reid’s foundationalism consists of simply examining the facts that are laid out for every observer to see. Because Hume had made it clear that the only viable way to have knowledge was through direct experience, Reid saw a way to legitimize Christianity through scripture—the only “direct experience” we have of it. Scripture rests as the universal foundation (a major flaw that I will explore in due course), and, as the name of his school of thought suggests, one may rely on common sense to discover truth.

Kant, however, is more concerned with rescuing science from Hume’s critique. Hume was so skeptical of our rational capacities, that he claimed the only definite knowledge one could have a priori is analytic knowledge, i.e. definitions such as “all bachelors are unmarried” or “all triangles have three sides.” The problem is that such statements do not contribute any new knowledge—they are equative statements. If these statements constitute the only indubitable knowledge that we have, science is up against a serious issue because we have nowhere to build from them. Kant however argued that not all a priori knowledge is analytic. Theorems such as the Pythagorean theorem are, in Kant’s view, synthetic a priori knowledge because we do intuit such things but only after we experience their truth. We see the theorem in action, working again and again, but once we’ve seen it a few times, we intuit that all right angled triangles are subject to it. There’s no need to work out the problem on every triangle, since such a thing would be impossible in the first place not only because there is an infinite number but also because precise measurement of irrational numbers is not possible. In making this argument, Kant secured a new foundation for science, one that seemingly retained the universality foundationalism requires.

The Flaws of Foundationalism and the Need for an Alternative

Foundationalsim and its currents, as I have described above, suffer from some serious flaws and are thus subject to a number of critiques.  It will be helpful to break them up into four categories, logical critique, cultural critique, scientific critique, and theological critique, with some interplay and overlap between them. I will begin with the first two, which provide us ample reason to seek out an alternative to foundationalism. Describing what I see as the best alternative, holism, will give us a good opportunity to see how both scientific and theological inquiry actually work and how they also offer an inherent critique of foundationalism

The first critique is a general one and is quite simple: Foundationalism is a self refuting system. The requirements for a suitable foundation set forth by Descartes and accepted by foundationalists are that it be universal, indubitable, and known a priori; however, foundationalism itself meets none of those requirements. The fact that one can sit down and work out a logical, well reasoned case against it makes far too strong a case for it not being universal, indubitable, and a priori. Given this, what good reason do we have to accept foundationalism as a, let alone the only, viable epistemology?

This notion of good reasons is important in the cultural critique of foundationalism. In my brief sketch of the history of foundationalism, we can see that there were some very important cultural motivations behind Descartes’ search for a universal epistemelogical foundation. In other words, it was not without context as he had wanted it to be—indeed, it could not be. We also notice that there are a number of things Descartes did not doubt, such as his capacity for philosophical method once his radical doubt was achieved. Alasdair MacIntyre states the Cartesian problem this way:

[The doubt] is to be contextless doubt. Hence also that tradition of philosophical teaching arises which presupposes that Cartesian doubts can be entertained by anyone at any place or time. But of course someone who really believed that he knew nothing would not even know how to begin on a course of radical doubt; for he would have no conception of what his task might be, of what it would be to settle his doubts and to acquire well-founded beliefs.[3]

Hume falls victim to the same problem. In all of his frustration over not being able to find a suitable foundation based upon experience, he never stopped to ask himself what experience he had for needing a foundation in the first place. He had none. As I mention above, it just seemed so obvious given his training, and the culture in which he was living, that one could question whether or not he, or Descartes for that matter, even had the capacity at all to ask such questions that seem just as obvious to us today as foundationalism seemed for them.

Before we approach the critiques that both the nature of scientific and theological inquiry provide, it is important to note that the logical and cultural critiques do more for us here than simply refute foundationalism. Both point to something that is much closer to the way that our epistemology actually works. The fact that foundationalism is self-refuting must in turn mean that our knowledge is not based upon universal absolutes, and observing how cultural context so strongly affected both Descartes and Hume points to a strong case for including that (along with other factors, as we shall see) as viable justification for our knowledge claims. We are in need of an epistemelogical alternative.


W.V.O. Quine, in his 1951 essay entitled “Two Dogmas of Empericism,” refutes the foundationalists of his day (logical positivists) and provides us with a new image for knowledge, a web or net—a holistic epistemology. In this system, there is no difference between beliefs that seem to us “more foundational” and beliefs that are not this way; rather, beliefs differ in their distance from experience, which marks the boundaries of knowledge.[4] The following diagram will help to explain:

We can imagine each point where lines either end alone or meet together as various beliefs and knowledge claims. The point in the center we could refer to as the most basic of beliefs for someone and also note that a number of other beliefs are connected to (justified by) it. But there are others that have a number of beliefs connected to them as well, and some beliefs are supported by a number of others which are supported still by others. There are a few near experience that are justified directly by perhaps only one other belief. These are the most volatile since new experience can either erase or alter them.

