One of the main philosophical problems is the problem of how to conceive of reality. This is also a problem for our understanding of religions. The difficulty consists in finding a middle way between the claim that reality is wholly mind-independent and idealism. In this regard, the pragmatic tradition in philosophy promises to provide us with some fruitful ideas for steering a path between the two. I will apply some of these ideas in developing a pragmatic realist philosophy of religion which, I will argue, is not reductionist and therefore acceptable for religious as well as non-religious philosophers of religion. The structure of the paper will be as follows. First, I will give a very short summary of pragmatism as a historical background to my own way of doing philosophy of religion. [1] Second, in contrast to the notion of realism in the pragmatic tradition I will sketch the presuppositions of what is labelled religious or theological realism in present analytic philosophy of religion. Third, I will distinguish between ontological commitments that are metaphysical in character and ontological commitments that are not, drawing on Rudolf Carnap’s idea of the difference between internal and external questions of existence. Fourth, I will present Hilary Putnam’s criticism of a metaphysically realist conception of existence and fifth, his defence of what he calls internal realism. Sixth, I will put forward a pragmatic idea of the difference between observational experiences and existential ones in our lives. Finally, I will apply this pragmatic philosophy of religion to the question of whether it is reasonable to claim that belief in God presupposes God’s existence.

Most pragmatists try to find a middle way between metaphysical realism and relativism, between dogmatism and scepticism by using the pragmatic maxim, that in order to ascertain the meaning of a conception we should consider what practical consequences might conceivably result from the truth of that conception.

Belief in the sense of what is taken to be true is therefore not conceived of as the propositional content of a statement detached from praxis, but as guiding our actions in that it is a habit, a disposition to behave. Its opposite is disbelief and doubt. Unlike René Descartes’ methodical doubt, this type of doubt is involuntary and unpleasant. It is usually caused by some surprising phenomenon that is inconsistent with one’s previously accepted beliefs. When doubt arises, we usually start a process of inquiry by which we strive to obtain a new equilibrium with our environment in which our doubts are removed; this new equilibrium implies new habits and revised beliefs. The question of coherence is relevant to this process. Coherence should here, however, be understood not as mere logical coherence, but as including an affective equilibrium with one’s environment as well. A successful inquiry leads to stable opinion. This stability, however, is only temporary, and will in the long run be followed by new doubts.

Whereas Charles Sanders Peirce elaborated this idea of inquiry concerning science, John Dewey widens it to take on universal scope. After Dewey, most pragmatists conceive of this method simply as the way we actually think and ought to think in all areas of life. Cognitive experience is the result of inquiry. The process starts in a felt difficulty, proceeds through the stage of conceptual elaboration of possible resolutions, and results in a final reconstruction of the experience into a new unified whole. This unified whole is not, logically speaking, a closed system. We call it a unified whole because of the felt immediacy of the whole as an experienced whole.

With this idea of a unified whole, pragmatists question what is called ‘the spectator theory of knowledge’, according to which knowledge is a kind of passive recording of antecedent facts. Instead, knowing is seen as a constructive conceptual activity, anticipating and guiding our adjustment to future experiential interactions with our environment. We cannot, therefore, ascribe an absolute status to the classical ontological distinctions between mind and body, between means and end, and between fact and value in particular. These distinctions should rather be understood functionally and contextually. Consequently, most pragmatists deny truth as correspondence of thought to things-in-themselves. Instead, truth is a matter of successful interadjustment of our ideas to problematic situations.

William James shares this view. However, in his conception of truth he does not focus only on the testable consequences of a belief. Rather, he shifts the emphasis to the consequences for the person having a belief. True beliefs work. Not surprisingly, this conception of truth has been taken as an outspoken identification of truth with utility. James, however, distinguishes between different ways in which different beliefs work. Concerning empirical judgements, ‘true’ means ‘verified through observation and experiment’. Thus, the accusation of identifying truth with utility cannot be applied to empirical judgements. Neither can the accusation be applied to a priori truths since they are truths which we are prepared to accept as the conceptual presuppositions by means of which we talk about reality. It is only with regard to a third class of truths, the moral, the aesthetic and the religious, that there is a pragmatic interrelation between truth and utility. These judgements cannot be empirically verified. Their truth-value is given by their practical function in our lives. If religions are not to become idle talk, they must have practical consequences for the people who adopt them, i.e. they have to work in an existentially satisfactory way in their lives.

