AGNOSTICISM – All you want to know
There are various approaches or methods to the question of God, some positive and some negative. Perhaps the most widely held in the latter category is agnosticism. There are two basic kinds of agnostics: those who claim that the existence and nature of God are not known, and those who hold God to be unknowable. Since the first type does not eliminate all religious knowledge, attention here will center on the last one.
The term agnosticism was coined by T. H. Huxley. It means literally no-knowledge, the opposite of a gnostic.1 However, over a hundred years before Huxley the writings of David Hume and Immanuel Kant laid down the philosophical basis of agnosticism. Much of modern philosophy takes for granted the general validity of the types of arguments they set forth.
The Basic Arguments of Agnosticism
Even Immanuel Kant was a rationalist until he was “awakened from his dogmatic slumbers” by reading David Hume.
The Skepticism of David Hume
Technically Hume’s views are skeptical but they serve well the agnostic aim. Hume set forth the basis of his position in the concluding lines of his famous Enquiry Concerning Human Understanding.
If we take in our hands any volume; of divinity or school metaphysics, for instance; let us ask. Does it contain any abstract reasoning concerning quantity or number? No. Does it contain any experimental reasoning concerning matter of fact and existence? No. Commit it then to the flames: for it can contain nothing but sophistry and illusion.2
That is, any statement that is neither purely a relation of ideas (definitional or mathematical) on the one hand or a matter of fact (empirical or factual) on the other hand is meaningless. Of course all statements about God fall outside these categories, and hence knowledge of God becomes impossible.
Only Two Kinds of Propositions.
At the basis of Hume’s conclusion that all meaningful propositions are reducible to two is a radical empiricism that may be summarized as follows. All of our knowledge or ideas is derived either through sensation or by the reflection on ideas (derived from sensation) in the mind. There is nothing in the mind that was not first in the senses.
Furthermore, all sensations are experienced as “entirely loose and separate.” Causal connections are made by the mind only after one has observed a constant conjunction of things in experience. All one really experiences is a series of unconnected and separate sensations. Indeed, there is no direct knowledge even of one’s “self,” for all we know of ourselves is a disconnected bundle of sense impressions.
It does make sense of course to speak of connections among ideas, even necessary connections. But these are connections made only in the mind a priori or independent of experience. A posteriori (i.e., from experience) there are no known and certainly no necessary connections. In fact, all matters of experience imply a possible contrary state of affairs. For anything we experience in one way could be otherwise.
Causality Is Based on Custom.
Many who believe in God are willing to admit that they have no direct knowledge of God but claim nonetheless to have access to the existence and nature of God via his effects or the things he has made or said. Hume’s epistemology (theory of knowledge), if true, would seem to eliminate this possibility as well.
For, according to Hume “all reasoning concerning matters of fact seems to be founded on the relation of cause and effect. By means of that relation alone can we go beyond the evidence of our memory and senses.”3 And knowledge of the relation of cause and effect is not a priori but arises entirely from experience. And the idea of a causal relation appears in the mind only after there has been an observation of constant conjunction in experience.
That is, only when we observe death to occur after holding another’s head under the water for five minutes do we assume a causal connection. Once one event is observed to happen after another repeatedly, we begin to form the idea that one event happens because of the other. In brief, the idea of causality is based on custom.
Customary conjunction of events leads one to believe in or posit a connection between them. Of course this connection cannot be known but is simply believed because of the repetition of the conjunctions. There is always the possibility of the post hoc fallacy—namely, that things happen after other events (even regularly) but are not really caused by them.
For example, the sun rises regularly after the rooster crows but certainly not because the rooster crows. One can never know causal connections. And without a knowledge of the Cause of this world, for example, one is left in agnosticism about such a supposed God.
Knowledge of God by Analogy Is Highly Problematic.
Even if one were to grant that every event has a cause, nevertheless he cannot build any knowledge of God upon this fact because the analogy is weak at best.
In his famous Dialogues Concerning Natural Religion4 Hume contends that the cause of the universe may be:
- different from human intelligence since human inventions differ from those of nature;
- finite, since the effect is finite and one only need infer a cause adequate for the effect;
- imperfect, since there are imperfections in nature;
- multiple, for the creation of the world looks more like a long-range trial and error product of many cooperating deities;
- male and female, since this is how humans generate; and
- anthropomorphic, with hands, nose, eyes, and so forth, such as his creatures have.
