Can Epistemology Be Saved?

by Paul Pardi, IBD Faculty and Speaker, Secretary-Treasurer of the Evangelical Philosophical Society and member of the American Philosophical Association and Society of Christian Philosophers

Much has been written in recent years on the topic of faith and reason. In most of these works one will find a clear dichotomy between epistemology in general and religious epistemology. Much of the time one is reading a defense of the rationality of religious knowledge vis-à-vis a developed, broader system of epistemology. In this post-enlightenment age, rarely will one find an epistemology that presents the act of knowledge in a context of pre-established theism (or Christian theism) where theism serves as the ground for all knowledge whatever.

Perhaps rightly so. For in discussions of religious epistemology, it is not knowledge of the world one is seeking to establish (that is usually a given and specific claims to knowledge of objects of the world tend to be used as a clear cases of knowledge) but rather it is claims to knowledge of the objects of religion that are under scrutiny. Most of us have little problem with reason. Rather it is faith we are seeking to bolster and show that it is indeed reasonable. What would be the epistemological advantage of showing reason to be faithful? However, could such an approach be developed where a system of knowledge was developed out of Christian philosophy of the world? Would such an approach solve any problems that plague theories of knowledge (and solve more problems than it creates)? This paper will take up those two questions and look at an attempt to answer them.[1]

As stated above, in more recent discussions of Christian epistemology, reason is set up as the standard against which the claims of theism are judged. In a later chapter in Philosophy of Religion entitled “Faith and Reason”, William Rowe writes, “The central question that has occupied our attention since the first chapter is whether there are rational grounds supporting the basic claims of theistic religions.”[2] Similarly Geivett and Sweetman note, “The question of whether or not it is rational to believe in the existence of God is one of the most important of all human concerns. . . . However, it has proved difficult to decide the issue of the rationality of belief in God, and philosophers have debated it from the beginning of time without producing any clear-cut or decisive solution which has come to be generally accepted.”[3] In chapter one of his Reasonable Faith, William Lane Craig states, “Before presenting a case for Christianity, we must come to grips with some very fundamental questions about the nature and relationship of faith and reason. Exactly how do we know Christianity is true?”[4]

The view towards faith and reason represented by these statements is one of the reasons, I believe, why modern theistic philosophy has seen such a boon in apologetics—the discipline of “defending the faith.”[5] Defending it against what? Against the inroads of modern, post-enlightenment “reason”, upon the modern Weltanschauung. Certainly this is a noble and necessary endeavor.

What options have theistic philosophers taken in developing this defense? There seems to be two approaches that have been traversed in modern history. The first is to accept the enlightenment view of rationality (at least in a broad sense) and try to demonstrate that Christianity is rational on these grounds. This approach falls under the rubric of what many have called the Thomistic approach to apologetics.[6] It also has been characterized under the broader category of natural theology (or natural theologizing if one wishes to see it as applying the principles of a natural theology). The other, under the general aegis of Reformed thought, has sought to broaden the boundaries of what it means to be rational and in so doing demonstrate that analysis of belief in God may actually aid in defining what rationality is.[7]

The latter approach has been viewed by many academics as more auspicious and is gaining a wider and wider hearing among Christian thinkers. Even so, Reformed epistemologists are not trying to redefine the classical view of what is deemed as rational, but are trying to broaden what might be included within that idea. Certainly this is the view of Mark McLeod in his book Rationality and Theistic Belief. He evaluates the epistemologies of Plantinga and Alston characterizing their basic premises essentially as parity theses. By this he means to say that both Plantinga and Alston are attempting to explain the rationality of theistic belief by arguing “that certain beliefs about God are just as rational as beliefs about perceived physical objects.” [8] If McLeod is right, then these Reformed epistemologists see perception (or some phenomenological analogue) as paradigmatic of rational belief against which religious belief can be compared and, finally, to which it can be equated.

Though McLeod rejects these Reformed approaches he does not reject the idea of grounding the rationality of theism via parity with non-religious modes of belief formation. For after thoroughly examining and criticizing Plantinga and Alston, McLeod develops his own “New Parity Thesis” (which is an amalgam of Reidian epistemology and holistic coherentism) and argues that religious belief is justified in the same way as beliefs about persons as unique individuals. What is noteworthy about McLeod’s characterization of Plantinga and Alston, and McLeod’s own position is that the rationality of religious belief is being formulated in terms of some other generally accepted rational belief (or mode of belief formation) and not vice versa. It may not be natural theology pure and simple but it is certainly a not too distant relative.

There is a third option, however, that is available to the theist. This third approach attempts to tear down the enlightenment view of reason altogether and erect in its place an approach to reason that places knowledge of God and His word at its center. This view has come largely out of the Dutch Reformed[9] churches developing from the thought of Abraham Kuyper (though its proponents argue that it had a much earlier origin[10]). This approach has had a much narrower hearing in the Christian academic community due to (I believe) its lack of systemization and its decidedly apologetic approach. It does however represent what I believe to be a largely clandestine view in many conservative evangelical churches and seminaries and represents the most developed view of an approach to a thoroughly Christianized epistemology qua Christian epistemology as I will define it.

Perhaps the question of an exhaustive epistemology based on religious constraints is actually penultimate. The prior question may have to do with the extensibility of religion as a comprehensive philosophy of the world and life. If it is so extensible the idea of a thoroughly religious epistemology isn’t so far off. Opinions vary as to the intellectual viability of such a project. Some see the task being hindered by enlightenment thought and if we can shed our enlightenment fetters, the doors are open for a comprehensive philosophy based in religion (the way it used to be). For instance Evans and Westphal write, “Maybe religious knowledge looks dubious because we have the wrong idea about what it is to know something and how we know what we know. In the last thirty years there has been a marked resurgence of Christian Philosophy, and this suspicion has developed into a full-fledged assault on Enlightenment epistemologies and those philosophies of religion which rests on them.”[11] Even though they appear to be open to the idea of a “new Christian view of the world,” the bifurcation between religious and non-religious epistemology is clearly present throughout the papers in their book.

In evaluating the Reformed answer to the epistemological question, Peterson, Hasker, Reichenbach, and Basinger draw some observations about the viability of what they call a “common philosophical rationality” based on religion. “It does seem that in deciding [whether religious experience grounds religious belief], an important role is played by one’s willingness (or unwillingness, as the case may be) to give up the goal of a ‘universal method’ for reaching philosophical agreement and to accept that the different starting points and belief-commitments of philosophers may give rise to disagreements that are not decidable by means of a common philosophical rationality.”[12] They go on to conclude that the prospects for a religious epistemology whose application goes beyond the objects of religion alone are probably rather nil.

It appears that religious epistemologies are developed in order to vouchsafe specific religious truth claims. However it seems to me that the development of broader epistemologies that begin with Christianity as a philosophical foundation is motivated by other concerns as well. As mentioned above, a Christianized epistemology may serve as an heuristic for general epistemological problems vouchsafing not only religious beliefs but any beliefs whatever. Further, there may be theological motivations . One may, for instance, wish to preserve the absolute sovereignty of God by completely removing man from the act of knowing. [13] This might best be done in an epistemological system that is driven either by God’s nature or His will.

There is an important distinction between evaluating knowledge claims (which would be normative epistemology) and evaluating systems that evaluate knowledge claims (what we might call meta-epistemology). This paper is engaging in the latter type of evaluation and will seek to determine the viability of a system of epistemology. In order to evaluate whether or not a Christianized epistemology solves any of the problems that plague attempts to develop a comprehensive theory of knowledge, the problems themselves must be defined. What an adequate epistemology must do is a difficult question in and of itself.[14]

What would a Christian epistemology look like? Obviously such an approach would have to be theistic. It would have to explain both events of belief formation and doxastic justification in terms of an ontology of knowledge that includes an explanation of the noetic relation between mind and world in terms of the creative act of God. Also it would have to include Christian revelation within the scheme. The latter is important for a Christian epistemology, as it is revelation that distinguishes a general theism from Christian theism.

Specifically, I believe a fully christianized epistemology might have the following features. (1) It would explicate the justification (on the standard JTB account) conditions for all knowledge whatever. That is, its range of discussion will not be limited to religious knowledge alone.[15] (2) The justification conditions for knowledge would include an act of God that causes a given belief to come to be known. I will not specify at this point whether the causal relation needs to be efficient or final. However a fundamental tenet of Christianity is that God is not creator only but sustainer as well.[16] (3) Christian revelation somehow would serve as a standard by which the truth-value for all belief is measured. (4) Full or complete justification of beliefs (knowledge) would be attainable only by those who stand in some personal relationship with God.[17]

These features are derived in part from the Christian belief that not only did God create man,[18] but He did so with the capacity to know Him. That is, man was created with a specific teleology: to enter into a personal relationship with God. Further, such a capacity, when not exercised leaves man incomplete.[19] A Christian epistemology then might include this noetic capacity insofar as it seeks to explain the conditions under which knowledge is possible.

This precisely is the project of Cornelius Van Til. Instead of asking what one can know about religious matters based upon a non-religious view of knowledge, he asks what one can know of the world based on certain “religious” truths. The rest of this paper will examine Van Til’s approach to this question. One may ask, “Why Van Til?” I have chosen Van Til not because I believe his position to be the most clearly defined and presented, nor even because I hold him to be the most adequate representative of this particular position (in fact I lean the other way on both accounts). I have chosen to examine his view for the following three reasons.

First Van Til’s epistemology represents what I believe to be the approach taken by many in (especially) conservative Christian churches at least in the west. The Christian scripture is the epistemological starting point of faith for many believers though most churchgoers do not usually understand the importance of that. Second, though Van Til’s writings are not popular outside of certain Reformed circles, he has had considerable influence on the evangelical church at large through his students, most notably Francis Schaeffer and Greg Bahnsen. Third, Van Til, has done specific work in the area of epistemology relating to theistic belief as an epistemologist and more importantly believed that the appeal to scripture was not only important for religious knowledge but for all knowledge. This relates directly to the project at hand.

Essential Van Til

My analysis of Van Til will be an attempt to draw out key characteristics of his view of knowledge as it might apply to a thoroughly Christian philosophy of the world. I cannot, in this short paper, hope to answer objections to his position (nor would I be qualified to do so). I can, however, attempt to build a framework that is representative of the key ideas of Van Til. My hope is that in constructing this framework I will be to provide enough insight into this view such that its viability (and potential for further development) can be fully appreciated.

As a starting point, I will lay out what I see as essential ideas in Van Til’s approach. Let F stand for any fact whatever. Let B stand for the self-attesting Christian revelation or Scripture. If we assert also that for any F,

(1) F sustains a relation R to B such that B interprets F

then S knows F just in case

(2) S believes F

(3) S presupposes God’s existence

(4) S stands in some relation R*­ to R such that R completes F for S.

These four propositions capture the essence, I think, of Van Til’s epistemology. As mentioned earlier, his approach uses the language of metaphysics but he attempts to draw epistemological conclusions. Can the system based on these propositions form an adequate basis for understanding what knowledge is? To answer this we will have to carefully examine each of the above propositions then look at how they relate to one another to form Van Til’s approach to Christian epistemology.

