This paper aims at setting forth a perspective on religious epistemology. As will hopefully become clear through the course of this essay, I understand accurate discussion of issues pertaining to religious epistemology, or more specifically of an epistemology of belief, as necessarily grounded in general epistemology. Thus this essay will begin with a discussion of general epistemology in order to set a foundation for its discussion of belief.
It may be the case that the need for such a dependency is obvious and uninsightful. Highlighting this relation between belief ideas and knowledge ideas does, however, set an initial boundary as to the scope of this paper by suggesting a central argument (namely, that beliefs are grounded in knowledge) which itself will require the support of careful discussions of general and religious epistemologies. It will also become evident that based on the epistemology outlined here, certain theories of general and religious epistemology are precluded. By the conclusion of this essay I hope to have laid out a very specific position on these matters which is both consistent and arguably plausible.
existential and abstract ideas
One should probably begin this discussion of epistemology with an explanation of what is meant in the use of the term "idea". When such an explanation is attempted, however, an immediate need for a distinction presents itself, a distinction between what we will call "ideation" and the idea itself. By ideation I mean the process through which the particular idea is derived. I will turn to ideation in more detail shortly. For purposes of this discussion, by idea I mean to emphasize the notion of content, that cognitive representation of what exists outside the mind. This narrow definition of ideas as those representations of things existing outside the mind would seem to neglect entire categories of "ideas" which we realize do not correspond to reality or at least to actual states of affairs. Such categories would include the fantastic (ideas of unicorns and centaurs, for example), the hypothetical, remembrances, and possible others. However, this narrow definition of idea, as we will see, will provide a very adequate means of accounting for these secondary categories.
More precisely, I would like to distinguish between what we may call existential and abstract ideas. Existential ideas are those which occur (through ideation) from encounters with actual particulars, things which exist, and thus such ideas are consistent with the narrow definition suggested above. Historically, this category of ideas has gone by the names simple (John Locke), intuitive (William Occam), atomistic (Wittgenstein), and others. I prefer the name existential to these others since it provides at least initial reference to what the idea (representationally) consists of and from what it is derived, namely, existents. The existential idea, in addition to containing the collection of perceptions of the existent's various attributes, also carries with it the indelible conviction that 'this exists'. Whether such existence belongs properly to the existent's attributes I will not here discuss. It would seem, however, that one's conviction of the existence of an object does not take place in the same manner one's conviction, for example, that the object is red, since we have no perceptual organ specifically designed to perceive existence. While it could be argued based on this difference that one's notion of the object's existence might more properly fall within the category of abstract ideas (for reasons which will become clear), and thus the term 'existential idea' used in the manner I intend is from the offset problematic, at present I maintain my preference for the term for its explanatory value.
One inevitably encounters numerous existential ideas throughout the course of each day, as a myriad of objects are confronted and perceived. Each subject, then, has a wealth of such existential ideas through his or her life experience. These core ideas serve as the basis for rationation wherein the rational capacity of the subject is able to manipulate or further investigate the content of the existential ideas. Those subsequent ideas resulting from such rationation I designate abstract ideas. This designation points to the fact that the content of such ideas ground in abstractions of the more basic existential ideas. Abstraction itself is not a single process, but is possible through several operations of the mind. John Locke (Essay, I.x-xi) lists the following faculties of the mind: retention, memory, discerning, distinguishing (as to clarity and determinateness), comparing, compounding, naming, and abstraction ("whereby ideas taken from particular beings become general representatives for all of the same kind"). All of these functions of rationation involve a manipulation of or extrapolation from existential or simple ideas.
Such abstraction results in the more commonly recognized variety of ideas alluded to earlier. Using Locke's terms, we see that ideas of unicorns, centaurs, and other fantastic things result from the compounding or combination of existential ideas. Remembrances or memories result from the mind' ability to retain and recall such existential ideas. And, as we will see, hypothetical ideas are possible through our comparison of existential ideas and experiences.