In one respect there is, according to pragmatism, no difference between science and religion. Both activities have something to say about who we are as human beings. In neither of these activities one talks about reality-as-it-would-be-independent-of-human-experience. However, whereas science deals with experimental, observational experience, existential experience is what is in focus in religion. Obviously, this kind of pragmatic approach is opposed to mainstream analytic philosophy of religion.

We are told that religious believers as well as atheists maintain that belief in God presupposes God’s existence. Underlying this is the combination of two theses. Firstly, if a person believes in God, he or she also believes that God really exists. Secondly, the question whether or not God exists is a question of fact. The second thesis is an essential part of what is called religious or theological realism, when we define this as the position according to which ‘there is a transcendent divine reality, the principal object of religious belief and language, the existence of which is not contingent upon (or, positively, is independent of) our thoughts, actions and attitudes.’ [2] It is claimed emphatically that

Obviously, when realist religious believers talk about a transcendent divine reality, they manifest a special ontological commitment. Let me illustrate what can be meant by ontological commitment, drawing on Carnap’s distinction between internal and external questions of existence put forward in his ‘Empiricism, Semantics and Ontology.’ [4] Although Carnap is not a pragmatist, his way of reasoning about existence shows some affinity with the pragmatists’ approach to the concept of existence.

When we talk about existence, we need, according to Carnap, to decide which conceptual framework should be used: The framework of everyday life? That of mathematics? That of fairy tales? That of religious language? The framework of quantum physics? Or still another framework? Certainly, in sentences within these frameworks words occur in the material mode of language use, i.e. these sentences are expressed in the object-language and appear in Carnap’s view misleadingly to refer to real objects and to express factual commitments concerning them. The question, however, is not, whether or not the entities that are constituted in the one or other conceptual framework, such as numbers, so to speak, really exist. Which conceptual framework should be reasonably applied, will be decided by its relevance for the activities we need or wish to perform and for the problems we need to cope with. After a specific conceptual framework has been accepted, substantial internal questions can be raised as, for instance, whether there are prime numbers between 10 and 20.

Thus, Carnap does not reckon with a metaphysical concept of existence, in relation to which the supposed existence of any kind of entity should be investigated. Instead, if the conceptual framework that explains our experience in the most satisfactory way presupposes some entity, we assume that this entity exists. However, why should we accept Carnap’s distinction between internal and external questions of existence ones and his idea that conceptual frameworks are important? Do not metaphysical realists maintain emphatically that the issue of God’s existence is not an issue of conceptual frameworks? In contrast to the non-metaphysical conception of ontological commitment, according to which there is no metaphysical concept of existence which the existence of pretended real entities should be related to, the metaphysical realist advocates an idea of ontological commitment on which there actually is a general metaphysical concept of existence. The existence of all allegedly real entities, whether they be trees, black holes or God, has to be related to this concept. The phrase ‘metaphysical concept of existence’ I use to indicate what according to the metaphysical realist is important to safeguard, namely, that what really exists and what not, is independent of human beings. If something exists, then it does so, irrespective not only of how we conceptualise it, but also of whether or not there is anybody who could conceptualise it. Using Putnam’s criticism of metaphysical realism I want to demonstrate what is problematic with this position, namely its disregard of the aspect of conceptualisation.

My reason for choosing the neo-pragmatist Putnam’s way of doing philosophy is the following. Common to metaphysical realists is that they maintain the objectivity of truth by defining truth in terms of correspondence with a reality that is independent of human consciousness. This reality in itself is an unconceptualised reality. Together with this definition of truth goes the conviction that the only conceivable alternative to this view consists in denying the objectivity of truth. Putnam’s goal is to find a way which enables one to keep truth objective, while rejecting the idea that truth is a correspondence between sentences and unconceptualised mind-independent facts. Putnam’s main reason for rejecting the idea of truth as correspondence with unconceptualised reality is, that if we accepted the idea of such a correspondence, we would need to presuppose that we have access to and are able to identify those facts or objects before we have concepts for them. The idea of correspondence is useful only if we can judge whether or not there is correspondence between, on the one hand, our judgements, i.e. the result of our conceptualisations, and on the other hand, facts or objects as such, i.e. unconceptualised facts or objects. To be able to judge whether or not there is correspondence we should have access to facts or objects as the one pole of the relation before we have concepts for them. However, if we want to identify facts or objects, they have to be conceptualised, and as a matter of fact, they can be conceptualised in different ways. Certainly, there are some fundamental categories or concepts or whatever we want to call them such as substance, time, place, person, identity, quantity and quality, without which we human beings could not experience reality as reality for us. Nevertheless, when we actually apply those categories, we can do this in different ways, thus creating different correspondences. What is a pencil in my hand, is conceptualised as something else by people who live in conditions where there is no use for a pencil and the concept of a pencil therefore has not arisen. The trouble with the idea of correspondence, thus,

Putnam does not deny realism but develops a form of internal realism: the question as to which objects the world consists of, can only be answered when we have elaborated a kind of conceptual net which enables us to describe the world. Let us call this conceptual net a theory in a very wide sense of the term. Objects in the world are therefore always objects conceptualised by us, and they differ depending on the theory at hand. This means that what is said to be true about the objects, presupposes a theory. This does not mean, however, that what is true about the objects is caused by the theory. Nevertheless, there is no practical point in claiming that truth consists in some correspondence with so to speak unconceptualised objects, since what we say about objects is theory-laden. The question of how one can bring about a relation of unique correspondence with objects that exist so to speak out there, becomes superfluous, since not only our conceptions but also the objects as they manifest themselves to us are internal in relation to some conceptual framework and the descriptions inside it. Therefore, it is preferable to see truth as ‘some sort of (idealised) rational acceptability – some sort of ideal coherence with each other and with our experiences as those experiences are themselves represented in our belief system.’ [6]

The internalism Putnam is advocating, is nevertheless a form of realism:

Let us assume for the sake of the argument and against what Putnam claims, that truth could be defined in terms of evidence or in terms of justification. Alvin Goldman shows the unreasonableness of such a procedure by means of the following example. A person is accused of having committed a serious crime. Several witnesses have identified the person as the one who has committed the crime, and the person lacks an alibi for the time during which the crime has been committed. Although the person is innocent, he cannot prove his innocence. Furthermore, the person who actually has committed the crime is dead. Under such conditions the accused person has no chance to defend himself. Now, if a definition of truth in terms of evidence is accepted, by necessity the result will be that the accused person is deemed guilty. Actually, however, it is not the case that the person has committed the crime. Goldman, therefore, draws the conclusion that

I agree, but this view is still compatible with Putnam’s idea of truth as ‘some sort of (idealised) rational acceptability.’ It does not need to imply the claim that truth is correspondence between judgements and words on the one hand and facts independent of conceptualisation on the other hand. At this point, Putnam’s internalism can get support from an argument in an article by Murat Baç who shows, using Goldman’s example, what tremendous practical difference a rejection of defining truth in terms of justification or in terms of evidence means for the actually innocent person. Baç points out

The normative function of the concept of truth I want to summarise as follows. We need the normative view of truth as independent of our epistemic abilities as a warning against arbitrary decisions and categorical judgements of different kinds. Thus, like the metaphysical realists Putnam, too, can distinguish between truth and justification and still stick to his emphasis on the internal character of our conceptions of reality. Putnam’s internal realism excludes neither the concept of truth nor the concept of objectivity. These concepts, certainly, are framed in relation to which beings we human beings are, but this does not amount to defining them in terms of our abilities. Let me develop this aspect further.

Putnam asks his readers to perform the following thought experiment.

