A PRAGMATIC CASE AGAINST PRAGMATIC THEOLOGICAL REALISM

Authors


Abstract

Pragmatic theological realism (PTR) urges us to take up the realist aim of theology or the goal of truth although we have good reason to think that the goal can neither be attained nor approximated. Rescher contends that pursuing an unreachable goal can be rational on pragmatic grounds so long as pursuing the unreachable goal yields indirect benefits. I have blocked this attempt at providing a pragmatic justification for the realist aim of PTR on precisely the same pragmatic grounds: since there is a competing alternative to PTR and the alternative can provide whatever indirect benefits PTR can offer while being less risky than it is, prudential reasoning favours the alternative to PTR. This undermines the pragmatic case for the realist aim of theology since the instrumentalist alternative does not aim at the truth.

INTRODUCTION

The cognitive status of the claims of religion is a perennial issue in theology and the philosophy of religion. If theological statements and theories are not merely useful fictions devoid of cognitive content, can they – and to what extend can they – claim to be reliable representations of reality? If the cognitive dimension of religion underlies its emotive, evaluative and pragmatic aspects, truth must be a focus in religion and theology. This focus on truth leads to an important question: is there any justification for the belief that theology is a truth-orientated inquiry? Theologians and philosophers arguing for some forms of theological realism often draw analogies from and model their defences upon scientific realism. One prominent example is Wentzel van Huyssteen (1989, 155), who avers that ‘scientific realism, in the form of a qualified critical realism, has tremendous resources for supporting the reliability and validity of theological assertions’. In this paper I shall focus on van Huyssteen's proposal of a pragmatic form of theological realism modelled upon Nicholas Rescher's pragmatic scientific realism, and argue that it is unworkable. But I shall first trace the backgrounds that led to his pragmatic realist turn.

THE REALIST CLAIM OF THEOLOGY AND ITS DISCONTENTS

Scientific realism can be defined as consisting of three theses: (1) semantic thesis – theoretical statements of science should be taken at face-value in its reality depiction; (2) teleological thesis – scientific theories aim at ‘true descriptions of the world’; and (3) epistemic thesis – this aim or goal is legitimate (Goldman 1986, 157–60). Convergent Scientific Realism (CSR) is the view that theories of mature sciences progress towards increasing verisimilitude or truth-likeness in terms of their referential and reality-depicting capacity (Psillos 1999, xix; Boyd 1991, 195), and CSR is usually meant when one mentions scientific realism. An upshot of this view is that, if scientific theories approximate the truth or reality in its referential capacity, theoretical or unobservable terms postulated in scientific theories should be seen as ‘putatively referring expressions’ (Boyd 1991, 195).

Theologians addressing the issue of realism in science and religion seem to favour the term ‘critical realism’, for they think that ‘realism’ implies naïve realism, which takes the world exactly as we experience it. They contend that realism should be qualified as ‘critical’ since ‘knowledge of the real world can only be acquired through critical reflection upon experience’ (van Kooten Niekerk 1998, 51; van Huyssteen 1999, 214). But since scientific realists do not take realism as implying naïve realism, the qualifier ‘critical’ is not necessary. What theologians mean by ‘critical realism’ is what scientific realists mean by realism.

But modelling theological realism on scientific realism has its problems. Whilst attention has too often been focussed on identifying the parallels in the use of models and metaphors in science and religion, the crucial question of the legitimacy of extending the notion of realism from science to religion has often been overlooked. When arguing for a realist outlook on religion/theology, these authors often emphasise that although scientific theories should be construed realistically, in most cases they can only capture some aspects of reality. This characteristic makes scientific theories function like models, analogies, and metaphors. They highlight the analogical and metaphorical nature of religious language and the prevalence of models in theology, and argue that religion/theology should therefore also be seen from a critical realist perspective (e.g. Peacocke 1984, 40–50; van Huyssteen 1999, 218; McGrath 1998, 141–205; McGrath 2002).

Although models and metaphors are used both in science and religion, it does not follow without further arguments that a parallel to the realism in the former can be found in the latter. While models and metaphors are features of scientific theories, they are not the features that provide grounds for our belief in scientific realism, or the ‘credentials’ that secure its claim to realism. As pointed out by William Abraham (2003, 274), ‘the crucial features securing the [sic] scientific realism,’ viz. predictive power, technological control, and carefully constructed experiments, ‘are not at all available to theology’. In fact, a central focus of the defence of or evidence for scientific realism lies in ‘the instrumental reliability of scientific methodology’ (Boyd 1991, 207), as indicated by the predictive and technological success of science (pp 205, 208). In other words, the realist claim of science rests on the success of scientific theories in producing ‘substantially correct predictions’ and ‘efficacious interventions in the natural order’ (Laudan 1981, 1118). Theological theories clearly cannot claim such a notion of success.

IN SEARCH OF A VIABLE MODEL OF REALISM: FROM CONVERGENTISM TO PRAGMATISM

Unlike many theologians who model upon scientific realism in their defence of theological realism, van Huyssteen shows awareness of this crucial disanalogy: theology cannot claim the instrumental reliability or empirical success underlying the mainstream defence of scientific realism. But, as I shall argue, what he proposes in response to this disanalogy is multiply problematic. He believes that any form of convergent theological realism (theological theories progress towards increasing verisimilitude) is indefensible, but thinks that this should not lead to the abandonment of the realist claim of theology because it is still possible to defend a form of theological realism without the idea of gradual progress to the truth in convergentism (van Huyssteen 1989, 150, 159). In place of CSR, he appeals to Ernan McMullin's scientific realism as a model of non-convergent realism for theological realism on the one hand (van Huyssteen 1989, 150–5), and to Rescher's pragmatic scientific realism (PSR) on the other. PSR does not lay its realist claim on the notions of ‘progress’ and ‘success’ (van Huyssteen 1999, 216). As such, this form of scientific realism is not ‘a justified position’, but a ‘methodological presupposition of our praxis of inquiry’ (van Huyssteen 1996, 259; 1998, 39), which can only be defended on pragmatic grounds.

Nevertheless, appealing to these two forms of scientific realism in an attempt to construct a non-convergent form of theological realism is incoherent since they are incompatible: McMullin's is in fact a form of CSR and, given PSR's rejection of convergentism, CSR is incompatible with PSR. Van Huyssteen (1989, 150–1) obviously and wrongly assumes that McMullin's scientific realism is not a version of CSR. The crux of the matter lies in van Huyssteen's problematic and incoherent view of convergence and approximate truth, thinking that in both scientific and theological realism one can have the latter without the former. Thus, he seems to think that McMullin's version of scientific realism affirms approximate truth without convergentism. As I shall argue below, one cannot reject convergentism without also rejecting approximate truth at the same time. However, even if we granted van Huyssteen's acceptance of approximate truth and rejection of convergentism and granted, counterfactually, that McMullin's is a version of scientific realism that affirms science's ability to attain approximate truth without convergentism, this would still make his appeal to PSR incoherent because PSR denies science's ability to achieve approximate truth.

To further complicate the problem, van Huyssteen's claim that theology can attain approximate truth is incompatible with his view of theory change in theology. His application of Thomas Kuhn's idea of radical conceptual revolutions in science to theology indicates that he not only thinks that we have no good reason to think that theology progresses towards increasing verisimilitude, but also believes that we have good reason to think that theology does not and will not progress towards increasing verisimilitude (van Huyssteen 1989, 63). This Kuhnian model of theological theory change leads to his denial of a cumulative growth of knowledge (p63). Paradigm shifts in systematic theology often lead to ‘divergent systematic-theological traditions’ and ‘fundamental … transformation of certain formerly standard axioms and procedures’ (p64). Moreover, he asserts that ‘[w]hat Kuhn called the … incommensurability of paradigms is … particularly noticeable in systematic theology’ (p65). In parallel with Rescher's (1987) depiction of the ‘constant radical changes in science’ (p15) that lead to ‘outright abandonment and replacement’ (p28) of old concepts, van Huyssteen (1989, 67) maintains that ‘radical breakthroughs’ in theology ‘always imply the rejection and replacement of a framework or essential parts of it’.

Elsewhere however, van Huyssteen (1988, 177) inconsistently claims that approximate truth is possible for a theological theory because it is possible to argue for explanatory progress in the theory. He defines explanatory progress as a ‘better theory or hypothesis’ that can solve experiential problems and reduce conceptual problems arising from religious experience (p177). There are two problems here. Firstly, the claim that explanatory progress in a theological theory gives us grounds to claim its approximate truth is inconsistent with his Kuhnian view of revolutionary theory change in theology and his view that any form of theological convergentism is indefensible. If there is explanatory progress and approximate truth in theology, can he consistently think that any form of theological convergentism is indefensible? It seems not. Suppose a theologian proposes Theory B and contends that it shows explanatory progress in relation to Theory A. Since van Huyssteen thinks that explanatory progress in a theological theory provides us with grounds to think that it approximates the truth (to a higher degree), Theory B has greater verisimilitude than Theory A, which implies that the change from A to B is a process of convergence. If it is not convergence, what else can it be? It cannot be divergence, or else it will rule out approximate truth. Can it be a process of change that is neither convergent nor divergent? But if the process of change is neither convergent nor divergent, the change from A to B cannot lead to approximate truth unless A and B already have the same degree of approximate truth. But if so where does the idea of explanatory progress arise? Furthermore, even if a theory is already approximately true, a process of change that is neither convergent nor divergent but is nonetheless revolutionary will not leave its approximate truth intact.

Perhaps what van Huyssteen meant to say is that we should infer that a theory is approximately true if it has explanatory power, in the sense of solving experiential problems and reducing conceptual problems. But this simply begs the question that this problem-solving ability is an indicator of the truth, and leads to the second problem: the appeal to explanatory power or progress renders no support to the idea of approximate truth in theology. In ordinary life and science, we have many counter-examples against this suggestion. Many successful explanations are neither true nor approximately true, which include, for example,

virtually all those geological theories prior to the 1960s which denied any lateral motion to the continents[,] … the chemical theories of the 1920s which assumed that the atomic nucleus was structurally homogeneous[, and] … those chemical and physical theories of the late 19th century which explicitly assumed that matter was neither created nor destroyed … (Laudan 1981, 1128).

Van Huyssteen appeals to the principle of inference to the best explanation and avers that this principle allows us to infer that a theological theory that has explanatory progress is approximately true. Put differently, we should infer from the ‘explanatory progress’ of a theory to its truth (van Huyssteen 1988, 177). But Bas van Fraassen (1980, 20) challenges the idea that we should infer from the explanatory progress or success of a theory to its truth, arguing that there is a rival hypothesis for the inference: we could infer from the explanatory progress or success of a theory to its empirical adequacy:

In this way I can certainly account for the many instances in which a scientist appears to argue for the acceptance of a theory or hypothesis, on the basis of its explanatory success … For remember: I equate the acceptance of a scientific theory with the belief that it is empirically adequate (van Fraassen 1980, 20).

In short, van Huyssteen's Kuhnian view of theological theory change indicates his rejection of convergentism, which renders his belief in the approximate truth of theological theory incoherent. At any rate, he has no basis to claim approximate truth for theological theories by appealing to their explanatory power and success.

In fact, van Huyssteen's problem with this has its root in his problematic view of scientific realism. He clearly looks for a version of scientifc realism that can provide an analogue to theological realism, and realises that CSR as advocated by philosophers such as Newton-Smith and Richard Boyd cannot be a good candidate because of CSR's emphasis on the empirical success that theology lacks (van Huyssteen 1999, 216). In order to provide a space for theological realism to manoeuvre, he argues that ‘progress as such need not at all be defined in the sense of a gradual growth toward truth’, and that ‘[e]ven if the history of at least some of the natural sciences shows progressive approximation to truth in the physical world, the realist position could never depend on ‘achiving truth’ as such’. (One wonders how progress can be understood without the idea of growth towards the truth, given van Huyssteen's view that explanatory progress indicates approximate truth.) What he tries to argue is that, even if CSR is possible, it is not the only viable form of scientific realism. In fact, he thinks that McMullin's scientific realism is a better version, and most importantly, it can be applied to theology (van Huyssteen 1989, 150–1). However, he mistakenly assumes that McMullin's scientific realism does not appeal to convergence and gradual growth towards the truth. In fact, McMullin (1984, 266) refers to the central idea of his scientific realism as ‘convergences of structural explanation’. In adducing evidence from the history of science for this idea McMullin appeals to the gradual and steady growth towards the truth and even to cumulative knowledge:

There was a lively debate about the mechanisms of mountain building and the like, but gradually a more secure knowledge of the past aeons built up … The long-vanished species of the Devonian are theoretical entities about which we have come to know more and more in a relatively steady way … But the very considerable theory changes that have occured since Hutton's day do not alter the fact that the growth in our knowledge of the sorts of life forms that inhabitated the earth aeons ago has been pretty cumulative (McMullin 1984, 266–7).

This indicates that McMullin's is a version of CSR, despite its differences from the CSR of ‘Maxwell, Salmon, Newton-Smith, Boyd, Putnam and others’. At any rate, he does not reject their versions of CSR. He simply claims that his version can provide the best case for scientific realism (p266). Whatever improvement he thinks he has made to other versions of CSR, it is certainly not an abandonment of convergentism or the idea of gradual growth towards the truth, as suggested by van Huyssteen.

All these undermine the applicability of McMullin's scientific realism to theology, given van Huyssteen's rejection of convergentism and growth towards the truth in theology. Moreover, a central characteristic of McMullin's scientific realism lies in the idea that the fertility of a theory, which ‘is usually associated with the ability to make novel predictions’ (p270), provides the grounds for the theory's realist claim. Empirical success, including predictive power and technological control, is clearly unavailable to theology. Therefore, van Huyssteen seems to be left with only one option, viz. Rescher's PSR. Can PSR provide a model of realism that will work for theology?

The purpose of this paper is to argue that, independent of whether or not PSR is a defensible form of scientific realism, this pragmatic form of realism is incapable of supporting the realist claim of religion or theology. But before arguing to this effect it is instructive to take a brief look at why PSR seems so promising (at least to van Huyssteen) as a potential model of realism that works for theology.

PRAGMATIC SCIENTIFIC REALISM

PSR is attractive to van Huyssteen because of its preservation of the realist aim of science, which it takes to be the sine qua non of scientific realism, despite its rejection of convergentism, and with it the idea of progressive approximation to the truth. Given van Huyssteen's acknowledgement that theology has no claim to convergent realism, it is not difficult to see why a form of scientific realism that is independent of convergentism looks promising to him.

The image of science Rescher depicts in (1987, Chapter 2, 3) is that of ‘constant radical changes’ (p15), which often bring about a ‘wholesale revision of ‘established facts’’ (p23). He clearly denies the cumulative progress of scientific knowledge when he says that scientific change is a matter of replacement rather than supplementation, and that the declarations of science are subject to ‘revision and even to outright abandonment and replacement’ (p28). In view of this, science's aim of getting at the truth is unreachable: ‘the real truth’ lies not with present nor future science, nor the future convergence of science (pp19, 25). This implies that science cannot even achieve gradual approximation to the truth (pp23–5). Such a view of science is very close to that of instrumentalism or antirealism. But Rescher stops short of becoming an instrumentalist by refusing to give up the realist aim/goal of science – science is an inquiry that aims at the truth.

However, if, as maintained by Rescher (1987, 23), both ‘historical experience’ and considerations of general principle' indicate that science can neither attain nor approximate the truth in its current and even future developments, there seems to be no legitimate reason why the truth or approximation to the truth should be taken as one of the goals of the scientific enterprise. For, as Newton-Smith (1981, 14) puts it so aptly, ‘how can it be rational to pursue that which we have evidence for thinking can never be reached? Is it rational to try and get to the moon by flapping one's arms if one has evidence that it will not work?’ But Rescher disagrees with Newton-Smith and contends that holding the realist aim of science can still be rational on pragmatic grounds:

The view that it is rational to pursue a goal only if we are in a position to achieve its attainment or approximation is a mistaken one. The goal can be perfectly valid, and entirely rational if the indirect benefits of its adoption and pursuit are sufficient – if in striving after it we realize relevant advantages to a substantial degree. An unattainable ideal can be enormously productive (Rescher 1987, 29).

In his 1987, Rescher attempts to provide a pragmatic case for the realist aim of science by arguing that holding the realist aim leads to indirect benefits that give one advantage over instrumentalism. In my 2005 (13–29), I argue against Rescher's pragmatic case by developing the so-called ‘prudential argument against PSR’. My purpose in this paper, however, is not to deal with PSR. What I wish to establish here is that, independent of whether or not PSR is defensible, the attempt to provide an analogous pragmatic justification of the realist aim of theology can be refuted by an analogous prudential argument.

A PRAGMATIC CASE FOR THE REALIST AIM OF THEOLOGY?

Theology is an inquiry in which we attempt to account for religious experience by formulating theological theories (van Huyssteen 1989, 144). From a realist perspective, theology aims at true theological theories. Pragmatic theological realism (PTR) contends that although the realist aim of theology is unreachable and cannot be justified by comparing the representation with the represented, it can be defended on pragmatic grounds.

It should be noted that van Huyssteen merely appeals to PSR as a form of realism applicable to theology and has not addressed the issue of the philosophical defensibility of both PSR and its theological analogue, viz. PTR. In other words, he has neither made a case nor provided any argument for PTR. Moreover, unlike Rescher in the scientific counterpart, he has not even developed the key tenets of PTR in any detail, not to mention clarifying and analysing the concepts involved. Therefore, as I develop the prudential argument against PTR, I will have to fill in these lacunae for him.

As argued above, theology can claim neither current approximation to the truth nor progressive growth towards the truth. This implies that theology has no epistemic or evidential grounds for the realist aim and that the realist goal is unreachable. The rationality of holding the realist aim of theology depends on whether a pragmatic case can be made for it, which depends on whether holding the unreachable realist goal can provide substantial advantages or indirect benefits. In order to examine whether ‘striving after’ PTR's realist aim provides substantial advantages or indirect benefits, we have to ask what these advantages or benefits are. There are two important points to bear in mind before we inquire into this question, and Rescher's strategy in his pragmatic argument for the realist aim of PSR provides important clues for the appropriate approach to be adopted for the pragmatic justification of the realist aim of PTR.

Firstly, it is essential to distinguish between the pragmatic justification for a choice or course of action and the pragmatic endorsement of a belief. Rescher's pragmatic justification for the realist aim of science belongs to the former. The pragmatic justification for the realist aim considers the pragmatic benefits or advantages brought about by a course of action or decision, and has nothing to do with the pragmatic validation of belief discussed in Rescher's (2000). The same should be said of the pragmatic justification for the realist aim of theology. In other words, the pragmatic justification for the realist aim of theology is not the pragmatic justification for the belief that theology aims at the truth. Rather, it is the justification for the choice or action of taking up the realist aim of theology. As such, it is not susceptible to the objection that the pragmatic validation of a belief is not truth-linked, which Rescher (2000, 93) happily agrees. Although holding the realist aim of theology involves the belief that theology aims at the truth (unless one is a radical sceptic), it does not render the pragmatic justification for the realist aim of theology susceptible to the objection that the pragmatic validation of a belief is not truth-linked because what seeks justification is the choice or action of taking up the realist aim, not the belief that theology aims at the truth.

Secondly, apart from the question of what the indirect benefits are, it is essential to consider what kind of benefits is acceptable. Is the pragmatic case for the realist aim of theology established if holding the realist aim brings just about any benefits to an individual or a community? All the benefits Rescher suggests for the realist aim of science have a common feature: they are the benefits that are only of interest to the effective implementation of the functions of science, i.e. purported benefits that help science perform its task effectively. I think Rescher's delimitation (if there is any conscious delimitation at all) is very wise, for it will provide a stronger case for the pragmatic justification for the realist aim. A pragmatic case for the realist aim that appeals to the benefits vis-à-vis science's effective implementation of its functions is stronger than one that appeals to other (non-cognitive) benefits such as the glory, power, and wealth of individual scientists, since only the former deals with the question that is of interest to the philosophy of science, viz. what course of action can best help science fulfil its functions?

Therefore, I shall develop an analogue of Rescher's delimitation and focus my considerations of the benefits on those that are of interest only to theology. As such, the type of acceptable indirect benefits considered here are only those that are of interest to the effective implementation of the tasks or functions of theology. This delimitation is especially appropriate since the philosophical discussion of the issue of theological realism is concerned about the nature of theological inquiry, and the question of the effective implementation of its functions is directly relevant to this concern. Thus, the non-cognitive indirect benefits of the realist aim for people who hold that aim (e.g. that it brings about better chance for securing the reward of eternal bliss, social cohesion, psychological well-being and etc.) are outside the purview of the present discussion.

THE PRUDENTIAL ARGUMENT AGAINST PTR

Corresponding to my prudential argument against PSR (Lee 2005, 13–29), I shall develop an analogous prudential argument against PTR to undermine the pragmatic case for the realist aim of theology. The argument compares PTR with theological constructive empiricism (TCE), an analogue of constructive empiricism (CE). It is instructive to take a quick overview of CE before considering my proposal of TCE and its comparison with PTR.

PTR Compared with TCE

CE is a form of non-eliminative instrumentalism that takes the language of science literally, i.e. scientific theories are not construed as metaphors (van Fraassen 1980, 11). As such, CE accepts the semantic thesis of scientific realism. However, it rejects the teleological and epistemic theses, and thus implies that science should not be seen as pursuing the goal of truth, because it maintains that one should only accept a scientific theory construed literally without believing it. Believing a scientific theory entails belief in its truth, which includes both its observable and unobservable aspects, whilst accepting the theory only involves the belief that it is empirically adequate, which means that what it says about ‘the observable things and events in this world is true’ (pp11–2). Its attitude to the existence of theoretical entities is that of agnosticism rather than denial.

CE's stance that allows acceptance of scientific theories also means that the postulation of theoretical entities for the purpose of prediction and explanation is not prohibited (van Fraassen 1980, 33–4). Van Fraassen contends that scientists can make use of the heuristic functions of theoretical entities without believing them. He avers that while a scientist is ‘totally immersed in the scientific world-picture' (p80), a world where one finds theoretical entities like electron, such immersion does not involve belief and commitment to the ontological implications of the world-picture (p81). ‘For to say that someone is immersed in theory, ‘living’ in the theory's world, is not to describe his epistemic commitment’ (p82). Some may question the possibility of being immersed in the scientific world-picture without belief or epistemic commitment. But there are examples in ordinary life when one gets totally immersed in an imaginative world in order to perform the relevant task effectively but can do so without any epistemic commitment. An actress has to get immersed in her role and imagines that she is the person she is playing. But the ability to perform this task effectively requires no epistemic commitment – she does not have to believe that she is the person she is playing. Likewise, Peter Lipton suggests to me in a private conversation that one who plays computer games need no epistemic commitment to his role – he need not believe that he is the boxer or pilot – to perform his ‘task’ effectively though he certainly gets immersed in that imaginative world.

Let's suppose TCE is a theological analogue of CE. It agrees with PTR in its denial of any form of theological convergentism. However, PTR and TCE differ in their epistemic attitudes to unobservable entities postulated by theological theories and their views on the realist aim of theology. Whereas PTR contends that we should believe that ‘what we are provisionally conceptualizing in theology really exists’ (van Huyssteen 1989, 155), and that we should hold the realist aim of theology even though theological theories cannot and will not attain nor approximate the goal of truth, TCE maintains that theology should aim at empirical adequacy, and that in some cases we can accept the theoretical entities of theology for the purpose of explanation without believing in their existence.

Readers who have been persuaded by my argument above that a rejection of convergentism entails a simultaneous rejection of approximate truth may find that PTR's claim that what we are provisionally conceptualising in theology really exists seems to be inconsistent with its rejection of convergentism in theology. If theological theories cannot approximate the truth, can one still believe the existence of what is postulated in a theological theory? Fraser Watts (1998, 161–5) appeals to Ian Hacking's scientific realism and suggests that one can still do so if she adopts realism about entities rather than realism about theories. The fact that theological theories cannot approximate the truth leads only to antirealism about theories, but one can still be a realist about entities.

However, theology's inability to achieve verisimilitude within the PTR framework does not allow van Huyssteen to hold entity realism. Any form of entity realism (e.g. Hacking's experimental realism) relies on the theory-independence of unobservable entities. But as David Resnik (1994) points out in his critique of Hacking's experimental realism, such independence is not possible since an experimental realist can only justify her belief in the existence of the unobservable entities if she has reason to think that the theories postulating these entities are approximately true. Realism about entities that is committed to antirealism about theories is unreasonable in that any antirealist about theories has no justified belief in the entity postulated by a theory (Resnik 1994, 1180). Setting aside the objection to Hacking's realism about entities, it should be noted that realism about entities cannot be applied to theology. Hacking's realism about entities or experimental realism relies on the experimental manipulability of an entity postulated by a theory, and this is unavailable to theology. If an entity postulated by a theory (e.g. God) has no experimental manipulability, there is no reason to think that it exists at all, since according to Hacking the reason for our belief in the existence of an unobservable entity postulated by a theory that is neither true nor approximately true is its experimental manipulability. In short, PTR's claim that what we are provisionally conceptualising in theology really exists is indeed incompatible with its rejection of convergentism and indefensible in virtue of its lack of experimental manipulability, and should thus be rejected.

An Overview of the Prudential Argument against PTR

The principle of prudential choice, which is a form of pragmatic reasoning, contends that, when facing some options underdetermined by epistemic or evidential considerations, the most rational action is to opt for the choice that implies maximal gain and minimal loss. My prudential argument against PTR is developed according to this principle of prudential choice:

  • (P1) PTR has no advantage over TCE vis-à-vis the indirect benefits (they are on a par in terms of potential gains)
  • (P2) PTR is riskier than TCE
  • (C) prudential reasoning favours TCE over PTR

The advantage of this strategy lies in the fact that it defeats PTR's pragmatic justification for the realist aim on its own terms – on pragmatic considerations. In the following sections I shall first show that holding a realist aim does not give one advantage over holding the aim of empirical adequacy because the indirect benefits that can be gained by holding the realist aim can also be gained by holding the aim of empirical adequacy. This renders PTR and TCE on a par vis-à-vis their abilities to bring about the indirect benefits. But I shall go on to argue that PTR is riskier than TCE, and this provides us with pragmatic grounds to favour TCE over PTR.

PTR has no Advantage over TCE

What are the indirect benefits of holding the unreachable realist aim of theology? Since van Huyssteen does not develop PTR in any detail, I will have to fill in the lacuna on this question by drawing parallels from PSR.

It should be stressed that only indirect benefits can provide a pragmatic justification for PTR's realist aim. The benefits are ‘indirect’ vis-à-vis the goal of truth because they are gained even when the goal of truth is not attained. The benefits that are obtained by holding a certain goal are ‘direct’ if they are obtained only when the goal is achieved or approximated, and they are ‘indirect’ if they can be obtained whether or not the goal is achieved or approximated. For example, if my goal is to get a PhD after three years of postgraduate studies, being awarded a PhD after three years is the direct benefit, while gaining the first-hand experience of PhD studies is an indirect benefit because this benefit can be obtained whether or not I achieve the goal of gaining a PhD after three years of studies. The qualifier ‘indirect’ is essential given PTR's rejection of theology's attainment of and approximation to the goal of truth. If a benefit can only be obtained on the assumption of theology's attaining a certain degree of verisimilitude, then it is actually a direct benefit. But a direct benefit is not obtainable given its dependence on the attainment or approximation of the goal and PTR's rejection of theology's attainment of the truth and verisimilitude.

As to the specific indirect benefits, perhaps we can draw some parallels from Rescher's PSR. Rescher (1987, Chap 4; cf. 2000, 105) contends that holding the realist aim of science gives one advantage over instrumentalism on these five counts. Firstly, since ‘depiction of nature’, which means the description of the physical world in its phenomenal and extra-phenomenal domains or getting at ‘the truth about the world's way’ (Rescher 1987, 33), has traditionally been viewed as the ‘quintessentially cognitive aspiration’ of science, holding the realist aim gives one advantage over instrumentalism because only a realist intent can do the job of science viewed traditionally. Secondly, holding the realist aim is more (perhaps psychologically) conducive to the exertion of the instrumental power of science in prediction and control. Thirdly, only a realist aim can give a fair trial to scientific theorising about observation-transcending reality by allowing explanation of observable phenomena in terms of unobservables. While the realist may be wrong in her extra-phenomenal explanation of a certain observable phenomenon, this is still better than instrumentalism, which urges us to ‘forego any attempt to explain the phenomena in non-phenomenal terms’ without giving it a fair trial (p44), because the realist at least has a chance to provide a true explanation, however slim the chance might be. Fourthly, thinking that all forms of instrumentalism prohibit the postulation of ‘real entities or processes’ as the extra-phenomenal causes, Rescher (1987, 50) avers that it deprives science of the ability to provide a naturalistic explanation of the unobservable phenomena, because ‘[w]hen the range of our acceptable claims is confined to the phenomenal sphere, we are deprived of the mechanisms through which alone we can construct a picture of ourselves as one item of physical reality among others …’ Finally, Ontological parsimony or economy seems to be the only advantage that instrumentalism can offer. But since instrumentalism purchases ontological economy, whose virtue and value are questionable, at the expense of explanatory economy, the insistence on ontological economy is actually a disadvantage (p53).

In summary, Rescher suggests that holding the realist aim within the PSR framework will bring about the benefits of (1) enabling us to describe the physical world, which includes both its phenomenal and extra-phenomenal aspects; (2) providing a more conducive environment for the instrumental power of science (i.e. prediction and control); (3) giving a ‘fair trial’ to theorising about observation-transcending reality by allowing explanation of the phenomenal in terms of the extra-phenomenal; (4) enabling us to develop a naturalistic explanation of phenomena; and (5) maximal gain and minimal loss in the trade-off between ontological economy and explanatory economy.

Setting aside the problems of Rescher's suggestions that I highlight in my 2005 (16–25), (4) can be immediately ruled out for my task in hand since a naturalistic explanation of phenomena is incompatible with a realist outlook of theology. (1) should also be excluded because the benefit of obtaining the description about the physical world in its phenomenal and extra-phenomenal domains is not an ‘indirect’ benefit that will be gained whether or not the goal is realised/approximated. Rather, whether or not this benefit is gained is directly related to whether the goal is realised or approximated, for if the goal is not realised or approximated, the information about the physical world it provides will have been false. Therefore, this cannot be accepted as an indirect benefit for the pragmatic justification of the realist aim of science, and it follows that a theological analogue should not be sought. (2) seems to be out of place because prediction and control have no close parallels in theology. But perhaps there is a loose parallel in a theology that is associated with a functional view of religion, which sees the crucial function of religious traditions as ‘providing a guiding vision which shapes our way of life’ (Drees 1996, 279). Though this is far from paralleling prediction and control, the idea of a theological theory exerting a certain degree of impact on individuals and communities in the functional view of religion has a loose analogy with the idea of a scientific theory exerting control over nature in science. Of course, a crucial disanalogy lies in the fact that there is a close responsiveness between nature and the scientific theory that allows one to exert control over nature, while such a close responsiveness does not exist between the theological theory that exerts a certain degree of impact on individuals or a community and these people. Recognising this disanalogy, the PTRist may nonetheless still argue that holding a realist aim is more conducive to the implementation of the functions or tasks of a theology that is associated with a functional view of religion. Finally, (3) and (5) have close parallels in theology.

Therefore, the PTRist may suggest that holding the realist aim in theology may provide the indirect benefits of (1) giving a ‘fair trial’ to observation-transcending reality by allowing explanation of the phenomenal in terms of the extra-phenomenal; (2) providing a more conducive ‘environment’ (in a metaphorical sense) for the implementation of the functions of a theology associated with a functional view of religion; and (3) maximal gain and minimal loss in the trade-off between ontological economy and explanatory economy. Next, I shall take issue with each of these purported indirect benefits of holding the realist aim of theology.

Fair Trial to Theorising about Observation-Transcending Reality

A PTRist may contend that only a realist aim can give a fair-trial to theorising about observation-transcending reality by allowing explanation of observable phenomena in terms of unobservables. Take the example of the origin of the universe. The orderliness of the universe is the observable phenomenon. Theists infer from the orderliness of the world to the existence of a divine creator as the cause of this orderliness, while naturalists explain this orderliness in entirely naturalistic terms. A PTR theist postulates God and attributes the orderliness of the universe to God in her theological theory. The PTRist may accuse the naturalist of not giving a fair-trial to observation-transcending reality, and may even borrow Rescher's critique: ‘[a] general epistemic policy which would make it impossible, as a matter of principle, for us to discover something which may (for all we know) responsibly be supposed actually to be the case is clearly irrational’ (Rescher 1987, 45).

TCE, however, is not susceptible to this charge, for it allows the postulation of an unobservable theoretical entity to explain the orderliness of the universe. Therefore, holding PTR's realist aim does not give us advantage over TCE, since TCE also gives a fair trial to observation-transcending reality by allowing the explanation of observable phenomena in terms of unobservables. The intelligent design theory (ID) may be a case in point. ID theorists postulate an unobservable theoretical entity called intelligent designer to account for the origin of the universe and the appearance or indication of design observed in the biological world. This intelligent designer could be God or an intelligent alien from another planet or even a demon. According to TCE, ID theory can be accepted for the purpose of explanation as long as one does not believe its truth. However, this does not mean that TCE will necessarily be favourable to ID, not to mention endorsing it, for van Fraassen (1980, 24) avers that it is not always necessary to explain the observable phenomena in terms of the unobservables unless doing so can (1) lead to gains in predictions and (2) account for wider empirical regularities (pp33, 34). Therefore, ID theory may still lose out to naturalistic theories on this basis. But it is at least open to it, and is willing to give it, in Rescher's terms, a fair trial. Given van Fraassen's two caveats, there is no need to worry that TCE will lead to the acceptance of many theological theories in place of naturalistic theories.

One might question the propriety of accepting these two caveats of van Fraassen's and argue that they seem to stem from an empiricist bias that focuses on empirical gains. In this paper I shall remain neutral about whether this emphasis on empirical gains per se is justified, but the empiricist focus is entirely appropriate within the context of the debate of scientific realism because disputants on both sides take the focus on empirical gains equally seriously. Even a scientific realist will frown upon a PTRist's attempt to postulate an unobservable entity to explain observable phenomena when such a postulation does not bring about empirical gains. While one might complain that such an empiricist emphasis puts theology at a disadvantage, this is the price that theologians such as van Huyssteen have to pay for modelling their defence of theological realism on scientific realism.

A PTRist might argue that TCE's focus on the truth about the observables makes it impossible to discover the truth about God if it turns out that God really exists. TCE asks us to focus on empirical adequacy whilst being agnostic about whether a theory is descriptively true, which implies a kind of openness to the possibility that the theory might be descriptively true. But a PTRist may charge that, notwithstanding this openness, the focus on empirical adequacy makes us inattentive to when the theory is descriptively true. But can PTR do better? According to PTR's rejection of convergentism, it is highly improbable that theology ever attains or approximates the truth. Those who believe in this are also inattentive to when the theory is descriptively true. In short, PTR's realist aim does not give us an advantage over TCE vis-à-vis a fair trial to theorising about observing-transcending reality.

Conduciveness to a Theology Associated with a Functional View of Religion

Whereas Andrew Moore (2003, 56) thinks that the functional view of religion is an argument against theological realism, the PTRist may argue that holding a realist aim is more conducive to the implementation of the functions or tasks of a theology that is associated with a functional view of religion. According to the functional view, religion serves to provide action-guiding visions, motivational power and pedagogical force in the ethical, familial and social aspects of human life. While such a view does not necessarily entail a denial of the ontological aspects of religious claims, the chief focus is their functional efficacy.

PTRists may suggest that holding a realist aim of theology is more conducive to the effective implementation of these functions in a religious community. Given that theologising is a crucial element in the formation of the action-guiding visions and teaching in a religious community, holding the realist aim may provide the theologians within the community with the indispensable motivational power for their theological endeavours, thereby contributing to the better implementation of the functions of religion. By holding the realist aim, the theologians will think that their theological activity is in the business of truth, and this will provide a much more powerful motivational force than the belief that they are engaging with ‘useful fictions’. (Note that what seeks justification here is the action or choice of holding the realist aim rather than the belief in the realist aim.)

This is a benefit of holding the realist aim that is subjectivistic, individual-, cultural-, community, and time-relative. (Note that this is a subjective benefit vis-à-vis the better implementation of the functions of theology and should not be confused with subjective non-cognitive benefits that are not related to the effective implementation of the tasks of theology, e.g. fame and power of the theologians.) It is not a matter of principle, and thus cannot be determined by conceptual analysis or even empirical studies. Even if an empirical study were to suggest that holding the realist aim was more conducive to the work of a community of theologians, it would be doubtful if this result could be generalised to other communities, time, locations and cultures.

However, some may not be satisfied with the above response and may argue that unless the working theologians believe that they are in the business of truth, they are unlikely to find sufficient motivational power to sustain them through. In other words, if one holds the realist aim of theology, she will or will be more likely to believe that theology is in the business of truth and such a belief can provide the indispensable motivational power. Believing that theology is ‘in the business of truth’ can be interpreted as believing that theology is an inquiry that aims at the truth or as believing that theology deals with (at least approximate) truths in its current state. But holding the realist aim within the PTR framework cannot provide the motivational power. PTR's claim that the realist aim of theology is unreachable in virtue of its rejection of convergentism will offset the potential motivational power of the belief that theology aims at the truth. Moreover, PTR's rejection of convergentism and its Kuhnian view of theological theory change do not allow one to believe that theology deals with (at least approximate) truths in its current state.