I have categorized ideas into two broad categories, existential and abstract. Existential ideas derive through encounter with existents and serve as the foundation for all other types of ideas. The abstract category contains these other types of ideas, all of which share the characteristic of deriving from a form of rationation of existential ideas. This categorization has left some important questions unanswered, many of which we cannot address in this paper. For example, how existential ideas actually emerge from such encounters is one such question. Here one's stance on empiricism, a priori ideas, and universals would become quite clear. A related question is how one might be sure his or her existential ideas actually correspond to actual states of affairs. In answering this second question a precise definition of knowledge would emerge and the problem of skepticism would be dealt with. Other questions which will be addressed involve notions my system apparently precludes or would seem to given fuller exposition, notions such as a priori knowledge or truths of reason. Before such answers are attempted, though, we must briefly turn to the topic of belief.
knowledge and belief
Given the system outlined above we must find a place for belief. Although I will not suggest a strict definition of knowledge, a functional definition will be necessary if we are to talk also about belief. In this discussion, knowledge will be attributed to only those ideas which correspond to actual states of affairs in the sense outlined above. This implies that knowledge can only be had of actual existents and states of affairs, and that existential ideas, then, serve as the ground and criteria for claims to knowledge. Whereas abstract ideas to varying degrees relate to existential ideas, they may also be said to convey knowledge. But it is clear that the level of knowledge conveyed through abstract ideas is different from that conveyed through existential ideas. The knowledge I now have of my sitting at my desk is of a different sort than my 'knowledge' that I was sitting here an hour ago or my 'knowledge' that the sun will be rising several hours from now. We would say that the knowledge of my sitting here could reliably be a claim of knowledge based on fact that I am indeed sitting here. (Here, we again intentionally skip the discussion of how I in fact know this.) My claim to know I was sitting here an hour ago, however, is no longer based on fact in the same sense, but on my remembrance of a previous state of affairs. My knowledge of the sun rising in a few hours does not rest on present or past fact in the same manner as the previous two instances, but on a reasonable assumption that the sun will appear tomorrow morning in the same way it has throughout my previous experience. Although numerous other examples are possible, these three should suffice to make the intended point.
Strictly speaking, I would want to attribute knowledge to the first instance, that of my current sitting, but not to the last instance, that of the rising sun. It would be better for the sake of our discussion to say I believe the sun will rise in a few hours. To say that this is necessarily a belief in no way speaks as to whether or not it is reasonable for me to believe it. It does mean, however, that the truth or reality of my claim will only be demonstrated should it come to pass. As for my claim to know past states of affairs, I must here confess I have not yet been able to make a determination as to whether this is more properly a knowledge claim or a belief. This determination ultimately rests upon how one distinguishes between degrees of knowledge or between degrees in reasonability of beliefs, or at what point knowledge stops and belief begins. I am wont to also consider memory a belief, with the understanding that such recollection naturally provides very reliable grounds for such belief. Such reliability, however, would be supported through a more thorough development of perception and operations of the mind which we will not here address.
Beliefs are not simply to be understood as any idea without the attribution of knowledge, for we are clearly free to think whatever we want irrespective of actual states of affairs or the conviction that what we think corresponds to reality. Beliefs would be limited to ideas which are other than existential ideas yet which we deem correspondent to actual states of affair. The very fact that we 'deem' certain ideas thus results in their status as beliefs. Based on an increasing number of existential ideas it is possible that current belief ideas may later be recognized as non-correspondent while current abstract ideas which are not beliefs may in time be deemed correspondent. As with all other abstract ideas, belief ideas find their ground in existential ideas. Although one is free to believe anything one can conceive, the reasonableness of one's belief is said to be determined through the degree to which it could plausibly adhere to actual states of affairs.
a posteriori and a priori knowledge
Given the system outlined thus far certain significant problems are apparent. How, for example, is this primary status of existential ideas to be understood? The discussion at this point lends itself equally well to empiricism, positivism and materialism. And if this be the case, what possibility does this create for a religious epistemology whereby the belief in (a non-material, non-empirical) God might be justified? In order to answer such questions, we must turn in greater detail to a discussion of a posteriori and a priori knowledge.
Traditionally the distinction made between a posteriori and a priori knowledge has been one of qualitative difference. Representative of this recognition of a qualitative difference is Leibniz's statement that, "There are also two kinds of truths: those of reasoning and those of fact. The truths of reasoning are necessary, and their opposite is impossible. Those of fact, however, are contingent, and their opposite is possible" (Quoted in Chisholm, 26). Similarly, Roderick Chisholm offers the general definition of a priori truth as those "propositions that are necessarily true and such that, once one understands them, one sees that they are true" (Ibid). A set of such propositions is the axiom which is traditionally taken to be "incapable of proof"; that is, there is no proposition which is better known. Such an understanding of axiom naturally implies then that the axiomatic proposition is "better known" to us than any demonstration thereof. This, for example, is the basis of Thomas' claim that "those who have knowledge of the principles [i.e., the axioms] have a more certain knowledge than the knowledge which is through demonstration" (Posterior Analytics, quoted in Ibid, 29). Upon such axioms a greater body of a priori knowledge is then established.