This is one of the primary benefits of holism over foundationalism. William Placher writes: “Quine’s point is that we can always solve the problem in more than one way. Since a solution at one point will always have consequences elsewhere, we cannot decide single questions in isolation.”[5] The points and lines in the web can change due to experience, shifts in reasoning, the disappearance of other points and lines—a whole host of reasons. The points closest to the center are more unlikely to change, but their justification, the lines that connect to them, may change depending on the factors above. In other words, when we encounter something that contradicts a particular belief, whether it be sensory experience or something else, we look for a new way to justify that belief. If we cannot find a new way, then the belief is most likely erased. For instance, imagine a child on Christmas Eve, sneaking downstairs to catch a glimpse of Santa Claus because she heard some rustling under the tree. The moment she sees her parents wrapping presents and placing them under the tree, she is forced to either seek new justification for her belief (i.e. “My parents are helping Santa Claus who had to take care of the reindeer”), or she decides to abandon the belief altogether based upon this new experience.

This system of knowledge is able to account for the two major errors discussed above that foundationalism has encountered. There is no need in this system to justify any belief as indubitable and universal. Beliefs can be justified by experiences, but they can also just as rationally be justified by agreed upon cultural assumptions, history, or any number of reasons. In other words, beliefs rationally justify other beliefs. This must be the case if we can arrive at no indubitable foundation for beliefs. The phenomenon of belief-justified knowledge leads us right into seeing how scientific knowledge is holistic and thus the scientific critique of foundationalism.

Science as Theory-laden

The foundationalist model of science, as we have seen with Hume, assumes that we must build knowledge upon indubitable facts, which Hume argues we can only have from direct experience. However, as Murphy points out, “It is widely recognized that data are not just given [. . .] they are made (‘facts’ comes from facere) by means of their interpretation in light of “ideals of natural order” (Toulmin) or theoretical assumptions (Hanson, Kuhn).”[6] In other words, the way we gather experience does not work in the ways that Hume assumed it did: we do not merely experience data as if it were simply lying around the universe, waiting to be collected and used to compose hypotheses and theories. Rather, science works the other way around. Placher writes: “What we see is shaped by what we already know, by other things we see, and by what we expect. The uninterrupted bare impression proves an illusion.”[7] We cannot even begin to collect data if we do not already know what data we are seeking out. Scientists may observe certain phenomena that are surprising, out of the ordinary, seemingly brand new; however, the only reason scientists are able to see phenomena this way in the first place is because they already have the conceptual framework necessary to recognize the phenomena as novel as well as the “theories of instrumentation” (the physical, scientific tools) needed to collect the data which also come from  a conceptual framework about how optics or thermometers operate.[8]

If we attempt to imagine this structure in terms of foundationalism, as a building, we see that it is rather difficult. We can no longer say that raw data, pure experience as Hume would say, is foundational since what we see as data in the first place is shaped by the theories that, under foundationalism, we assumed were derived from the data. We have a problem of circularity. Imagining how science works in terms of the holistic model, however, solves this problem by allowing us to accept that we do not have an indubitable starting point.

Holism and Self-contained Systems

Finally, we must address the Kantian Synthesis, which houses the last foothold of foundationalism to address. As discussed above, Kant provided the reasoning by which we could say we have synthetic a priori knowledge—knowledge that, observed once, can then be intuited and applied to all cases of that knowledge. These knowledge claims seem foundational because we can intuit them to be universally true. We have seen how this works with the Pythagorean Theorem. Let us now look at a different case, one that raises a problem for the foundationlist: the parallel postulate. In Euclidean geometry, the postulate serves us very well. We can see two parallel lines on a Cartesian plane, note how, mathematically, they will never intersect, and intuit that this must be the case for all parallel lines. However, we know now, thanks to Einstein’s theory of relativity that no straight line parallel to the original line could be drawn through a point not on the line.[9] In other words, space time is curved, and on a cosmic scale, the parallel postulate is actually false. Something we thought to be so basic and foundational does not actually describe the way the universe is as a whole.