Setting aside these problems, even if we concede that holding the realist aim can provide the motivational power, TCE can also provide the same motivational power, because van Fraassen's notion of immersion could be applied to TCE. Immersion in the theological and religious world within the TCE framework is a mental state that is so close to belief that it is able to provide the psychological effect that belief can offer. Theologians who are immersed in the worlds of their theories can gain the psychological power equivalent to that of believing in the truth of these theories though they do not in fact believe it. Some may think that such blurring of the distinction between immersion and belief is a sleight of hand that sneaks belief back into TCE. While the psychological states of belief and immersion are difficult to distinguish, this does not at all imply that a clear conceptual distinction cannot be made between belief and immersion, and a conceptual distinction is all it takes to differentiate between them.

Between Ontological Economy and Explanatory Economy

PTR has an advantage over naturalism because naturalism pursues ontological economy at the expense of explanatory economy while PTR strikes a balance between ontological economy and explanatory economy by allowing the postulation of unobservable theoretical entities for explanation. However, PTR has no advantage over TCE because the latter does not proscribe the postulation of unobservable theoretical entities for explanation.

With this I have made a case for the claim that PTR has no advantage over TCE vis-à-vis indirect benefits.

PTR is Riskier than TCE

In order to show why PTR is riskier than TCE, I shall develop an argument that closely parallels my corresponding argument in the case of science – that PSR is riskier than CE (Lee 2005, 28–9). Alan Musgrave (1985, 199) contends that both the realist and the constructive empiricist face the same risk vis-à-vis the sceptical challenge. Both the realist's belief in the truth of a scientific theory and the constructive empiricist's belief in the empirical adequacy of a scientific theory can turn out to be wrong. The same could be said of theology – theological realists and theological constructive empiricists face the same risk. However, the realist faces two additional risks. If a theory that postulates God to explain the orderliness of the universe – the God theory – turns out to be false, the realist will have believed in falsehood, whilst the constructive empiricist cannot be said to believe in falsehood because she only accepts the God theory without believing in it. Furthermore, if the God theory turns out to be false, the realist will have made a false assertion about the existence of God, while the constructive empiricist cannot be said to make a false assertion because she does not assert that God exists. Therefore, a realist incurs the risks of believing in falsehood and making false assertions.

But what if the God theory turns out to be true? The constructive empiricist will have failed to believe in the truth and failed to make a true assertion. Therefore, the realist contends that her own risk is offset by the constructive empiricist's risk –‘I risk believing in falsehood and making false assertions but you risk not believing in the truth and not making true assertions. So we break even’.

But PTR can still be shown to be riskier than TCE in another sense in virtue of its rejection of convergentism and its adoption of the Kuhnian view of theological theory change. According to PTR, the probability of a theory's being true is very low.

  • p=the probability that a theological theory is true
  • 1−p=the probability that a theological theory is false
  • p<1−p

Since the PTRist suffers loss when the theory is false (believing in falsehood and making a false assertion), and the TCEist suffers loss when the theory is true (failure to believe in the truth and make a true assertion),

  • p=the probability of the TCEist's suffering loss
  • 1−p=the probability of the PTRist's suffering loss
  • since p<1−p
    the probability of the PTRist's suffering loss>the probability of the TCEist's suffering loss

In this sense, PTR is still riskier than TCE.

In other words, PTR and TCE face the same risk in relation to the sceptical challenge, since the PTRist's belief in a theological theory's truth and the TCEist's belief in its empirical adequacy can be wrong. But the PTRist has two further risks of believing in falsehood and making a false assertion if the theory turns out to be false. The TCEist can avoid these two risks in virtue of its agnostic attitude to the truth of the theory and the existence of the theoretical entity it postulates. However, the TCEist faces two risks that can be avoided by the PTRist: if the theory is true, the TCEist will have failed to believe in the truth and to make a true assertion. Nonetheless, PTR is still riskier than TCE in this sense: since the probability of a theory's being false is higher than its being true given its inability to approximate the truth, the probability of a PTRist's suffering loss will be higher than a TCEist's suffering loss because the PTRist suffers loss when the theory is false and the TCEist suffers loss when the theory is true.

With this I have made a case for the prudential argument against PTR.

CONCLUSION

PTR urges us to take up the realist aim or the goal of truth although we have good reason to think that the goal can neither be attained nor approximated. While Newton-Smith thinks that pursuing what we know we cannot achieve is clearly irrational, Rescher disagrees and contends that pursuing an unreachable goal can be rational on pragmatic grounds so long as pursuing the unreachable goal yields indirect benefits. I have blocked this attempt at providing pragmatic justifications for the realist aim of PTR on precisely the same pragmatic grounds: since there is a competing alternative to PTR and the alternative can provide whatever indirect benefits PTR can offer while being less risky than it is, prudential reasoning favours the alternative to PTR. This undermines the pragmatic case for the realist aim of theology since the instrumentalist alternative does not aim at the truth. In short, I have presented a pragmatic argument against a pragmatic form of theological realism.

My conclusion that pragmatic considerations favour TCE over PTR should not be taken to imply that I accept TCE. While the merits of TCE compared with those of PTR on some of the premises of van Huyssteen's have been discussed here, its intrinsic merits and its merits in relation to other rivals have not been dealt with here. In fact, I realise that to some TCE per se is perhaps a rather unsatisfactory view of theology. Although it allows the postulation of unobservable entities for the purpose of explanation in principle, in practice it will still yield many grounds to naturalistic explanations given van Fraassen's two caveats. But since it is less unsatisfactory than PTR (and thus the lesser of two ‘evils’ to be chosen by prudential spirit) and my purpose here is not to propose a satisfactory philosophical view of theology, but to argue against PTR on pragmatic grounds, I believe my appeal to TCE despite its perceived unsatisfactoriness is appropriate to the task in hand.

Ancillary