Before examining the propositions, a word must be said about how this analysis relates to what is going on in epistemology as a discipline. Much of contemporary epistemology is concerned with addressing the ambiguity that exists regarding the third condition of the standard, tripartite view of knowledge. Namely, epistemologists are seeking to provide insight into that feature that turns true belief into knowledge. Historically, the study has concentrated on the justification condition though of late some epistemologists have replace justification with some other condition. In any case, examining that feature (whatever it is) will be important in our study of Van Til. More specifically, we will have to surface from Van Til’s view that feature (or those features) that justifies or warrants Christian belief.[20]

There are a couple of features that can be surfaced immediately about the set of propositions I have laid out. The first is that knowledge on this view is structured in terms of entities and relations that obtain between them. That is, when certain relations obtain between a person and some object, the person knows that object. Second this system relies heavily on Christian revelation. This is important to notice for our present purpose, as it is the Bible (insofar as it the logos of Christ) that defines Christianity qua Christianity. If Van Til’s system did not include the Bible in a fundamental way, it would be questionable whether or not it could be called a Christian epistemology. Third the set attempts to define all knowledge, not just “religious knowledge.” In fact, Van Til held that knowledge of anything in the world had to be understood in terms of Christian theism. For him, the claims of Christianity were “truer” than anything else one could know and so in some sense grounds all knowledge. Because of this, knowledge of the world could be defined only in terms of Christian theism. The following will examine the set of propositions in light of these considerations.


Van Til’s system has been called presuppositional. If that label is accurate (and I believe it is) one would expect presuppositions to be at the core of his system and (3) satisfies this expectation. Van Til’s use and meaning of the term “presupposition” is somewhat varied.[21] Essentially, the notion of presuppositions appear to include both a cognitive element—an entity of some sort which is the direct object[22] of some noetic function—and a relational element.[23] I think, then, presuppositions can best be understood in terms of objects and relations. Each of these must be taken in turn.

Taking the former idea first, the objects of knowledge for Van Til are facts. Relating this to the concept of presuppositions, Van Til develops the idea that there is a fact (or are facts) upon which all other facts depend for their existence. For Van Til this has ontological as well as epistemological implications. Ontologically this means that certain facts cannot exist independently but are dependent upon other facts for their existence (ostensibly this would include their coming to be and their continuing to be). More to the point, Van Til believes that no fact can exist unless the fact stands in an ontological dependency relation to a single fact: the fact that God exists (FG).[24] This latter fact however does not stand in a dependency relation to anything else. To use modal language, God’s existence is a necessary fact without which any other fact could exist. This ontological map is implied (if not explicitly stated) in Van Til’s use of presuppositions.[25]


To understand this better, we must digress and look briefly at how the notion of facts function within Van Til’s epistemological system. Since facts are the objects of knowledge then when a subject knows (or believes) a thing it is a fact that he knows. Although a systematic presentation of the nature of facts is hard to find in Van Til, they appear to have the following features: (a) they are extra-mental entities, (b) they are not propositional in nature,[26] (c) they can be apprehended by the mind, (d) they must be interpreted to be known, (e) facts, in some sense, just are objects and states of affairs (O/SA) and are not to be distinguished from them.[27] Essentially facts appear to be the objects and states of affairs that make up the world. All physical objects are facts as are states of affairs and metaphysical objects.

This description, however, does not really explain matters. The problem arises in the way Van Til employs the use of facts in his system. Facts appear to have properties that would differentiate them from objects and states of affairs (and vice versa). For example, (d) indicates that facts are apprehended by the mind meaning that facts are the objects of comprehension. But certainly when one comprehends the existence of a tree, one does not have a physical tree in one’s mind (imagine the headache that would cause!). Yet Van Til wants to avoid saying that a fact is a proposition or some other entity of that sort. Given this, the best I can say is that facts appear to be some property or properties of objects and states of affairs that relate to the mind in some way. Thus facts are not separated ontologically from these entities yet they do not necessarily encompass every property of these entities. This ultimately may be an inaccurate rendering of Van Til’s idea but will have to suffice as a working definition for our purposes.