However, that does not mean that the postulate is useless. It serves an important purpose within the self-contained system of Euclidian geometry. Indeed, without the parallel postulate being true in some cases, we would not have railroads or solidly built houses. The fact that it is not true universally does not negate its usefulness or necessity in some cases. Placher writes of the postulate that we

cannot judge it in isolation, either by empirical test or by some criterion of intuitive self-evidence. On the face of it, the postulate does seem reasonable, but in order to preserve it, we have to give up something else that also seems reasonable. We can do that, but we then have to face a different set of consequences, and we have to look at connections all through our system.[10]

In other words, even something as seemingly basic as synthetic a priori knowledge is subject to the give and take of the holist system—which is acceptable since this example proves that even mathematical postulates cannot serve as foundations for knowledge either.

Christian Scholarship: Finding a New Playing Field

One of the primary consequences of Enlightenment foundationalism as discussed at the beginning of the paper is that it sets the standard for rational thinking as tossing out all assumptions and beginning at something objective and foundational. This presents a huge problem for Christianity. Let us return briefly to Thomas Reid and the Common Sense School and our final critique of foundationalism as an example. Reid claimed that one could use Scripture as the foundation for knowledge and rather than interpret or guess at what the Bible was saying, one simply had to open one’s eyes in order to see the inherent truth laid out plainly in the text. Given what we have already discussed with science, this should give us pause. If we cannot use something as obviously basic as a geometric postulate as a foundation, how can we use something as culturally rich and contextual as the Bible? The challenge to Christianity on the basis of science has always been that while science is just so obvious and basic, Christianity requires assent to a particular worldview, a particular narrative.

Christian scholarship has then made the unfortunate mistake of attempting to prove, using scientific means, that it is as viable a theory of how the world operates and why as any scientific theory is. However, both the scientist who expects scientific proof for religious belief and the Christian who thinks he or she has found it are operating under a foundationalist epistemology. Science, even though it deals primarily with what we can observe directly, is no more foundational or basic than any religion. It has the same limitations in terms of arriving at absolute knowledge.

Furthermore, Christian scholars must also ask themselves what foundational reasons they have for seeking a foundation upon which to base belief. This means that we must question the motivation behind critical redaction and historical criticism in biblical hermeneutics. Certainly, those methods can be useful given the proper circumstances; however, if the goal is to arrive at the most basic, absolute understanding of the biblical text, then we are essentially falling into the same trap that Reid does. We fail to recognize our own cultural conditions and subjectivity in an attempt to somehow stand from an objective position. In doing this, scholars tend to devalue other hermeneutical methods, such as those found in literary criticism, on the basis that they are not close enough to the actual meaning of the text. But why do we need that, and can we ever know it? The critique I have provided thus far will hopefully tell us that the answer is that we do not and we cannot, and it seems that our motivation behind seeking these things has been driven primarily as a reaction against scientific doubt. Therefore, scholars must leave the field of foundationalism behind.

by Joel Harrison


MacIntyre, Alasdair. “Epistemological Crises, Dramatic Narrative, and the Philosophy of    Science.” The Monist. 60. 1977.

Murphy, Nancey. Anglo-American Postmodernity: Philosophical Perspectives on Science, Religion, and Ethics. Boulder: Westview Press, 1997.

Placher, William. Unapologetic Theology. Louisville: Westminster/John Knox Press, 1989.

[1] Murphy, Nancey, Anglo-American Postmodernity, 9

[2] Murphy, Nancey, Anglo-American Postmodernity, 10.

[3] MacIntyre, Alasdair, “Epistemological Crises, Dramatic Narrative, and the Philosophy of Science”, 59.

[4] Murphy, Nancey, Anglo-American Postmodernity, 27.

[5] Placher, William, Unapologetic Theology, 32.

[6] Murphy, Nancey, Anglo-American Postmodernity, 27.

[7] Placher, William, Unapologetic Theology, 27

[8] Murphy, Nancey, Anglo-American Postmodernity, 27.

[9] Placher, William, Unapologetic Theology, 30.

[10] Ibid., 31


3 thoughts on “Finding a New Playing Field: The Failure of Foundationalism, the Benefits of Holism, and the Implications for Christian Scholarship

  1. […] Finding a New Playing Field: The Failure of Foundationalism, the Benefits of Holism, and the Implica… […]

  2. […] written many, many posts that deal with the sort of complaint that Olson is raising. In the first point, he’s assuming […]

  3. […] that support them shift slightly. I’ve written about this sort of epistemological holism here. If we take that to be the case, then there doesn’t seem to be any good reason to hold […]

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

%d bloggers like this: