« Prev A. The Real Referent and the Available Referent Next »

A. The Real Referent and the Available Referent

According to Gordon Kaufman,

The central problem of theological discourse, not shared with any other “language game,” is the meaning of the term “God.” “God” raises special problems of meaning because it is a noun which by definition refers to a reality transcendent of, and thus not locatable within, experience. A new convert may wish to refer the “warm feeling” in his heart to God, but God is hardly to be identified with this emotion; the biblicist may regard the Bible as God’s Word; the moralist may believe God speaks through men’s consciences; the churchman may believe God is present among his people—but each of these would agree that God himself transcends the locus referred to. As the Creator or Source of all that is, God is not to be identified with any particular finite reality; as the proper object of ultimate loyalty or faith, God is to be distinguished from every proximate or penultimate value or being. But if absolutely nothing within our experience can be directly identified as that to which the term “God” properly refers, what meaning does or can the word have?3333   God the Problem (Cambridge: Harvard University Press, 1972), p. 7. Henceforth GP.

So the claim is that God is not to be identified with any particular finite reality—on the grounds, presumably, that God is not in fact identical with any particular finite reality. From the Christian perspective, this is, of course, no more than the sober truth: God is infinite and therefore not identical with any finite reality. So far, so good. Kaufman apparently infers from this, however, that “absolutely nothing within our experience can be directly identified as that to which the term ‘God’ properly refers”; he adds that if this is so, then there is a real problem for the reference of our term ‘God’: if “nothing within our experience can be directly identified as that to which the term ‘God’ properly refers, then what meaning does or can the word have?” I realize this last is a question, but it looks like a rhetorical question; the idea is that if nothing within our experience can be directly identified as that to which the term ‘God’ properly refers, then the term ‘God’ doesn’t refer to anything, or at any rate there is a real problem about its referring to something.

Here, therefore, we have two claims:


(a) if God is not a finite reality, then absolutely nothing within our experience can be directly identified as that to which the term ‘God’ properly refers.


(b) if nothing within our experience can be directly identified as that to which the term ‘God’ properly refers, then the term ‘God’ doesn’t refer to anything, or at least it is problematic that it does.

These claims awaken Kantian echoes—echoes that get stronger as we move further into Kaufman’s thought. And surely both are initially dubious. Consider (a). First, we must ask what it means to say that “nothing within our experience can be directly identified as that to which the term ‘God’ properly applies.” What is it, as Kaufman is thinking of it, for something to be within our experience, and to be such that it can be directly identified as that to which a certain term properly applies? What about my friend’s cat Maynard: is Maynard something within our experience which can be directly identified as that to which the term ‘Maynard’ properly applies? I should think so: else the problem is not merely with reference to God, but with reference to anything at all; Kaufman’s suggestion, I think, is that the problem is specifically with respect to God. According to (a) it is because God is infinite that the term ‘God’ doesn’t properly apply to anything within our experience. Now why, precisely, is that true? Maynard, I take it, is something within our experience, and this is because we can experience Maynard. We can perceive him: we can see, hear, touch, and sometimes smell him. The idea must be, then, that if God is not a finite reality, then we cannot experience him; we cannot perceive him (we cannot see, hear, or touch him) or in any other way experience him. An infinite being—one that is omnipotent and omniscient, for example—cannot be perceived or experienced in any way whatever.

Is that really true? How does the fact that God is infinite mean that we cannot experience him? Many Christians and Jews believe that God spoke to Moses from the burning bush; Moses heard him. He spoke to Abraham in a dream. He spoke to several people when he said, “This is my beloved Son in whom I am well pleased”; these people all heard him. Christians may also believe that the Holy Spirit works in their hearts, producing conviction and faith, as well as the religious affections of which Jonathan Edwards spoke; are they not then experiencing God? The term ‘experience’ (taken as either a noun or a verb) is notoriously slippery, but if these things do in fact happen, do not the people involved experience God? Christians may go still further and hold that in some circumstances some people perceive God, a theme that has received explicit and powerful treatment in William P. Alston’s Perceiving God. If they are right, then in these 34cases too they experience God. Now Kaufman apparently thinks the fact that God is infinite—unlimited along several dimensions—means that these people are mistaken: whatever they think, they do not experience God. Again, why so? God is infinite with respect to power, that is, omnipotent: how does that so much as slyly suggest that God cannot make himself heard or that he cannot be experienced? He is infinite with respect to knowledge, that is, omniscient; does that somehow show that he could not speak to Abraham or anyone else? Is it perhaps the combination of omnipotence and omniscience that shows this? It is certainly hard to see how.

If God is omnipotent, infinitely powerful, won’t he be able to manifest himself in our experience, bring it about that we experience him? He will be unable to do so, presumably, under those conditions, only if it is logically impossible (impossible in the broadly logical sense) that an omniscient and omnipotent being should be able to make himself heard. But so far as I can see, there isn’t even the slightest reason to think that; certainly Kaufman gives us none. I will go into the question of the nature of experience of God in more detail in chapters 6, 8, and 9; here I only want to point out that it seems initially implausible to declare that God, if he is infinite and omnipotent, could not bring it about that we experience him.

The second premise (b)—the claim that if nothing within our experience can be directly identified as that to which the term ‘God’ refers, then the term ‘God’ doesn’t refer to anything (or it is at least problematic that it does)—also seems dubious. Cosmologists tell us of the Big Bang, an event that occurred several billion years ago in which an explosion of enormous energy caused an expansion from an initial configuration of enormous density. I suppose the Big Bang is not something within our experience, something that can be directly identified as that to which the term ‘the Big Bang’ correctly refers; does it follow that there is a profound problem with this term? Is the real problem with contemporary cosmology not just the speculative nature of those suggestions about many universes and what happened during Planck time, but rather the very idea that we can refer to and think about that initial Big Bang? It isn’t easy to see why: at the least, a powerful argument would be required. And if there is no particular problem here, why is there a special problem in the case of God?3434   One source of Kaufman’s views here may be a sort of lingering allegiance to the “Verifiability Criterion of Meaning” mentioned above (pp. 7–8): “Since seemingly no clear experiential evidence can be cited for or against that to which the word ‘God’ allegedly refers, the question has been repeatedly raised whether all talk about him is not in the strict sense cognitively meaningless” (p. 8). As we saw in chapter 1, however, there is little to be said for the Verifiability Criterion.


Well, then, someone might say, if there is no problem about referring to an infinite being, how do we refer to God? In chapter 1, I suggested that we could do so, first, by way of definite descriptions such as ‘the creator of the heavens and the earth’, ‘the omnipotent and omniscient creator of the world’, ‘the divine father of our lord and savior Jesus Christ’, ‘the divine person who spoke to Abraham’, ‘the divine person I am presently experiencing,’ and so on. Each of these descriptions will refer to something if there is exactly one thing exemplifying the properties mentioned in the description; if not, then the description will not refer. (If Christian belief is true, of course, then each of these terms does refer to something, indeed, to the same thing.)

Furthermore, we can use the proper name ‘God’ to refer to the being denoted by those descriptions. That term can serve as a proper name, for me, of God, in several ways. For example, I might ‘fix the reference’ of the term ‘God’ by one of the above descriptions, such as ‘the creator of heaven and earth’; if, indeed, just one person created the heavens and the earth, and if that person is also denoted by those other descriptions, then my name ‘God’ will be a proper name of the same being as that denoted by those descriptions. My name will be a proper name of a being who is omniscient, omnipotent, the creator of the world, the father of our lord and savior Jesus Christ, and the like. Under these conditions, my name ‘God’ will express an essence of that being.3535   See my The Nature of Necessity (Oxford: Oxford University Press, 1974), pp. 77ff. Perhaps my name, introduced in that way, will not express the same essence of God as your name, introduced by way of a different description. Even so, however, they will express logically equivalent (even if epistemically inequivalent) essences of God.3636   See my “The Boethian Compromise,” in American Philosophical Quarterly (1978).

Alternatively, I might not get my proper name of God by using a definite description to fix the reference and then officially baptize the thing to which the description refers: I might instead just catch the name, so to speak, from others. In fact, this is the more usual way. Proper names, like colds, are ordinarily caught from our associates. As a child, I hear talk of God, talk in which the name ‘God’ occurs; I pick up the name, tacitly or implicitly intending to use it to refer to the same being to which those from whom I get the name refer. If they do indeed succeed in referring to God by using that name, then so will I. (Here is another way in which the success of my noetic ventures depends on the success of similar ventures on the parts of those around me: see Warrant and Proper Function, pp. 77–78.)

In any event, Kaufman holds that we can neither know nor experience what he calls ‘the real referent’ of the term ‘God’:

The real referent for “God” is never accessible to us or in any way open to our observation or experience. It must remain always an unknown X. . . . (GP 85)


When Christians use the term ‘God’, therefore, they do not refer to the real referent of that term (but then why call it “the real referent”?). To whom or what (if anything) do they refer when they say such things as that God was in Christ, reconciling the world to himself, or that God created the heavens and the earth, or that God is our faithful and loving father? The answer, says Kaufman, is that when they say these things they are referring to the “available referent” of the name ‘God’, and the available referent is an imaginative construct, something we have somehow created:

For all practical purposes it is the available referent—a particular imaginative construct—that bears significantly on human life and thought. It is the “available God” whom we have in mind when we worship or pray . . . it is the available God in terms of which we speak and think whenever we use the word “God.” In this sense “God” denotes for all practical purposes what is essentially a mental or imaginative construct. (GP 85–86)

God is a symbol—an imaginative construct—that enables men to view the world and themselves in such a way as to make action and morality ultimately (metaphysically) meaningful. (GP 109)

So the available God, the God whom we have in mind when we worship and pray, the being to which we refer when we use the term ‘God’—this being is a human creation, an imaginative construct, something we ourselves have created. The view seems to be initially that there is this available referent, but also a real referent of the term ‘God’, a being with whom we have no noetic contact and about whom we cannot speak. Or rather, the view is, I think, that there might be a real referent, and that if there is, it is a being we cannot think about:

This fact, that the God actually available to people is an imaginative construct, does not necessarily mean that God is “unreal” or “merely imaginary” or something of that sort. That question remains open for further investigation. (GP 86)

Does this mean, then, that the conclusion is, after all, that God really does not exist, that He is only a figment of our imaginations? If those words are intended to put the speculative question about the ultimate nature of things, then, as we have seen, there is no possible way to give an answer. (GP 111)

In essence, then, Kaufman’s view in God the Problem appears to be the following. The term ‘God’ has an available referent: this is a human construction, something we have created; when we speak of God in worship or to him in prayer, it is this available referent about which (or to which) we are talking. Perhaps the term also has a real referent. If so, however, it transcends our experience and is hence something to which our concepts do not apply: a mere unknown X, to adopt Kaufman’s Kantian terminology.


Now I’ve already argued that there seems no good reason to hold this position. Here I must go on to add that there is excellent reason not to hold it. As it stands, the view is incoherent. First, the ‘available referent’: the suggestion is that when Christians pray and worship and speak about God, they are talking about the available referent. When they say such things, for example, as ‘God created the heavens and the earth’, they are really attributing this property—the property of having created the heavens and the earth—to the available referent. But the available referent is a human construct, and hence presumably did not exist before there were human beings. How then did it manage to create the heavens and the earth? Could it somehow do this before it existed? In any event, an imaginative construct, a symbol, a structure of meanings of some kind is just not the sort of thing that could create the heavens and the earth or, indeed, anything else. A symbol, an imaginative construct, may have properties: being a construct, for example, or being a symbol, or being appropriately used by human beings for such and such a purpose; it certainly won’t have such properties as being omniscient or creating the world. I suppose it could be that Christians are confused: they think they are referring to and talking about something that created them, but the fact is they are referring to something they themselves have created. Is it really plausible to think they are as confused as all that, however? Those who believe there is no such person as God will see Christians as mistaken in thinking there is, and perhaps it is at least sensible to think them mistaken in that way. Is it really sensible to think them mistaken in such a way that they predicate the properties of God of a mere construct? Well, perhaps that could happen; but surely a strong argument would be required to make this even reasonably plausible.

Say that a property P entails a property Q just if it is necessary in the broadly logical sense that everything that exemplifies P also exemplifies Q; and say that a concept C contains a property P if the property of which C is a grasp entails P. Then it is clear that a concept might contain such properties as being omniscient or having created the world (even if it couldn’t exemplify them), and equally clear that the concept corresponding to the definite description ‘the omniscient creator of the world’ contains the properties being omniscient and being the creator of the world. Could it be that what Kaufman really means is not that Christians assert that the available referent—which is something like a concept containing salient properties of God—exemplifies those properties, but rather that it contains them? This too seems wrong. It is indeed true that certain concepts, including some associated with descriptions of God, contain those properties. When Christians make their characteristic claims, however, they are not merely saying such things as that the concept being the omniscient creator of the heavens and earth contains the properties being omniscient and being the creator of the heavens and earth. That would, of course, be true; it would also be wholly trivial. It wouldn’t be at all distinctive of 38Christians or theists: even the most hardened atheist would agree that this concept contains those properties. What Christians claim entails rather that these properties are exemplified, that there really exists a being who has them.

The above seems to be the literal construal of Kaufman’s words; of course there are other possibilities in the neighborhood. Perhaps, for example, he thinks of the available referent not as a being with the properties Christians ascribe to God, but as something like a certain type with which those properties are associated.3737   See Nicholas Wolterstorff’s so-far-unpublished From Presence to Practice; Mind, World, and Entitlement to Believe, chapter 1. This may seem a more sympathetic construal of Kaufman; I doubt that it really is. If Kaufman’s claim is that Christians ordinarily worship that type, then his claim is outrageous in just the way I suggest. If his claim, however, is only that Christians believe they worship a being having the properties associated with the type but are possibly mistaken, then is his claim more than the uninteresting suggestion that Christians may be wrong about whether there is such a person as God?

Now consider the real referent. The idea is that our concepts do not apply to the real referent, if indeed there is such a thing. It follows that this being is not wise, almighty, or the creator of the heavens and the earth. For consider our concept of wisdom. This concept applies to a thing just if that thing is wise. So a being to which this concept did not apply would not be wise, whatever else it might be. If, therefore, our concepts do not apply to the real referent of the term ‘God’, then our concepts of being loving, almighty, wise, creator, and redeemer do not apply to it, in which case it is not loving, almighty, wise, a creator, or a redeemer. It wouldn’t have any of the properties Christians ascribe to God. And of course so far this is in accord with Kaufman’s intentions.

I suspect, however, that his official position has other consequences Kaufman does not intend. If this being, this real referent, is really such that none of our concepts applies to it, then it will also lack such properties as self-identity, existence, and being either a material object or an immaterial object, these being properties of which we have concepts. Indeed, it wouldn’t have the property of being the real referent of the term ‘God’ or any other term; our concept being the referent of a term will not apply to it. The fact is this being won’t have any properties at all because our concept of having at least one property does not apply to it. Kaufman’s view seems to entail that there could be a being that had no properties, didn’t exist, wasn’t self-identical, wasn’t either a material object or an immaterial object, and didn’t have any properties.

Taken strictly, therefore, Kaufman’s position is incoherent.

« Prev A. The Real Referent and the Available Referent Next »
VIEWNAME is workSection