Since no theist will admit that analogy leads to these anthropomorphic deities, it leaves us in skepticism about the nature of any supposed Cause of the world.
The Agnosticism of Immanuel Kant
The writings of Hume had a profound influence on the thinking of Kant. Before reading them Kant held a form of rationalism in the tradition of Leibniz. Leibniz and Wolfe following him believed reality was rationally knowable and that theism was demonstrable. They followed a long line of Western thinkers from Plato through Augustine, Anselm, and Aquinas who held that there were proofs for the existence of God. It was the pen of Kant that put an abrupt end to most of this thinking in the philosophical world.
The Impossibility of Knowing Reality.
Kant granted to the rational tradition of Leibniz that there was a rational, a priori dimension to knowledge, namely, the form of all knowledge is independent of experience. On the other hand, Kant granted Hume and the empiricists their basic contention that the content of all knowledge came via the senses.
The “stuff” of knowledge is provided by the senses but the structure of knowledge is attained eventually in the mind. This creative synthesis solved the problem of rationalism and empiricism. However, the unhappy result of this synthesis is agnosticism, for if one cannot know anything until after it is structured by the a priori forms of sensation (time and space) and the categories of understanding (such as unity and causality), then there is no way to get outside one’s own being and know what it really was before he so formed it.
That is, one can know what something is to-him but never what it is in-itself. Only appearance can be known, but not reality. In Kant’s words, we know the phenomena but not the noumena. There is a great impassable gulf between the real world and our knowledge of it; we must remain agnostic about reality. We know only that it is there but can never know what it is.5
The Antinomies of Human Reason.
There is another argument for Kant’s agnostic conclusion. Not only is there an unbridgeable gulf between knowing and being, between the categories of our understanding and the nature of reality, but there are also the inevitable contradictions that result once we begin to trespass the boundary line. In other words, when we take the necessary forms of sensation or categories of understanding, such as the principle of causality, and apply them to reality we run headlong into unavoidable contradictions.6
There is, for instance, the antinomy of time. If we assume that the form of sensation known as time (the “whenness” with which we time bound creatures sense things) applies to reality, we must conclude the following contradictions. On the one hand, if the world had a beginning in time, then an infinity of moments must have elapsed before the world began. But this is impossible because an infinity of moments can never be completed.
On the other hand, if the world did not have a beginning in time, then there must have been a time before time began—which is impossible. But either the world began in time or it did not, and both positions are impossible. Hence, by applying time to reality one eventuates necessarily in contradictions. And since contradictions do not yield knowledge, reality is unknowable.
Another antinomy concerns the category of causality. First, not every cause can have a cause or else a series of causes would never begin to cause—which they in fact do. On the other hand, if everything has a cause, then there cannot be a beginning cause and the causal series must stretch back infinitely. But it is impossible that the series be both infinite and also have a beginning. Such is the impossible paradox resulting from the application of the category of causality to reality.
There is also the antinomy of contingency. We must posit that not everything is contingent; otherwise there would be no basis or condition for contingency. On the contrary, everything must be contingent for necessity applies only to thought and not to things, since any state of affairs could be otherwise. But again reality cannot be both contingent and necessary. The way to avoid such contradiction is to acknowledge that reason cannot know reality, viz., by agnosticism.
These arguments do not exhaust the agnostic’s arsenal, but they do lie at the heart of the contention that God cannot be known. However, even some who are unwilling to admit to the validity of these arguments opt for a more subtle form of agnosticism. Such is the case with the school of thought to which we turn our attention next, logical positivism.
The “Acognosticism” of A. J. Ayer
Following up on Hume’s distinction between definitional and empirical statements, Ayer offered the principle of empirical verifiability. This affirmed that in order for statements to be meaningful they must be either analytic (Hume’s “relation of ideas”) or synthetic (Hume’s “matter of fact”)—that is, definitional or empirical.7
The former are devoid of content and say nothing about the world; the latter have content but tell us nothing about any alleged reality beyond the empirical world. Furthermore, the latter are only probable in nature and are never philosophically certain. They are useful in empirical and practical matters but not at all informative about reality in any metaphysical sense.
All God-Talk Is Nonsense or Empty. The result of Ayer’s logical positivism is as devastating to theism as is traditional agnosticism. God is unknowable and inexpressible. In fact, it is even meaningless to use the term God. Hence, even traditional agnosticism is untenable, since the agnostic assumes that it is meaningful to ask the question whether God exists. For Ayer, however, the word God or any transcendent equivalent, has no meaning.