Putnam’s example shows two things. First, that according to one conceptual framework, there are three objects, whereas according to another conceptual framework, there are seven objects. Second, that the one conceptual framework need not be less useful than the other one. That is the reason why there may be more than one true description of a state of affairs. When one has started to apply a certain conceptual framework, then what is said within it, vague statements discarded, is either objectively true or objectively false. Within the first conceptual framework, it is objectively true that there are three objects, and the person who claims that there are seven, claims what is objectively false. The other way round, if one were to use the second conceptual framework, it is objectively true that there are seven objects, and the person who claims that there are only three, claims what is objectively false. Let me apply the import of this example to religion.

We remember that according to Putnam’s internal realism, it is not necessary to claim that reality is independent of conceptual frameworks in order to talk about objectivity. Depending on our conceptual framework, what is said within the conceptual framework at issue, is objectively true or objectively false. From that, however, it does not follow that in principle one could choose whatever conceptual framework fits one’s own purposes, and then could claim that what one maintains is objectively true or false. It is definitely more complicated than that. Let me explain this by pointing out how science on the one hand and religion on the other hand are part of our life depending on the different functions they have in human life.

A theory, I suggest, can roughly be defined as a structural conception of what reality could be considered to consist of and how its constituents could be considered to relate to each other. On a first step, a theory is conceived of as something we have created or developed, but it is still undecided whether or not, so to speak, there is something substantial in it. On a second step, the proposed theory is applied to a given area of investigation in order to generate hypotheses about different kinds of entities. On a third step, finally, the hypotheses are tested to find out whether or not there is correspondence between statements about observations that the hypotheses lead us to expect on the one hand and statements about observations that have actually occurred on the other. If this correspondence obtains, the hypothesis on trial is at least not falsified and perhaps even confirmed. Let us now assume a situation in which due to new knowledge, a certain theory no longer can be used in order to generate testable hypotheses. This theory will no longer be used for claiming something about reality. In other words, ‘to accept a theory is (for us) to believe that it is empirically adequate—that what the theory says about what is observable (by us) is true.’ [11] What is at issue, is a continuously ongoing interplay between observations, hypotheses and theories, the latter functioning as conceptions of what reality could be like. The interplay is concretely performed in that we relate theories via testable hypotheses to our observational experiences of the empirical resistance reality offers.

I will now reason analogously concerning religion. Religion is not about observational experiences but about existential ones of what it means to be a human being. For these experiences we need conceptualisations, too. With regard to the different functions these two kinds of experience have in our lives, I will not use the term ‘knowledge’ in connection with religions but prefer instead the term ‘insight.’ Contrary to a common view in religion, I want to maintain that religion does not provide us with knowledge. My reason for reserving the word ‘knowledge’ for the sciences and for everyday well-tried experience, is that there we have experimental procedures, related to observational evidence, by means of which we can test whether or not our knowledge claims are justified. In religion we have no such experimental tests of putative knowledge claims.

The fact that on my definition we cannot speak of knowledge claims in religion, [12] does not entail that religion has nothing to say about the important things in life. It can still tell us, for instance, about love, joy and happiness on the positive side and about suffering, guilt and death on the negative one. Moreover, many religions put forward important insights about god(s) or other transcendent beings. Remember that what we talk about is always conceptualised by us humans as the beings we are. In this regard, religion is no exception. Let us assume that we are in a situation of crisis, so that there is no equilibrium among our beliefs any longer. In the first place, we have to find a conception of what it means to be a human being, i.e. to live in the tension between what life could be at its best and how it actually is. In the second place, we have to apply these conceptions to our concrete life with its felt experience of love, joy, happiness, suffering, guilt and death. These are concrete realities in our lives, although we can neither explain them nor eliminate them. Nevertheless, we need expressions for them, if we want to live a good life. When we can recognise ourselves and our life situation in them, we experience them as existentially adequate.

Now, let us assume that because of technical, political, cultural and other changes in society we can no longer experience the historically given expressions of suffering, guilt and death and of love, joy and happiness as existentially adequate expressions of what it means to be a human being, i.e. to live in the tension between what life could be at its best and how it actually is. What happens is that these existential expressions fade and cease to play an important role in people’s life so that they have to be modified or replaced by new ones. This is in analogy with what happens in the sciences when theories are no longer empirically adequate.

