Worlds Apart: The Logic of Conversion in Wittgenstein’s On Certainty

Abstract

This paper evaluates Pritchard’s account of religious conversion as a quantitative process along which gradual shifts in our system of empirical beliefs lead to a new, holistic view of the world. Wittgenstein’s own presentations of transition, including religious transitions, help us to challenge Pritchard’s conception of conversion. It is shown that Wittgenstein was aiming at doing justice to the tension between continuity and discontinuity which is intrinsic to cases of transformative change. The main thrust of Wittgenstein’s view of conversion in On Certainty departs from a gestalt shift model that is complemented by a hermeneutical, imaginative model of reasoning. For this reason, we contend that hinge epistemologists are mainly blind to informal modes of reasoning and apprehension which are neither logical nor grounded in evidence. This, in turn, is a reminder of how Wittgenstein called attention to the importance of the imagination in the process of conversion, and of his remarkable appreciation throughout On Certainty of imaginative activity for genuine understanding. Wittgenstein’s logic of ‘internal understanding’ clearly challenges some widely assumed representations of Wittgenstein’s last notes.

1. Introduction

One way of approaching Wittgenstein’s thought in On Certainty is suggested by the discussions initiated by Duncan Pritchard’s quasi-fideism. Central to Pritchard’s position is the application of the Wittgensteinian account of hinge commitments to the religious case (see Pritchard, [1,2,3,4]). Pritchard’s proposal involves three aspects. First, there is an interpretation of On Certainty. Second, there is a version of hinge epistemology that is grounded in this interpretation. Third, there is the quasi-fideism position that as an account of the epistemology of religious belief, Pritchard understands as grounded in his rendering of hinge epistemology. Although these three aspects are all intertwined in Pritchard’s view, it is nonetheless important to be aware of the fact that, in a certain sense, each aspect could stand or fall independently of the others. Pritchard might be called to task for failing to provide a plausible interpretation of On Certainty, even if his hinge epistemology stands on its own as a proposal about the arational framework of rational evaluation¹. Or perhaps his rendering of On Certainty is correct but his version of hinge epistemology is theoretically defective. Or maybe quasi-fideism and the parity argument that is central to it is defensible but it is not grounded in hinge epistemology².

The fact is that Pritchard’s quasi-fideism can be challenged on two fronts. First, it may be argued that this position relies on a contentious reading of On Certainty, making it questionable whether it is a correct application of Wittgenstein’s insights to religious epistemology. Second, it may be contended that Pritchard’s project of defending the rationality of religious beliefs by means of a parity argument which is grounded in the arational nature of hinge commitments, whether quotidian or religious, collapses into full-fledged fideism and traditional irrationalism, and so, that it fails to provide any legitimation, or positive defense, of religious beliefs. After all, the danger for quasi-fideism lies in its proximity to fideism (Schönbaumsfeld, [9], p. 24), and also, in that if fideism entails a total rejection of reason, and no total rejection of reason can itself be argued to be reasonable, fideism is tantamount to irrationalism³. As it is well known, Annalisa Coliva has recently attacked Pritchard’s quasi-fideism on the interpretative and on the epistemological fronts, arguing that it is an inherently unstable position (Coliva, [7], p. 2). Pritchard’s opponents do not deny that there are such things as hinge beliefs, but rather deny that belief in God is one of those properly basic beliefs (Coliva, [7], pp. 9–11).

In this paper, we challenge Pritchard’s account of religious conversion, as conversion is crucial to understanding how generic hinge commitments and religious hinge commitments relate⁴. However, our challenge should be prefaced by some words of caution.

First, Pritchard deals with both the process of acquiring religious faith (positive conversion) and of losing it (negative conversion) with the aim to oppose what he sees as a potential disanalogy between fundamental religious convictions and mundane hinge commitments. The disanalogy lies in that, unlike common conceptions of transitions from religious non-belief to religious belief (or vice versa), changes in one’s hinge commitments do not seem to be the result of an all-or-nothing, qualitative change. This is why Pritchard is adamant both in rejecting instantaneous conversion experiences (religious epiphanies) and in redescribing change in one’s religious commitments as a quantitative process along which, like on his view it happens with transitions to alternative frameworks in general, gradual shifts in our system of empirical beliefs lead to a new, holistic view of the world. Under Pritchard’s description, religious epiphanies would be transformative experiences where the shift in one’s religious hinge commitments is not gradual but immediate.

Though we agree with Pritchard in that conversion is a function of what precedes it, what is important and correct in Pritchard’s understanding of the gradual and indirect process that leads to conversion is misrepresented because he uses the same word (‘instantaneous’) for quite different kinds of events—something ineffable and sudden which simply happens to us (grace); and a creative and qualitative culmination which is irreducible to its antecedents. It is small wonder that only one (false) option is left open for Pritchard: the description of conversion as the result of a cumulative, mechanical process.

We will claim that this view throws a false light both on the transition (or leap) in a conversion experience and on transitions between irreducible worldviews⁵. In what follows we suggest how the concept of a Gestalt shift, that is, of a qualitative change which occurs when a ‘critical threshold’ has been reached, can be useful for interpreting the character of transitions in general, and so, to provide one perspective on religious conversion⁶. Like the gestalt shift, conversion must occur all at once (though not necessarily in an instant) or not at all. Where conversion involves a qualitative reorganization of a person’s picture of the world, it must be understood as a unitary transformation rather than as the mere sum of gradual changes in belief, although this does not require that the transition be instantaneous or temporally isolable. As will be argued, the gestalt model embodies not only the perspective of discontinuity that Pritchard overlooks in his theory of conversion, but also and at the same time the perspective of continuity that he wants to emphasize. Gradual changes in the system of beliefs may provide motives for conversion, but they do not constitute conversion ([15], OC § 612).

Secondly, the model of gestalt shifts that we propose involves a parity argument that is almost an inverted version of the kind of parity argument put forward by Pritchard (see, for example, Pritchard [2])⁷. Pritchard argues for the legitimacy of religious conversion on the grounds that it is a gradual process on an epistemic par with other forms of change in one’s hinge beliefs. In contrast, the gestalt model argues that it cannot be thought to be a problem with religious conversion that it involves the emergence of a qualitatively different awareness at a critical threshold given that all transitions between worldviews have gestalt shifts at their heart, and that the heuristic tool of gestalt shift is required to explain such a change. Additionally, the discernment of a different pattern necessarily involved in gestalt shifts suggests a broadened concept of ‘reasoning’ and ‘rational’, as well as ties between these concepts and imaginative activity. The ‘narrow’ conception of rationality that quasi-fideism shares with its opponents is bound to lead to distortion and misdescription of religious reasoning in one way or another. Either religious conversion will be misshapen into the gradual or cumulative product of piecemeal changes in empirical beliefs, or it will be misrepresented as not resulting from a genuine mode of reasoning, and will accordingly be rejected altogether, because it fails to fit into the mould of rationality. There is a tendency in hinge epistemology to presuppose that the intellectual practices designated as ‘reasonable’ in our scientific society constitute the normative concept of ‘reason’, and so, there is in hinge epistemology in general a blindness of sorts to the possibility of alternative ways of thinking that are in fact truth-conductive.

Finally, though it is true that in his works Wittgenstein hardly ever discussed the question of the nature of reason, this question comes to the surface in his reflections on religious belief. However, the problem lies in the fact that religious belief is hardly discussed in On Certainty. It is not only that references to the religious language are almost lacking. It is rather that concepts usually related to religious transitions such as ‘conversion’ ([15], OC § 92) and ‘persuasion’ ([15], OC §§ 262, 609) appear to be in this context descriptions of transitions in general. This explains why Pritchard conceives quasi-fideism as an importation to the religious case of some of OC’s insights, and also, why he is careful to distinguish Wittgenstein’s fideistic view of religious belief as it is expressed in CV and LC (see Pritchard, [16]) from quasi-fideism, which is a position merely inspired by Wittgenstein’s last remarks.

In contrast, we will argue that in OC Wittgenstein raises crucial questions about religious belief, and so, about the philosophical problem of how faith relates to reason. For one, beliefs about the character of the universe and the place of human being in it—what Wittgenstein calls “a picture of the world” ([15], OC § 93)—are not religiously idle (see Pignocchi, [14]). Additionally, Wittgenstein is emphatic on the fact that whenever one recognizes that something ought to be believed, whenever one feels certain, one’s state of mind is not purely intellectual ([15], OC § 117), and so, that even those beliefs which are concerned with the establishment of matters of fact have a reference to a norm or value ([15], OC § 493). This reference leaves room for the introduction at the heart of mundane beliefs of axiological hinges, and so, for the possibility that the belief in the significance of moral obligation to the truth may be considered an act of generic faith. Thus, quasi-fideism illustrates the difficulty of doing justice to the complexity of OC without providing an analysis of religious belief. We aim at making some neglected aspects of Wittgenstein’s last remarks explicit. The perspective of gestalt shift can thus be useful in highlighting these possibilities for interpreting the various Wittgensteinian presentations of transition, including religious transitions.

However, the main stimulus for this article comes from the limits of the gestalt shift model in illuminating transitions to radically alternative worldviews—for example, that in cases of gestalt shift which are more subtle than that of the duck/rabbit figure, one cannot shift back and forth at will between the construals of the figures⁸. The upshot is that the gestaltic model of conversion needs to be qualified and complemented by a hermeneutical model that does not carry with it the same freedom between the two equivalent, symmetrical and reversible perspectives that we find in visual gestalt shifts. Reorientation, cognitive advance, and decisiveness are thus crucial to understanding conversion. This, in turn, is a reminder of the cognitive dimension of the imagination, and of the fact that the imagination is a genuine creative centre of meaning. One can see in Wittgenstein’s writings a striking preoccupation with imagination which expresses itself throughout OC in a remarkable appreciation of imaginative activity for genuine understanding⁹. It is this appreciation, of which epistemological readings of OC are usually silent, which most warrants a reconsideration of Wittgenstein’s last notes in our day.

The foregoing reflections make for a complex and ambitious argument, but we believe there is no other way to appreciate the scope and depth of the controversy. If the following proposal holds water, we may come to appreciate that, even where particular religious and non-religious orientations are ‘worlds apart’, there are grounds for constraint that prevent the distinctiveness of religious commitment from implying that ‘anything goes’. The argument concerns transitions between particular pictures of the world, not a general opposition between religion and secularity.

Our argument requires an interpretive commitment concerning the relation between Wittgenstein’s early and later writings. We do not assume that his philosophical vocabulary remains invariant across periods, nor that the later Wittgenstein can be read as simply restating conclusions already reached in the Notebooks or the Tractatus. Rather, we propose that certain structural problems persist across changes in philosophical method, even when the conceptual resources through which they are addressed undergo substantial revision. In particular, the relation between factual description and the significance of the world as a whole, together with the possibility of a transformation in how things are seen, provides a point of comparison between the early reflections on the world’s waxing and waning “as a whole” ([22], NB 5.7.16; [23], TLP 6.43) and the later discussions of aspect-perception, persuasion, and conversion in On Certainty ([15], OC §§ 92, 609–612). Campbell’s account of the relation between ethics, aspect-perception, and the unity of the Tractatus offers a relevant precedent for investigating these connections without reducing them to a merely verbal continuity ([24], p. 33). It does not, by itself, establish continuity across Wittgenstein’s entire philosophical development; that stronger claim would require a separate argument. Our present claim is consequently more limited: the earlier texts illuminate a structural problem that reappears, in a transformed conceptual setting, in the later discussion of how a person may come to see the world differently. If this continuity cannot be sustained, the argument concerning conversion in On Certainty would not thereby collapse in its entirety. Its central claims about imaginative understanding, the limits of formal justification, and the transformation of a person’s orientation would still have to be assessed on their own textual and philosophical merits. What would be weakened, however, is the further suggestion that the early and later texts articulate related expressions of a single underlying problem. We therefore treat continuity as a defeasible interpretive hypothesis, rather than as an assumption that renders differences between periods irrelevant.

2. Religious Conversion as an Object of Comparison

A passage from OC makes it clear that Wittgenstein thought that controversies where different pictures of the world are involved are not like differences in opinion, and that transitions between irreducible worldviews may be better understood in terms of an analogy with conversion:

Men have believed that they could make rain; why should not a king be brought up in the belief that the world began with him? And if Moore and this king were to meet and discuss, could Moore really prove his belief to be the right one? I do not say that Moore could not convert the king to his view, but it would be a conversion of a special kind; the king would be brought to look at the world in a different way.

([15], OC § 92)

We take it that, in speaking this way, Wittgenstein is making at least four claims that are closely connected: (i) Wittgenstein’s words appear to imply that there is no shared content between the beliefs of Moore and the beliefs of the king, and so, that in an empirical sense, they do not contradict each other (they are ‘worlds apart’). (ii) Wittgenstein appears to be also contending that how one perceives certain empirical facts is itself already shaped by one’s picture of the world¹⁰, and thus, it appears that he is making the notion of ‘neutral evidence’ questionable. (iii) Wittgenstein is also suggesting that there is a logical disproportionality (and discontinuity) between the reasons on which Moore’s worldview is grounded and his assurance that the scientific picture of the world is correct. (iv) Finally, Wittgenstein makes clear that conversion stands for a transforming shift in perspective that involves seeing the world anew, to wit, that it stands for new vision and conceptual reorientation¹¹. It is therefore as if in conversion the world one had been inhabiting shifted and rearranged itself along a new axis. In conversion, we are taught a new way of seeing and making sense of the world; conversion is a qualitative different awareness, and also, a non-mechanical extension beyond what it is actually given.