Hence, it is impossible to be an agnostic. For the term God is neither analytic nor synthetic; that is, it is neither offered by theists as an empty, contentless definition corresponding to nothing in reality nor is it a term filled with empirical content, since “God” is allegedly a supraempirical being. Hence, it is literally nonsense to talk about God.
It is true that Ayer later revised his principle of verifiability.8 But even in this form (that admitted the possibility that some empirical experiences are certain, such as those of a single sensory experience, and that there is a third kind of statement, viz., some analytic or definitional statements that are not purely arbitrary such as his own principle of verifiability) he did not thereby allow for the meaningfulness of God-talk. This third class would be neither true nor false nor factual but meaningfully definitional.
However Ayer believed that “it is unlikely that any metaphysician would yield to a claim of this kind,” even though he acknowledged that for “an effective elimination of metaphysics it needs to be supported by detailed analyses of particular metaphysical arguments.”9 In short, even a revised principle of empirical verifiability would make it impossible to utter meaningfully true statements about a transempirical reality such as God. There is no cognitive knowledge of God; we must remain “a-cog-nostic.”
“God” Is Inexpressible or Mystical.
Following a tip from Wittgenstein’s Tractatus, Ayer held that while God might be experienced, such an experience could never be meaningfully expressed. Wittgenstein believed that “how things are in the world is a matter of complete indifference for what is higher. God does not reveal himself in the world.” For “there are indeed, things that cannot be put in words.… They are what is mystical,” and “what we cannot speak about we must consign to silence.”10
If God could express himself in our words it would indeed be “a book to explode all books,” but such is impossible. Hence, there not only is no prepositional revelation, but there are no cognitively meaningful statements that can be made about any alleged or real transcendent being. Hence, whether one takes the more strict logical positivist’s principle of verifiability or even the broader Wittgensteinian linguistic limitations, God-talk is metaphysically meaningless.
To be sure, as Wittgenstein taught, language games are possible, even religious language games. God-talk can and does occur, but it is not metaphysical; it tells us nothing about the existence and nature of a being beyond this world. About this we must, because of the very necessary limitations of language, remain silent. In summary, for religious noncognitivists Ayer and Wittgenstein, metaphysical acognosticism is the net result of language analysis.
It makes little difference to the Christian or theist whether he cannot know God (as in Kant) or whether he cannot speak of God (as in Ayer). Both traditional agnosticism and contemporary acognosticism leave us in the same dilemma philosophically: there are no bases for making true statements about God.
The Unfalsifiability of Religious Beliefs.
The other side of the principle of verifiability is that of falsifiability. Taking his cue from John Wisdom’s parable of the invisible gardener, Antony Flew posed a challenge to believers as follows: “What would have to have occurred to constitute for you a disproof of the love of, or of the existence of, God?”11 For one cannot allow anything to count for his belief in God unless he is willing to allow something to count against it.
Whatever is meaningful is also fasifiable. There is no difference between an invisible, undetectable gardener and no gardener at all. Likewise, a God who does not make a verifiable or falsifiable difference is no God at all. Unless the believer can indicate how the world would be different if there were no God at all, he cannot use conditions in the world as evidence that there is a God.
In short, unless the theist can answer the challenge head-on, then it would appear that he must have what R. M. Hare called a “blik.”12 That is to say, he has an unfalsifiable belief in God despite all facts or states of affairs. It matters little whether the believer calls his “blik” a parable, a myth, or whatever; the fact remains that he is an acognostic believer with no meaningful or verifiable knowledge of God, and this is little or no improvement on Kant’s traditional agnosticism.
An Evaluation of Agnostic Arguments
As was indicated earlier, there are two forms of agnosticism: the weak form simply holds that God is unknown, that is, that we do not know God. This of course leaves the door open that one may know God and indeed that some do know God. As such this kind of agnosticism forms no threat to Christian theism. The second or strong form of agnosticism is mutually exclusive with Christianity.
It claims that God is unknowable, that is, that God cannot be known. Even here one must make an important distinction before embarking on a critique. There is unlimited and limited agnosticism about God. The former claims that God and all reality is completely unknowable. The latter claims only that God is partially unknowable because of the limitations of man’s finitude and sinfulness. We will take it that the latter form of agnosticism is both possible and desirable. Paul wrote, “For now we see in a mirror dimly.… Now I know in part” (I Cor. 13:12).
This leaves us with three basic alternatives with respect to knowledge about God. First, we can know nothing about God; he is unknowable. Second, we can know everything about God; he is completely and exhaustively knowable. Third, we can know something about God but not everything; he is partially knowable. The first position we will call agnosticism; the second, dogmatism; and the last, realism.
Now it is evident that the dogmatic position is untenable. One would have to be God in order to know God exhaustively. Finite man can have only a finite knowledge of the infinite, not an infinite knowledge. Few if any informed believers have seriously held this kind of dogmatism. However, theists sometimes argue against agnosticism as though partial agnosticism is wrong too.
They argue that agnosticism is wrong simply because one cannot know something is unknowable about reality without thereby implying a knowledge about that something. But this is faulty reasoning. There is no contradiction in saying, “I know enough about reality to affirm that there are some things about reality that I cannot know.”
For example, we can know enough about observation and reporting techniques to say that it is impossible for us to know the exact population of the world at a given instant (unknowability in practice). Likewise, one may know enough about the nature of finitude to say that it is impossible for men to know exhaustively an infinite being (who could not be exhaustibly knowable in principle for finite man as we know man).
In the following critique we will be concerned only with the complete agnostic who rules out in theory and practice all knowledge of God. This kind of complete agnosticism is self-defeating.
Agnosticism Is Self-Defeating
Complete agnosticism is self-defeating; it reduces to the self-destructing assertion that “one knows enough about reality in order to affirm that nothing can be known about reality.” This statement provides within itself all that is necessary to falsify itself. For if one knows something about reality, then he surely cannot affirm in the same breath that all of reality is unknowable. And of course if one knows nothing whatsoever about reality, then he has no basis whatsoever for making a statement about reality.
It will not suffice to say that his knowledge about reality is purely and completely negative, that is, a knowledge of what reality is not. For every negative presupposes a positive; one cannot meaningfully affirm that something is not-that if he is totally devoid of a knowledge of the “that.” It follows that total agnosticism is self-defeating because it assumes some knowledge about reality in order to deny any knowledge of reality.
Some have attempted to avoid the logic of the above critique by putting their skepticism in the form of a question: “What do I know about reality?” However, this does not avoid the dilemma but merely delays it. This question can and ought to be asked by both agnostic and Christian. But it is the answer that separates the agnostic from the realist. “I can know something about God” differs significantly from “I can know nothing about God.” Once the answer is given in the latter form a self-defeating assertion is made.
Neither will it help to take the mutist alternative of saying nothing. For thoughts can be as self-stultifying as assertions. The mutist cannot even think he knows absolutely nothing about reality unless in that very thought he implies that he does know something about reality.
Of course someone may be willing to grant that knowledge about finite reality may be possible but not willing to allow any knowledge about an alleged infinite reality, such as the God of Christian theism. If so, two things should be noted. First, the position is no longer complete agnosticism, for it holds that something can be known about reality.
This leaves the door open to discuss whether or not this reality is finite or infinite, personal or impersonal. Second, the latter discussion takes us beyond the question of agnosticism to the debate between finite godism and theism (which will be discussed later). Before we take up some of the specific arguments of agnostics it will be helpful to further illustrate how agnosticism involves a self-defeating assertion.
Reply to Kant’s Agnosticism.
Kant’s argument that the categories of thought (such as unity and causality) do not apply to reality is unsuccessful, for unless the categories of reality corresponded to those of the mind no statements could be made about reality, including that very statement Kant made. That is to say, unless the real world were intelligible no statement about it would apply.
A preformation of the mind to reality is necessary whether one is going to say something positive about it or something negative. We cannot even think of reality that it is unthinkable. Now if someone should press the argument that the agnostic need not be making any statement at all about reality but simply defining the necessary limits of what we can know, it can be shown that even this is a self-defeating attempt; for to say that one cannot know any more than the limits of the phenomena or appearance is to draw an unsurpassable line for those limits.
But one cannot draw such firm limits without surpassing them. It is not possible to contend that appearance ends here and reality begins there unless one can see at least some distance on the other side. In other words, how can one know the difference between appearance and reality unless he already knows both so as to make the comparison?
Another self-defeating dimension is implied within Kant’s admission that he knows that the noumena is there but not what it is. Is it possible to know that something is without knowing something about what it is? Can pure that-ness be known? Does not all knowledge imply some knowledge of characteristics?
Even a strange creature one had never seen before could not be observed to exist unless it had some recognizable characteristics as size, color, or movement. Even something invisible must leave some effect or trace in order to be observed. One need not know the origin or function of a brand-new he-knows-not-what. However, he must observe something of what it is or he could not know that it is.
It is not possible to affirm that something is without simultaneously declaring something about what it is. Even to describe it as the “in-itself” or the “real” is to say something. Furthermore, Kant acknowledged it to be the unknowable “source” of the appearance we are receiving. All of this is informative about the real: namely, it is the real, in-itself source of impressions we have. Even this is something less than complete agnosticism.
Reply to Hume’s Skepticism.
There are several levels on which we may reply to Hume. First, the overall skeptical attempt to suspend all judgment about reality is self-defeating, since it implies a judgment about reality. How else could one know that suspending all judgment about reality was the wisest course, unless he knew indeed that reality was unknowable? Skepticism implies agnosticism and, as was shown above, agnosticism implies some knowledge about reality.
Unlimited skepticism which commends the suspension of all judgments about reality implies a most sweeping judgment about the knowability of reality. Why discourage all truth attempts, unless one knows in advance that they are futile? And how can one be in possession of this advance information unless he already knows something about reality?
Second, Hume’s contention that all meaningful statements are either a relation of ideas or else about matters of fact is itself neither of these. Hence, on its own grounds it would be meaningless. It could not be purely a relation of ideas, for in that case it would not be informative about reality as it purports to be.
And clearly it is not purely a matter-of-fact statement since it claims to cover more than empirical matters. In short, Hume’s distinction is the basis for Ayer’s empirical verifiability principle, and the verifiability principle is itself not empirically verifiable.
Third, Hume’s radical empirical atomism that all events are “entirely loose and separate” and that even the self is only a bundle of sense impressions is unfeasible. If everything were unconnected there would be no way of even making that particular statement, since some unity and connection are implied in the affirmation that everything is disconnected.
Further, to affirm “I am nothing but the impressions about myself” is self-defeating, for there is always the assumed unity of the “I (self)” making the assertion. But one cannot assume a unified self in order to deny the same.
Reply to Ayer’s Acognosticism.
As has already been noted, the principle of empirical verifiability as set forth by Ayer is self-defeating. For it is neither purely definitional nor strictly factual. Hence, on its own grounds it would fall into the third category of non-sense statements. Ayer recognized this problem and engaged in recovery operations by way of a third category for which he claimed no truth value but only a useful function.
Verifiability, he contended, is analytic and definitional but not arbitrary or true. It is meta-cognitive, that is, beyond verification as true or false but simply useful as a guide to meaning. This is a classic but ill-fated move for two reasons. First, it no longer eliminates the possibility of making metaphysical statements. Rather, it admits that one cannot legislate meaning but must look at meaning of alleged metaphysical statements. But if it is possible that some meaningful statements can be made about reality, then we are not left with complete agnosticism and acognosticism.
Second, can cognitively restrictive meta-cognitive statements be made without self-stultification? It seems not, for to restrict the area of what is meaningful is to limit the area of what could be true, since only the meaningful can be true. Hence, the attempt to limit meaning to the definitional or to the verifiable is to make a truth claim that must itself be subject to some test. If it cannot be tested, then it becomes an unfalsifiable view, a “blik” of its own.
Reply to Wittgensteinian Mysticism.
Ludwig Wittgenstein engages in a self-stultifying acognosticism. He attempts to define the limits of language in such a way as to show that it is impossible to speak cognitively about God. God is literally inexpressible. And that whereof one cannot speak, he should not attempt to speak thereof.
But Wittgenstein can be no more successful in drawing the lines of linguistic limitation than Kant was in delimiting the realm of phenomena or appearance; for how can one know that God is inexpressible without thereby revealing something expressible about God? The very attempt to deny all expressions about God is an expression about God. One cannot draw the limits of language and thought unless he has transcended those very limits he would draw. It is self-defeating to express the contention that the inexpressible cannot be expressed.
In like manner even to think the thought that the unthinkable cannot be thought is self-destructive. Language (thought) and reality cannot be mutually exclusive, for every attempt to completely separate them implies some interaction or commerce between them. One cannot use the scaffold of language and thought about the limits of reality only to say the scaffold cannot be so used. If the ladder was used to get on top of the house, one cannot thereupon deny the ability of the ladder to get one there.