This understanding of a qualitative difference between a priori and a posteriori knowledge seems incompatible with the system I have suggested above for two reasons. First, the traditional view claims a complete independence of a priori knowledge from actual states of affairs, the latter of which Leibniz refers to as "truths of fact" or Thomas as "demonstrations". I have, however, suggested only one source of knowledge, namely the existential idea resulting from one's encounter with actual states of affairs. A second ground of incompatibility stems from the traditionals' claim that knowledge of the axiomatic is superior in quality to knowledge of the actual or particular, or what I have defined as existential ideas. I, on the other hand suggested that any claim of knowledge must be evaluated solely in terms of actual states of affair and thus must involve existential ideas.
It is clear that the axioms of which the traditional view of a priori knowledge speaks are meaningful and truly adhere in the real world. The law of non-contradiction and mathematics, for example, do seem to exist independently of this or that particular object. How then might my system account for such axioms and their 'independence' from actual states of affairs?
I would argue that rather than an independent status, such axioms are more primarily descriptive of actual states of affairs and to the degree our reality operates uniformly, they may be said to be prescriptive. Thus we may say axioms describe relationships which will always adhere among particulars within our reality. To state, as the traditional view has, that such axioms are "truths of reason" would seem to imply that such axioms are laws of reason or of thought and as such do not depend upon external fact. But as Russell rightly points out, belief in, for example, the law of contradiction is to be distinguished from the law of contradiction itself. On the distinction, he writes,
- "The belief in the law of contradiction is a belief about things, not only about thoughts. It is not, e.g., the belief that if we think a certain tree is a beech, we cannot at the same time think it is not a beech; it is the belief that if the tree is a beech, it cannot at the same time be not a beech. Thus the law of contradiction is about things, and not merely about thoughts; and although belief in the law of contradiction is a thought, the law of contradiction itself is not a thought, but a fact concerning the things in the world" (The Problems of Philosophy, 89).
This definition of a priori knowledge as consisting of relations which will always adhere among particulars within our reality, and the distinction between the fact of such relations and our knowledge and belief in them are quite consistent with what I have thus far suggested. They also allow for some interesting observations. First, it grounds the axiomatic in relations among particulars rather than in reason. I have already suggested that our core ideas are those which derive from encounters with particulars and that all other ideas are based on extrapolations thereof. We had also suggested that this would prove problematic (for our purposes) unless we could somehow move beyond simple positivism or materialism. I believe that the solution to this potential problem lies precisely in the emphasis upon relations we have now made. Such relations are unobjectifiable and thus lie outside the purview of strict empiricism or positivism. David Hume promptly pointed this out in his claim that the cause-effect relation, the centerpiece of the empirical method was in fact unknowable and empirically unverifiable. Without adopting Hume's skepticism, we can agree with him that such a relation is unknowable given our definition of knowledge as stemming from existential ideas and given the immaterialness of relations. Yet despite its being unobjectifiable, we rightly believe that the law of cause and effect is a relation which will always adhere among (in this case, certain) particulars, and is present wherever particulars exist. Similar unobjectifiable yet reliable relations include among others those between a subject and its attributes (predication), a subject and itself (non-contradiction or identity), multiple subjects (cause-effect), and between subjects and propositions.
We expect, based on the recognition of the consistency with which our reality demonstrates these relations, that these relations will adhere among potential, future and hypothetical particulars. Such "universal" application of these relations accounts for their being "a priori". They are a priori truths in that they adhere regardless of the presence of any thinking subject. Our knowledge of these truths is not, however, a priori if by this is implied an independence of fact or states of affairs. Such knowledge can be said to be a priori only in the sense that substantial justification exists for our belief that such relations will universally adhere, even when the particular relation does not belong to actual states of reality (e.g., regarding future affairs or hypothetical affairs). In this sense, our accurate knowledge of such relations could be said to be prior to the particular demonstration thereof.
truths of reason
Before we move to the problem of whether the empirically based system I have set forth can adequately provide a means of justifying belief in God, we must take a further look at the relationship between the axiomatic and reason itself. We might recall Thomas' claim that knowledge of the axiom is a more certain knowledge than that which comes through demonstration. In response to this I agreed with Russell that the certainty in our knowledge of the axiom is grounded in the axiom's correspondence to reality, and suggested such a priori truths are simply (though universally) descriptive of relations that adhere in actual states of affairs.
The relegation of axiom to description of relations, however, raises the very important issue of the status of reason itself. More precisely, we are now faced with the question of what reason itself is capable of telling us. As evidenced by Thomas' statement, it has traditionally been thought that reason itself provides a source of knowledge qualitatively superior to that produced through interaction with external reality. It has also been traditionally argued that such truths of reason provide the clearest (humanly possible) window into the divine nature. It is clear then, that my suggestion that such truths of reason and a priori knowledge are in fact the mind's recognition of unobjectified relations among existing particulars, will preclude much of the traditional argument involving belief in God. For this reason, we must answer the question of what reason is capable of telling us.
It is not the case that the axiomatic are "truths of reason" such that the mind is naturally compelled to adhere to them. It remains, however, that such axioms serve as a standard whereby thought or reason itself is rightly deemed truthful or erroneous. For example, logic consists of such axioms and the inferences drawn from them. Thus the role of such axioms or truths of reason is quite unique in that they are not subject to erroneous thinking in the same way other truths are. One might disbelieve or dispute their universal application, but to do so, we might say, seems unreasonable. It is precisely for this reason that these axioms have been traditionally deemed "basic truths".
How is it that these truths operate in such a distinct manner within reason? Several answers to this question are possible depending on how one accounts for the origin of the axioms. Some view such truths as belonging to the very nature of God such that God himself and all he creates necessarily adhere to them. Whereas all of God's dealings with humanity demonstrate adherence to such truths, the conclusion is possible that they stem from God's nature. It is clear that if this understanding is correct, Thomas' claim would be quite reasonable, since knowledge of the axiom is then knowledge of an aspect of God's nature and as such is easily superior to knowledge derived from material objects. This understanding would also support the very clear distinction traditional views have maintained between matters of fact and reason. Given the location of the axiom within the divine nature, and given the belief that rationality is an aspect of the imago dei, the relation between axiom and reason is made independent of actual states of affairs.
There are however several difficulties with this view. First, such a dependence of the axiom upon the divine nature is impossible to demonstrate. Appeal to the manner in which God deals with humanity as evidence for this relationship, as we will see, does not suffice. A second area of difficulty is that definitions of God's nature or aspects thereof are attempted in terms of natural states of affairs. God's logic, for example is said to be the same as our logic, or God's rationality the same as ours. This also is undemonstrable and assumes a very lofty status for aspects of human reality. A third and related difficulty, as I see it, is that this view ultimately subjects the divine nature to the same consistency we see in the created order such that God is said to necessarily abide by all such axioms. God, for example, must now abide by mathematics and geometry.
The position I have suggested above, however, is not completely incompatible with this understanding of truths of reason, for it does not preclude the possibility that such truths stem from aspects of the divine nature. My position differs, however, by subordinating the status of knowledge of such truth to knowledge of actual states of affairs. As we have seen, although knowledge of such relations adhere in the existential idea, our belief that such relations are universal principles involves an abstraction and as such is one step removed from what we have strictly deemed knowledge. However, if one allows that the natural relations among particulars from which we derive our understanding of such truths of reason themselves reflect the divine nature, then it would still be plausible to view such truths of reason as a reflection of the divine nature.
An alternative to this explanation and one I am more inclined to accept is that such axioms are inherently embedded within the created (natural) order as intentionally designed by God, but do not necessarily stem from the divine nature. This view understands such axioms as functionally necessary given the context of created reality. For example, it is evident that our material, three-dimensional bodies necessarily adhere to certain axioms such as the law of contradiction. For us, reality could not conceivable act in any other manner. This inability to conceive otherwise, however, does not preclude God from acting in another manner, in the same way our inability to conceive a five dimensional object does not preclude the possibility of its existence. From this is easily derived the type of dependence of "a priori" knowledge upon actual states of affairs I have proposed in this paper. It also accounts for the unique role such axioms play in regard to reason itself, by relating the axioms to the actual context, the state of affairs within which both reality and reasoning take place.
The difficulty with this position is, of course, that it does not lend itself to statements regarding the nature of God or his thoughts. It does not necessitate God's adherence to the axioms he employed in the construction of our reality. Instances of God's adherence to such axioms as found, for example, in God's communicated affirmations and negations in the form of human language would be explained in terms of the necessary conditions for human understanding rather than the necessary conditions for divine communication. As to whether or not God always abides by such axioms remains unanswerable according to this view. Here also, the proposed discrepancy between the axiom and God's nature is undemonstrable, just as the traditionalists' claim also proves unverifiable.