The most relevant aspect of Wittgenstein’s remark is, however, that Wittgenstein is actually employing the concept of ‘religious conversion’ to illuminate the question of change between incommensurable frameworks, a point which is reinforced by Wittgenstein’s own reference, in his discussion about the limits of reasoning and the role played by persuasion ([15], OC §§ 609–612), to “what happens when missionaries convert natives” ([15], OC § 612).

Pignocchi’s recent account of deep disagreement and conversion ([14]) provides an important point of comparison for the interpretation developed here. Pignocchi argues that disagreement between believers and non-believers cannot always be resolved by exchanging explanations or adducing considerations that both parties already recognize as reasons. Between initial incompatibility and possible conversion, he identifies an intermediate stage of partial emotive understanding, in which each party may come to appreciate something of the other’s perspective without yet adopting it. Conversion becomes possible when a blockage that sustains the former orientation is overcome. In Pignocchi’s account, the decisive role belongs to the will: the transformation of one’s way of seeing things cannot be secured by an appeal to reasons or rationality alone. This analysis is close to our own in its rejection of a purely cumulative model of conversion, its emphasis on affective engagement, and its recognition that a change in perspective can transform the significance of what remains factually unchanged. The disagreement between our accounts therefore concerns neither the existence of discontinuity nor the importance of emotion. It concerns the explanatory status of the will and the possibility of understanding the transition as an imaginative achievement.

Our proposal differs from Pignocchi’s insofar as it denies that the decisive transition is adequately explained by assigning the overcoming of a blockage to the will. The will may be involved in sustaining attention, remaining open to another perspective, or refusing to retreat from the demands of an emerging understanding. Yet these activities do not, by themselves, explain how the alternative perspective becomes intelligible or comes to reorganize the significance of the subject’s experience. An act of will can help make a transition possible without constituting the transition itself. To say that someone wills a new way of seeing does not yet explain what she comes to see, why this new configuration makes sense to her, or how it acquires the power to orient her life.

This role of the will finds support in Wittgenstein’s remarks on religious commitment. Wittgenstein describes faith as a passionate commitment to a system of reference ([25], CV p. 73) and speaks of being ‘seized and turned around’ by something, after which one must ‘stay turned round’ ([25], CV p. 61). These passages emphasize the will’s part in sustaining and stabilizing a transformed orientation. They do not imply that the will produces the transformation: the image of being seized and turned around suggests something received rather than simply decided, which is what our account of the imagination’s receptive dimension is meant to capture.

The alternative developed here gives the imagination a constitutive role in this process. Imagination is not merely a faculty that supplies vivid representations to a decision already made, nor is it an irrational force that substitutes for reasons. It is the activity through which relations, possibilities, and meanings that were previously unavailable within the subject’s operative perspective can become apprehensible. This does not mean that the subject freely invents a new world or chooses between two perspectives as one might select between equivalent visual figures. Rather, imaginative understanding allows another configuration of significance to become intelligible and potentially compelling before it has been fully appropriated as the subject’s own. The transition can therefore be both free and non-voluntaristic: it is not mechanically compelled by evidence, yet neither is it adequately described as an arbitrary decision. The will may condition or accompany the transition, while the imaginative reorganization of meaning accounts for its distinctive cognitive content.

This disagreement also clarifies the scope of our appeal to On Certainty. Pignocchi’s primary textual focus is the disagreement between believers and non-believers as presented in the Lectures on Religious Belief [26], whereas our argument concerns the use of conversion in On Certainty as a model for understanding transitions between pictures of the world. We do not take Pignocchi to be committed to the claim that the imaginative resolution of deep disagreement is primarily located in On Certainty. Rather, his account supplies a closely related interpretation against which the present proposal can be sharpened. Our claim is that the later remarks on conversion, persuasion, and looking at the world differently can be read as leaving room for an imaginative mode of understanding that explains more than the overcoming of a volitional blockage: it explains how a new orientation becomes meaningful to the person undergoing the transition.

Why is it that Wittgenstein thinks that distinctively religious conversions provide a fruitful vantage-point for assessing transitions in general? The answer does not only lie in that there are close similitudes between Wittgenstein’s description of general transitions and religious conversion, but rather in that, for all its stubborn resistance to attempts to subject it to any rationalist account, religious faith makes it visible, in an eminent sense, that irreducible pictures of the world have no common measure that allows them to be joined together. Some very Kierkegaardian and Newmanian aspects¹² in Wittgenstein’s conception of religious belief might throw light on what Wittgenstein might have in mind here.

To begin with, it is important to notice that logical disproportionality is particularly apparent in the case of religious faith, in that faith demands a kind of certainty and unconditional assent that cannot be provided by reasons. This is a crucial insight in John Henry Newman’s account of assent in the Grammar of Assent [34]. Newman’s notorious rejection of evidentialism is intrinsically related to his claim of the independence of assent and certitude from inference and his affirmation of the unconditionality of conviction ([34], pp. 160, 169)¹³. There is thus an unbridgeable gap between the qualitative character of the infinite interest that religious belief calls for and the quantitative character of the evidence on which is based. Wittgenstein took careful notice of this enormous disproportion ([26], LC pp. 57–58), and of the paradox that lies at its heart¹⁴.

Additionally, if one attempts to base religious conversion in rational deliberation and probability, then it would seem that conversion is the result of a decision that is made only tentatively and is subject to revision at any moment. Yet religious belief, according to Wittgenstein ([26], LC p. 57), calls for a decisive commitment and a radical self-transformation. Otherwise, one would treat religious beliefs no differently than ordinary empirical propositions, and religion “as a matter of reasonability” ([26], LC p. 58), something that “would in fact destroy the whole business” ([26], LC p. 56). There is, indeed, something ludicrous for the believer to be religious only up to a certain point.

Thirdly, there is a glaring asymmetry between ordinary beliefs and religious beliefs in that, for example, the question of God’s reality is a question about the character of the whole reality, and not about the existence of ‘one more thing’¹⁵. Religious beliefs are, in fact, linked to claims about the value and the meaning of the whole of reality, for the facts remain the same, whether there is a God or not. It is clear that this holistic and axiological aspect is not exclusive of the religious conviction. On the contrary, it is also of the nature of any generic Weltbild ([15], OC §§ 93–95).

Finally, let us mention that ‘belief’ in religious contexts is not a bare belief, but rather it is a thick concept which describes a state of mind that is not purely intellectual, namely, that is not a mere theoretical approval of propositions. At the very least, conversion involves an immediate feeling of conviction or approval, and ‘feelings’, in so far as they are emotions, are serious and concern-based (for example, one must be concerned about being in the condition of conversion one has come to be in). In addition, religious beliefs can never be fully divorced from ethical transformation and a new way of life. The crucial point is that we cannot sharply separate the process by which we come to have religious beliefs from the process by which we try to become transformed by the image of an ideal or better self. Therefore, religious conversion is not just a matter of making some intellectual adjustments in one’s system of beliefs, but rather it is more like assuming the task of self-actualization, and so, it is a (positive) response to a tempting possibility of self-redescription which is recognized as the demand of becoming oneself in the condition one construes oneself to be in.

In short, the whole person cannot be discounted in the process that leads to conversion, which is why, as it was previously remarked, Wittgenstein says that for conversion “you have to be seized and turned around” and, following Kierkegaard, calls faith “a passion” ([25], CV p. 61)¹⁶. Of course, that conversion is irreducibly first-personal has several implications. Though in passing, we noted one of these implications above, which is that conversion does not involve evidence that is available to everyone. On the contrary, religious conversion comes with a heightening sensitivity and the development of a capacity to see things in a particular way. Conversion may be thus characterized as a transforming vision.

The problem lies in that, even if one conceded that religious conversion throws light on transformative experiences in human life, one could still argue that irrationality is not redeemed by means of being extended. Transitions are traversed by paradox in multiple ways, to the extent of being impossible to be fully understood without it.

The problem is now how to reconcile the two opposites in tension on which the whole of Wittgenstein’s description of conversions rest: the epistemic discontinuity between competing pictures of the world, and the dimension of continuity with other human experiences which is required to do justice to the personal activity involved in the process that culminates in conversion. The process leads to a qualitative different awareness (in this sense the transition is discontinuous with what precedes it), but it cannot occur in a vacuum, and therefore, it is a function of preliminary material (in this sense the transition is continuous with what precedes it). Thus, conversion must be distinguishable from the gradual, quantitative process to which Pritchard reduces it as well as from separable fiats (so-called ‘acts of the will’, or an act of God’s grace) that would ground the claim that the acquisition of faith is ineffable.

The distinction between the gradual process that prepares a conversion and the qualitative transition that constitutes it requires further clarification. Noaparast’s discussion of the emotional problem of evil is particularly relevant here ([36]). When a person undergoes suffering that threatens her religious commitments, the ensuing crisis need not consist in a single decisive experience. She may repeatedly reinterpret what has happened, consider different explanations, reconsider her conception of divine goodness, and attempt to integrate her experience into a more comprehensive understanding of her life. Such interpretive work may unfold over an extended period. It can modify the person’s expectations, affective responses, and understanding of what it means to believe in God. If conversion is understood as a transformation of one’s orientation towards the world, it would be implausible to deny that such gradual processes can contribute to, and sometimes constitute part of, the transformation.

The relevant question, therefore, is not whether conversion has a history, but whether the history of a conversion exhausts what conversion is. We distinguish between the gradual elaboration of an experience and the reorganization of the perspective within which that experience becomes meaningful. During the former, the subject may acquire new information, revise particular judgments, or consider alternative explanations while retaining the basic criteria through which those judgments and explanations are assessed. In the latter, what changes is not merely the content of some beliefs but the organization of significance through which the subject understands herself, her experience, and the possibilities open to her. A person may spend months attempting to understand her suffering without undergoing a conversion; alternatively, the same interpretive process may culminate in a transformation through which the meaning of that suffering, and the place it occupies in her life, is reorganized. The difference is not necessarily observable as a discrete temporal boundary. It is a difference in the structure and function of the resulting understanding.

This distinction is intended to preserve the thesis that conversion is a qualitative transition without identifying qualitative change with instantaneous change. When we claim that conversion occurs ‘all at once’, we mean that the transformation, insofar as it constitutes a conversion, is not adequately described as the mere sum of the preceding modifications. We do not mean that it must occur at a single, temporally isolable instant, without preparation, hesitation, or subsequent development. A threshold may be reached through a process whose duration is indeterminate, and its crossing may itself be gradual in the phenomenological sense that the subject cannot identify a precise moment at which the new orientation became operative. What matters is that the transition establishes a new organization of significance that cannot be captured simply by enumerating the incremental changes that preceded it.

The distinction between gradual development and qualitative conversion cannot be established by reference to the duration of the process alone. What matters is whether the changes reorganize the practical and interpretive relations through which the subject understands her experience. Relevant indications include a transformation in what the subject takes to be salient, in the connections she recognizes between previously familiar experiences, and in the commitments that orient her subsequent judgments and actions. None of these indications is individually sufficient, and they need not emerge simultaneously. Together, however, they provide grounds for distinguishing a reorganization of the subject’s orientation from the cumulative revision of particular beliefs. The distinction is therefore an interpretive one, answerable to the details of particular cases, rather than a temporal boundary that must be independently identifiable in every conversion.

The distinction also clarifies the relation between conversion and the subsequent interpretation of one’s life. A transformation may acquire its full articulation only retrospectively. After a change in orientation, the subject may continue to reinterpret the suffering, doubts, and experiences that preceded it; she may also revise her understanding of the commitments she has adopted. This continuing interpretation does not show that the original transition never occurred. Nor does it imply that conversion is an accomplished event wholly detached from the life that follows. Rather, the transition and its subsequent appropriation belong to different but connected dimensions of the process. Conversion marks a reorganization of the subject’s orientation, while the practical and interpretive consequences of that reorganization continue to develop.

Noaparast does not offer a theory of conversion; he expressly sets aside the question of how people in fact come to acquire or lose religious belief ([36]). His description of the emotional problem of evil nevertheless shows that a person may reach or abandon a religious commitment only after repeated attempts to interpret and reinterpret her suffering. If so, we cannot treat all gradual interpretive change as merely external preparation for a conversion. Our claim should consequently be restricted to cases in which conversion names a qualitative reorganization of the person’s picture of the world, rather than every instance of religious change. We do not seek to establish that every loss or acquisition of religious belief has the same temporal structure. We argue instead that, where conversion involves a transformation of the subject’s overall orientation, a distinction remains necessary between the gradual history that makes the transformation intelligible and the qualitative change in the organization of meaning that constitutes it. Whether a particular case exhibits such a threshold is a substantive interpretive question, not something that can be settled by definition. This qualification matters for cases in which the subject’s transformation remains ambiguous even in retrospect. Our account does not require that every extended crisis culminate in a conversion, or that every conversion admit of a precise point of transition. It requires only that, where a conversion is plausibly identified, there be grounds for distinguishing a reorganization of the subject’s practical and interpretive orientation from the cumulative revision of particular beliefs. Where the available evidence does not support that distinction, the classification of the case as a conversion should remain open rather than being settled by the model itself. Our argument concerning conversion does not depend on Noaparast’s central thesis that ‘God exists’ cannot be a hinge commitment, which we leave open here.

The relation between rationality and conversion raises a further difficulty. If the transition to a new picture of the world cannot be established by a neutral procedure whose premises and standards are accepted by both parties, how can it be distinguished from an arbitrary choice? Conversely, if the reasons that make the new perspective intelligible can be appreciated only after conversion, does the account merely justify the transition from within the perspective that it has already produced? Our proposal must avoid both alternatives. It should neither assimilate conversion to an ordinary inference from shared premises nor render the transition immune to every form of rational assessment.

The difficulty can be clarified by distinguishing the reasons available within an initial perspective, the imaginative apprehension through which an alternative perspective becomes intelligible, and the retrospective articulation of the new orientation. These are not necessarily three successive stages, nor do they presuppose that every conversion follows the same psychological course. They identify different functions that may be involved in a transition. Reasons available within the initial perspective can disclose tensions, expose limitations, or make the alternative worth considering, without deductively entailing its adoption. Imaginative apprehension can then disclose relations of significance that the initial perspective does not adequately recognize. Finally, the new orientation may allow the subject to articulate why the transformation makes sense, how it reorganizes her experience, and what consequences follow from it. Retrospective articulation does not by itself establish that the transition was justified; it must remain answerable to the coherence of the resulting understanding, its relation to experience, and the ways in which it can be made intelligible to others.

The rationality at issue is therefore neither exclusively antecedent nor exclusively retrospective. It is informal in the sense that it cannot be reduced to a formal derivation from premises that are neutral between the competing pictures. Yet it is not arbitrary, because the imaginative transition is responsive to what the subject encounters, to the difficulties of her existing orientation, and to the intelligibility of the alternative that comes into view. Imagination does not create reasons ex nihilo, nor does it confer validity merely by making a perspective emotionally attractive. It mediates the apprehension of connections and possibilities that may subsequently be articulated, criticized, and assessed, even if no common procedure guarantees in advance that both parties will accept them.

This account also explains why the fact that a perspective becomes fully intelligible only from within it does not entail that it can recommend itself only to those who have already adopted it. A person may partially understand an alternative orientation without inhabiting it, just as one may recognize the significance of another person’s commitments without sharing them. Such partial understanding does not provide a neutral proof of the alternative’s truth. It nevertheless makes possible a form of engagement that is neither complete agreement nor sheer incomprehension. The imaginative appreciation of another perspective can precede conversion and contribute to the conditions under which it becomes possible. Conversely, the subject’s subsequent articulation of the transition remains open to criticism, including criticism concerning whether the new orientation genuinely makes better sense of the relevant experiences or merely redescribes them in its own favour.

We therefore distinguish between demonstrating that a conversion must occur and offering an account of how a conversion can be intelligible without being logically compelled. Our claim is the latter. The absence of a universally compelling inference does not establish the absence of rationality; it shows that the rational assessment of transformative understanding may require forms of comparison, imaginative apprehension, and interpretive criticism that are not exhausted by formal argument. This is the sense in which conversion can be a decisive engagement rather than an arbitrary choice, without being the conclusion of a proof that would compel any rational subject to adopt the same picture of the world.

This account of informal rationality does not imply that an alternative picture can be recommended only to those who already inhabit it. An interlocutor may be invited to reconsider an existing orientation through descriptions of shared experiences, the disclosure of previously neglected connections, and the presentation of possibilities whose significance can be partially appreciated without being fully endorsed. Such considerations need not function as premises in a neutral proof; they can instead make an alternative perspective available for critical consideration. Their force remains limited: they may fail to persuade, and their interpretation may itself be disputed. Nevertheless, they provide a basis for distinguishing imaginative engagement from arbitrary choice, insofar as the proposed transformation remains answerable to the experiences it seeks to illuminate and to the possibility of critical comparison between competing interpretations.

The crux of the matter is that a model should be provided that is able to make the dialectical tension between continuity and discontinuity intelligible. Additionally, this model should make sense of the irreducible first-personal and active character of conversion, while acts of the will are ruled out of the picture.

The solution is expressed, as was already suggested, in terms of a gestalt shift model that must be complemented by a hermeneutical, imaginative model of reasoning.

3. Subjective Understanding and the Imagination

Some have noted that the qualitative transition of conversion has similitudes with a change in visual gestalt (Schönbaumsfeld, [9], pp. 15–17): the marks of the paper that were first seen as a duck are now seen as a rabbit, or vice versa¹⁷.