Reply to Flew’s Falsifiability.
Two things must be said about Flew’s principle of falsifiability. First, in the narrow sense of empirical falsifiability it is too restrictive. Not everything need be empirically falsifiable. Indeed that very principle is not itself empirically falsifiable. But in the broader sense of testable or arguable, surely the principle is alive and helpful.
For unless there are criteria for truth and falsity, then no truth claims can be supported. Everything, including opposing views, could be true. But in this case nothing can be maintained to be true (as versus what is false), for no such distinction can be made.
Second, not everything that is verifiable need be falsifiable in the same manner. As John Hick pointed out, there is an asymmetrical relation between verifiability and falsifiability. One can verify his own immortality, for example, if he consciously observes his own funeral. But one cannot falsify his immortality, for if he does not survive death then he is not there to disprove his own immortality.
Nor could another person falsify one’s immortality unless he were omniscient or God. For it is always possible that my existence could be somehow beyond his limited knowledge. But if it is necessary to posit an omniscient mind or God, then it would be eminently self-defeating to use falsification to disprove God. So we may conclude that every truth claim must be testable or arguable but not all truth claims need be falsifiable or disconfirmable.
A total state of nonexistence of anything would be unfalsifiable, for example, since there would be no one and no way to falsify it. On the other hand, the existence of something is testable by experience or inference.
Reply to Some Specific Agnostic Claims
Hume denied both the traditional use of causality and analogy as a means of knowing the theistic God. Causality is based on custom and analogy would lead to either a finite manlike god or to a totally different God than the alleged analogue. Let us examine each of these in turn.
Causality Is Not Unjustifiable. First, Hume never denied the principle of causality. He admitted it would be absurd to maintain that things arise without a cause.13. What he did attempt to deny is that there is any philosophical way of establishing the principle of causality. If the causal principle is not a mere analytic relation of ideas but is a belief based on customary conjunction of matter-of-fact events, then there is no necessity in it and one cannot use it with philosophical justification.
But we have already seen that dividing all contentful statements into these two classes is self-defeating. Hence, it is possible that the causal principle is both contentful and necessary. In point of fact, the very denial of causal necessity implies some kind of causal necessity in the denial. For unless there is a necessary ground (or cause) for the denial, then the denial does not necessarily stand.
And if there is a necessary ground or cause for the denial, then the denial is self-defeating; for in that event it is using a necessary causal connection to deny that there are necessary causal connections.
Some have attempted to avoid the logic of the above objection by limiting necessity to the reality of logic and propositions but denying that necessity applies to reality. But this will not succeed because in order for this statement to accomplish what it intends to do, namely, to exclude necessity from the realm of reality, it must itself be a necessary statement about reality.
That is, it must in effect be claiming that it is necessarily true about reality that no necessary statements can be made about reality. It must make a necessary statement about reality to the effect that necessary statements cannot be made of the real. This is clearly self-canceling, for it actually does what it claims cannot be done.
Analogy Is Not Unfoundable.
Likewise, there is no way Hume can deny all similarity between the world and God, for this would imply that the creation must be totally dissimilar from the Creator. It would mean that effects must be entirely different from their cause. In actuality this statement too is self-destructive; for unless there were some knowledge of the cause there would be no basis for denying all similarity with its effect.
Comparison, even a negative one, implies some positive knowledge of the terms being compared. Hence, either there is no basis for the affirmation that God must be totally dissimilar or else there can be some knowledge of God in terms of our experience, in which case God is not necessarily totally dissimilar to what we know in our experience.
One should be cautioned here about overdrawing the conclusion of these arguments. Once it has been shown that total agnosticism is self-defeating, it does not ipso facto follow that God exists or that one has knowledge of God. These arguments show only that if there is a God, one cannot maintain that he cannot be known. From this it follows only that God can be known, not that we do know anything about God.
The disproof of agnosticism is not thereby the proof of realism or theism. In other words, agnosticism only destroys itself and makes it possible to build Christian theism. The positive case for Christian knowledge of God must be built later.
An Answer to Kant’s Antinomies.
In each of Kant’s alleged antinomies there is a fallacy. One does not end in contradictions when he begins to speak about reality in terms of the necessary conditions of human thought. For instance, we need not speak of the world beginning in time, as though time were already there and there was a time before time.