As to what the precisely the relationship between the axiom and God might be, two immediate options are apparent. One might suggest that God's nature is qualitatively different such that the law of contradiction does not apply in any way to God outside of his voluntary use of it in communicating to humanity. This view will either take the form of volutarism which sees God as willfully and without compulsion submitting to such axioms as regards relations with humanity, or possibilism which does not expect God to deal consistently with humanity. These views, though theoretically possible, open pandora's box in terms of theological discussion and religious belief. Voluntarism lends itself to statements regarding God in terms of his relation to humanity but not in terms of his divine nature, since we could not be sure which action if any by God stems from anything other than his voluntary behavior. Possibilism would preclude any statement from being made regarding God's relation or divine nature, since no consistency is expected from which a system of theology might be established.
A third option in explaining the relation between the axiom and God is that which suggests a quantitative difference such that God's nature is understood to operate at an infinitely higher capacity than the axioms require. Those axioms which constitute our "basic truths" would, in this view, be circumscribable by God in some way. This would be the moderate view between the traditionalists' and the voluntarist/possibilists' in that it posits a real relation between God and his design while at the same time understands the divine nature as unrestricted by our "basic truths". We may say an aspect of God is reflected in the order of creation without the need for understanding God's nature in terms of our reality. This view would place reason, axiom and natural order on a fairly level playing field, understanding each to be related to the others. It is perhaps this view and its interdependence of reason, epistemology and natural context that is most consistent with the position I have attempted to outline here.
belief in God
What, then, is left to be said regarding belief in God? By this point in the discussion certain parameters clearly present themselves as we attempt to deal with the topic of religious belief. First, we expect the basis of accurate belief will be found within existential ideas. That is to say, the criteria by which a belief is deemed plausible, justifiable or otherwise, will consist in a correspondence with actual states of affairs, and such a correspondence will be known by us only through existential ideas and extrapolations thereof. Second, we expect to find the primary evidence for belief in God within the unobjectifiable 'portion' of the existential idea, namely, in the recognition of relations. God is not to be sought in the object, for God is neither material nor empirically verifiable, and thus is not the object of our knowledge. If God is not to be found in the object, yet belief in him stems from actual states of affairs, we must look to the relations which exist among objects, to the axioms we believe to universally adhere. And it is precisely these relations upon which the classical arguments for God's existence are founded. Aquinas' Five Ways are grounded squarely in the cause-effect relation.
This being the case, however, a distinction between my own position and Natural Theology must be made. We can easily locate the basis of this distinction in my previous disagreement with Thomas' priority of axiomatic knowledge. What I expect to find from the axiom is not a window into the transcendent nature of God (as Thomas expects). Instead, I expect simply to a trace of God which will serve in making statements about him while recognizing he is not constrained by the aspect I see. This distinction also speaks as to the force of such demonstrations. It is clear that very few people are argued into a belief in God through such methods. I would suggest that this is due to the fact that what is argued toward in not the divine nature itself but instead an aspect which serves a necessary function in our reality. The compulsion to attribute such a function (such as prime mover or first cause) to the divine nature is not adequate to cause all who hear to convert. If the argument allows for alternative solutions, and those other solutions involve much less effort of the conscience, it is to be expected that many would dissent from the intended conclusion. Thus I would agree with Wolterstorff's critique of Natural Theology's emphasis on reason and neglect of (fallen) human will.
It is also important to note that our priority of relations in the search for evidence for God would also include, perhaps principally, the notion of religious experience. Religious experiences, unobjectifiable in the same manner we have considered all relations, remains a very real part of the actual states of affair. In the same way we can only believe (as opposed to know) axiomatic relations universally adhere, our religious experiences will be the object of belief rather than knowledge. This point, however, I cannot make dogmatically, for it again involves the initial qualms I had regarding whether or not to consider a memory belief or knowledge.
Chisholm, Roderick M. Theory of Knowledge. (New Jersey: Prentice Hall; 1989)
Locke, John.. An Essay Concerning Human Understanding. Woozley, A.D. ed. (Meridian Books; 1964)
Russell, Bertrand. The Problems of Philosophy. (New York: Oxford Univ Press; 1997)