To round out this ontological picuture, Van Til holds that God created man with the capacity to know the world around him and he created the world with properties such that man’s noetic structure is able to cognitively “fit” with those properties.[28] He writes, “God has created the human mind. In this human mind God has laid the laws of thought according to which it is to operate. In the facts of science God has laid the laws of being according to which they function. In other words the impress of God’s plan is upon his whole creation.”[29] God’s design of both the human mind and the world allows for the possibility of knowledge. Further, God must aid in interpreting what man apprehends through this mind-world interaction. “If the Christian position with respect to creation, that is, with respect to the idea of both the subject and the object of human knowledge is true, there is and must be objective knowledge. In that case the world of objects was made in order that the subject of knowledge, namely man, should interpret it under God.”[30]

Facts and Interpretation

This statement by Van Til brings to the surface another very important aspect about facts: facts must be interpreted to be known. This is the relational side of presuppositions. That is, facts are not “brute” but stand in relations to other facts and stand in ultimate relations to the plan of God. If one attempts to know facts apart from these relations, she is attempting to do the impossible.[31] There are no non-interpreted facts of science for example. Every thing that science discovers as a fact relates in some way both to the other facts of the world and to the plan of God. To know these facts truly, these relations must be a part of the epistemic process.

To reiterate, the human mind has functions and properties that allow man to know in this way. For Van Til, this cognitive fit is part and parcel of the human design. With this, Van Til’s approach avoids (at least on the surface) much of the sticky epistemological problems that plague many systems (such as objectively validating the correspondence relation between thought and world). If this relation is built into the cognitive structure of humans and God created the world with properties that immediately “interface” with that structure, the need for sense datum or idealist theories is done away. With this background, (3) can be modified as follows,

(3¢) S presupposes God’s existence when S stands in some relation R* to FG.

A presupposition then includes a fact (I might add of a certain class) and the relation one sustains to that fact. As Frame points out, a helpful analog to this idea is the notion of a priori knowledge where the latter consists of knowledge that is not experientially based. Van Til’s notion of presuppositions takes this idea much further. For him, presuppositions are necessary for any sort of knowledge whatever. Beyond this, the relation between subject and this presupposition does in fact obtain for all persons.[32] Without this relation no one could know[33] anything. Again the presupposition to which all subjects must necessarily be related is the fact of God’s existence. We now must explore this relation a bit further.

The Presupposition Relation

The first condition for knowledge, then, is that the subject presupposes the existence of God in terms of the relations described above. However, the epistemological significance of this relation has not been addressed. It seems at least prima facie false to say, for instance, that a person cannot know that Mount Rainier exists unless that person first believes the presupposition that God exists. Further, it is not at all clear what kind of a relation R* is. In order to address these problems, it will be necessary to introduce some key terms and distinctions that Van Til employs.

Denotation and Connotation

Van Til makes a critical distinction between the denotation and the connotation of a fact. Crudely, the denotation of a fact is its “thatness” and the connotation of a fact is its “whatness.”[34] These two terms specify the range of the relation that obtains between fact and subject. More specifically they designate the degree to which a subject is related to the object of knowledge. The denotation of a fact is, simply, the finite existence of that fact as it is grasped by the subject. What is denoted by a fact is nothing more than the prima facie properties and relations that can be predicated of the fact itself. It may be accurate to say that the denotation of a fact is those features that can be pre-reflectively apprehended of it. To say that the cat is on the mat is to say nothing more than that a cat exists, a mat exists and that the cat stands in an on top of relation to the mat. However there are much more to facts than mere denotation. Van Til brings out these further features in his idea of connotation.

The connotation of a fact is all that is entailed by a comprehensive predication of that fact. This involves understanding all the facts that can be predicated of that initial fact. The cat, for instance, not only exists, but it is a living thing, it has x number of physical properties, it is of such and such a breed, it didn’t come into existence yesterday, etc. There may be innumerable other facts that can be predicated of any given fact. The point is that what is connoted by a fact includes every other fact that can be predicated of that fact.