We may speak of the beginning of the world as the beginning of time. That is, time is a concomitant of a created world in process—in which case there would be no time before time. All that is prior to time is eternity.
It is likewise a mistake to view everything as needing a cause, for in this case there would be an infinity of causes and even God would need a cause. Only limited, changing, contingent things need causes. Once one arrives at an unlimited, unchanging, necessary being there no longer is a need for a cause. The finite must be caused, but the infinite being would be uncaused.
Finally, the so-called antinomy of contingency fails as well, for everything cannot be contingent. There must indeed be a ground for contingency that is beyond the contingent, namely, the necessary. And, as was previously noted, it is self-defeating to claim that necessity applies only to thought and propositions but not to being or reality; for that claim itself necessarily entails an affirmation about reality.
Thought and reality cannot be radically bifurcated without being irrevocably united; there is no way to affirm their separation unless they are in fact joined. This is not to say that the rational is the real, but it is to affirm that the real is rationally knowable.
Summary and Conclusion
There are two kinds of agnosticism: limited and unlimited. The former is no threat to Christianity but is compatible with its claim of finite knowledge of an infinite God. Unlimited agnosticism, however, is self-destructive; for it implies knowledge about reality in order to deny the possibility of any knowledge of reality. Both skepticism and noncognitivisms (acognosticism) are reducible to agnosticism.
For unless it is impossible to know the real, it is unnecessary to disclaim the possibility of all cognitive knowledge of it or to dissuade men from making any judgments about it. Skepticism and acognosticism imply agnosticism.
Finally, unlimited agnosticism is a subtle form of dogmatism. In completely disclaiming the possibility of all knowledge of the real, it stands at the opposite pole from the position that would claim all knowledge about reality. Each is equally dogmatic. Both are “must” positions regarding knowledge as opposed to the position that we may or doknow something about reality.
And there is simply no way short of omniscience that one can make such sweeping and categorical statements about reality, whether they are positive or negative. Agnosticism is negative dogmatism, and every negative presupposes a positive. Hence, total agnosticism is not only self-defeating but it is self-deifying. Only an omniscient mind could be totally agnostic, and finite men confessedly do not possess omniscience. Hence, the door remains open for some knowledge of reality. Reality is not unknowable.
SELECT READINGS FOR Agnosticism
Exposition of Agnosticism
- Hume, David. An Enquiry Concerning Human Understanding.
- Huxley, T. H. Collected Essays, vol. V.
- Kant, Immanuel. Critique of Pure Reason.
- Stephen, Leslie. An Agnostic’s Apology.
Evaluation of Agnosticism
- Collins, James. God in Modern Philosophy, chaps. IV and VI.
- Flint, Robert. Agnosticism.
- Garrigou-Lagrange, Reginald. God: His Existence and His Nature.
- Hackett, Stuart. The Resurrection of Theism, pt. I.
- Ward, James. Naturalism and Agnosticism.
1 See T. H. Huxley, “Agnosticism and Christianity” (1889), in his Collected Essays (London: 1894), vol. V.
2 London, 1748; modern edition by C. W. Hendel (New York: 1955).
3 Enquiry, see. IV, pt. 2.
4 London, 1779; recent edition by Norman Kemp Smith (New York: 1972).
5 See The Critique of Pure Reason, trans. Norman Kemp Smith (New York: St. Martin’s Press, 1965), especially p. 173 f.
6 Kant, Pure Reason, p. 393 f.
7 A. J. Ayer, Language, Truth and Logic (New York: Dover Publications, Inc., 1946; first published 1936).
8 See “Introduction” to rev. ed., p. 10 f.
9 Ayer, p. 16.
10 Tractatus Logico-Philosophicus, 6:44, 6432, 6522 (London: 1922; trans. D. F. Pears and B. F. McGuinness [London: 1961]).
11 “Theology and Falsification,” in New Essays in Philosophical Theology, ed. Antony Flew and Alasdair MacIntyre (London: SCM Press, Ltd.: 1955), p. 99.
12 Flew and MacIntyre, p. 100.
13 See David Hume, “A Letter from a Gentleman to His Friend in Edinburgh,” ed. Ernest C. Mossner and John V. Price (Edinburgh: University Press, 1967)
[1]Geisler, N. L. (1976). Christian apologetics. Includes index. (13). Grand Rapids: Baker Book House.