According to some philosophers and theologians this amounts to depriving religious faith of its truth content. According to them the truth of religious faith presupposes that there are supernatural facts irrespective of whether or not we know about them. Obviously, this is a kind of metaphysical realism that, as I have argued, is philosophically problematic. Seen from our human angle – and what other angle could there be? – the only form of realism for us is internal realism. This internal dimension concerns all of our conceptions. Neither the conceptions of science, nor those of religion are about unconceptualised reality. For us, reality is always reality as conceptualised and understood by us. The main difference between science on the one hand and religion on the other is that whereas our observational experiences are conceptualised in the former, our existential experiences are conceptualised in the latter. Seen from the internal perspective, religious conceptions of what it means to be a human being, and of experiences of God’s presence in one’s life, also are human conceptions. However, neither concerning the observational experiences nor the existential ones, it is only an individually psychological question. The question of what kind of biological and social beings we are, is also significant. Also our existential conceptions of what it means to be a human being should be capable of meeting the resistance that our lived reality offers. We therefore need to scrutinise these conceptions critically. Neither in science nor in religion can we decide empirical or existential adequacy once and for all. In the latter case, too, we require a continuous interadjustment of our conceptions of what it means to be a human being, through concretely experiencing the resistance life offers in a mixture of love, joy, happiness, suffering, guilt and death.

What does this pragmatic internal perspective mean for the question of whether belief in God presupposes God’s existence? Drawing on Putnam I have argued that it is only within a theory, i.e. in relation to a certain conceptual framework, meaningful to raise the question of what reality consists of. Analogously, I would like to add that it is only within a religion, an ideology or a view of life of some kind, meaningful to raise the question of what our concretely lived reality consists of. Just as our theories were developed in confrontation with observational experiences, our religions also have emerged from a confrontation with existential experiences. In both cases, our conceptualisations continue to be confronted with experiences of reality; and in both cases, this confrontation may lead to revision. I share the realist’s belief that we do not bring about what is true through our conceptualisations. Nevertheless, it is by means of conceptualisations by which we apprehend as adequate in relation to our observational and existential experiences, that we experience reality.

Now let us assume that the conceptions within a certain conceptual framework are experienced as existentially adequate, and that conceptions of God are an essential part of these conceptions. Then on the basis of having accepted the conceptual framework at issue, it is both logically and psychologically impossible not to reckon with the divine and ascribe reality to it. It is not meaningful to claim over and above that, that God exists really in some reality as such. As the criticism of metaphysical realism has shown, at an epistemic level this kind of truth or knowledge claim is useless. Instead, the concept of truth as independent of our epistemic means has a normative function in our actions by reminding us that religious conceptions also are human conceptions. Reasoning internally concerning religion, however, is therefore not an arbitrary affair, since also with regard to existential experiences the choice of a conceptual framework is not completely unrestricted. When conceptual frameworks no longer offer empirical or existential adequacy in relation to the problems we meet in life, they are actually abandoned in science as well as in religion. Both in science and in religion, objectivity is still possible, but ought to be seen in relation to a continuous interadjustment of our conceptions through observational or existential experiences.

In this paper I have proposed a pragmatic realist philosophy of religion. I have argued that the only perspective we human beings can adopt is the internal one, which takes account of the beings we are. This is the pragmatic aspect. Reality is always reality conceptualised by us building on the experience of resistance reality, so to speak, offers. This is the realist aspect. Furthermore, I introduced the distinction between observational experiences and existential ones drawing on the fact that science on the one hand and religion on the other have different functions in our lives. Let me conclude this paper by showing that this does not amount to reductionism with regard to religion. An essential part of my use of Putnam’s criticism of metaphysical realism and his defence of internal realism is the following distinction. It is one thing to define religion in terms of human needs and human perspectives. It is quite another thing to discuss the concepts of truth and objectivity in religion from the only perspective that is available to us, namely the human perspective. This is not reductionism. It is taking religion seriously as a human phenomenon. [13]