In employing the concept of connotation, Van Til seems to emphasize the idea of implication not fullness. What I mean by this is that Van Til was not trying to say that one had to have exhaustive knowledge of some fact before he could know that fact but that one had to understand that all facts imply an ultimate ground for their existence. I believe, then, there is an important distinction to be made between what Van Til meant by connotation and how he actually used the term in his system. The former is much richer yet the latter more easily employed from a practical point of view.

With this distinction in mind can we provide an answer to the question of the type and nature of R*? The answer does not come readily. If the relation is epistemic, (that is, if the person must know this presupposition in the connotative sense in order to know any other thing) we are involved in a vicious circle.[35] For how does one come to know F­G without first being in this epistemic relation? If the relation is not epistemic, how does it do any work in so far as it is functioning as a means by which F­­G enters into the noetic structure of a person as an item of knowledge? A fuller answer to this will have to be deferred until later. At this point, it may be sufficient to say that this relation is epistemic but in a special sort of way.

It seems that for Van Til there is a special subclass of beliefs that are justified on the basis of the means by which they were formed.[36] The closest analog may be the notion of basic beliefs (however the pool of what might be considered basic for Van Til is much smaller than that of most later Reformed thinkers). Provisionally I will use the term awareness[37] to refer this idea. Greater detail on how this awareness functions epistemically will be explicated below.

With this background we can return to our initial proposition and expand upon the idea of the relation that obtains between S and FG (the fundamental presupposition). S presupposes God’s existence just in case,

(5) ­God created S with an awareness of Himself.
(6) The awareness S has of God is validated by B.

What Van Til is attempting to do is bring his entire theology and philosophy into the knowledge event by collapsing ontology and epistemology into a unified whole. A key idea in this attempt is that of the nature of man as a being created by God. For only if God created man could he stand in a direct epistemic relation to the fact of God’s existence.

As mentioned above, Van Til emphasizes the unity of man within the “epistemically compatible” environment (i.e., the natural world) within which God created him. “God has created the human mind. In this human mind God has laid the laws of thought according to which it is to operate.”[38] The former idea is not fully expressed in (5) however. (5) would have to provide insight into what the awareness of God entails as well as how God’s creation of S relates to this awareness.

(5¢) God created S with a noetic faculty that is immediately aware of God’s existence such that S pre-reflectively believes the fact “God exists” is true.

With (5¢) not only is God designing the noetic faculty of man to be able to receive a belief about His existence, He somehow is causing man to have the very belief itself. Notice also that this belief is not limited to believers. This is not a belief that is somehow “inserted” at the time of salvation. By including the phrase “pre-reflective” I intend to bring in the idea that this assent to God’s existence is present in the noetic structure of all persons. Thus atheism would be learned and theism would be the normal, “default” position epistemically.

(5¢) is too strong however. It is not the case that everyone who reflects on the question, “Does God exist?” answers it positively and Van Til did not want to argue for that.[39] However what is true, Van Til argued, is that no one can consistently deny that God exists and affirm any other fact. Any attempt to do so leaves one with epistemic gaps or holes that cannot be filled with any other fact except the fact that God exists.


Here is where Van Til’s usage of denotation and connotation come to full play. By using these terms as he does, he indicates that knowledge is, in some sense, degreed. One can know fully or partially (one recalls Paul’s statement in 1 Corinthians 13:12). How connotation functions in this context is not to explain how one might know any given fact but how one is aware that God exists. That is, knowledge of God’s existence is directly related to one’s ability to fully know any given fact. (I will use the term intension for this property and will explain intension more fully below.) (5¢) can be modified to account for this lack of fullness knowledge of a fact can have.

(5¢¢) God created S with a noetic faculty that is immediately aware of God’s existence such that S’s belief in F is underintended for S when not based upon FG.

(5¢¢) locates S’s awareness of Go