The ‘duck–rabbit’ figure is one of a number of standard examples of a perspective-dependent reversible image (old woman/young woman, goblet/faces, bird/antelope) that are used to show how different appropriations of context will produce different visual experiences in people, even when the elements of their experiences are identical. These are cases where the shift in perspective involves seeing that certain things obtain. Wittgenstein reproduces a number of such images, including most famously the ‘duck–rabbit’. He speaks of the different organizations of these figures as ‘aspects’ and of reversals as ‘changes of aspect’. As Wittgenstein himself suggests when discussing the gestaltic phenomenon of “aspect-blindness” ([19], PI II, p. 213) and comparing it with what, in a different context, it is called “lacking a musical ear” ([19], PI II, p. 214), gestalt shifts in perspective may be extended to more than visual imagery, for example, to seeing at a glance what key a piece of music is written in. On Wittgenstein’s own terms they can be extended even further. This is why this kind of apprehension is a power of the trained imagination, the potential for the active recognition of patterns which are not explicitly present in the ‘neutral’ facts that are already well known to the subject before coming to see those marks as a particular figure¹⁸.

Why might conversions be understood as non-visual analogues of seeing-as? To what extent does this analogy hold true?

The first thing we can note regarding visual gestaltic transformations is that, though noticing, say, the ‘duck aspect’ of the figure, may take a preliminary effort of attention, the seeing comes, as it were, of itself. Although we can make ourselves look, we cannot make ourselves see¹⁹. This not only means that a gestalt shift is an all-or-nothing, holistic phenomenon in which the object becomes different while remaining the same object, but also, that the ‘seeing’ cannot be directly or immediately passed on to another person. Perhaps one can help another person to attempt to see the new aspect, but one’s ‘seeing’ cannot simply be transferred to the other person. Qualitative discontinuity is thus of the nature of gestaltic transformations.

The point is that the shift in perspective occurs only when a critical threshold has been reached. Initially we perceive only one figure. Although we can be told of another figure, this alternative is not already real for us. The shift is the point at which this assured possibility suddenly comes real for us, the point at which it seems to be the only way to see the marks of the paper, although we are free to shift back to the first figure by refocusing. It is also relevant to notice that in visual examples of gestaltic change, after reaching the point of the shift, there would be no ‘more or less’ in seeing the marks as a duck²⁰.

This does not entail, however, that there is not an intrinsic continuity between the marks of the paper and seeing that it is a duck. For one, although the data in any given case dictate the response, we cannot put any interpretation we would like on the data before us²¹. Additionally, it is relevant to distinguish between retrospective and prospective ways of looking at the same fact. In gestalt shifts, in one respect there is a qualitative ‘leap’, but in retrospect we experience the ‘leap’ as grounded in the physical properties and relations of the marks of the paper. This kind of ‘post hoc’ justification is never enough to explain the transition, but this is entirely consistent with the thought that, after experiencing the transition, one can conceive of epistemic support for it.

Finally, let us underline that visual gestalt shifts are ways of subjective recognition, meaning that, as it was noted earlier, they are incompatible with compulsion and irreducibly first-personal, and also, that they are cognitive experiences in which it is discovered that the world of experience is more than a list of ‘neutral’ facts. Our suggestion is that it makes sense to claim that the examples of gestaltic transitions used by Wittgenstein are designed to show limitations both in the epistemic conception of fact and in the rationalist conception of logic.

Let us compare now the foregoing description of visual gestalt changes to modes of reasoning and understanding which lead to a new perspective. These are those non-deductive and non-inductive modes of apprehension that are illustrated by examples of what, interpreting Wittgenstein, Genia Schönbaumsfeld calls “internal understanding” ([9], p. 39), namely, of what it is a further dimension of understanding that involves the perception of patterns and analogies whose discovery and description takes us off the narrow, beaten tracks of induction and deduction. Perhaps it is helpful to draw a distinction between an intellectual inquiry that is algorithmic and one that essentially involves ‘hermeneutics’ or interpretation. Tasks such as understanding the meaning of a poem ([19], PI § 531) cannot be algorithmic in character, since this kind of understanding cannot be broken in a series of clear and distinct operations, and so, it cannot be transferred and reinterpreted by means of ‘neutral’ data and passable paraphrases ([19], PI § 523). For Wittgenstein, these examples have a depth and range of importance which could hardly be exaggerated, but which has seldom, in the writings of philosophers, been anything but overlooked.

First, what impresses us about this comparison is how the model of visual gestalt shifts serves to illustrate how both the perspectives of discontinuity and continuity can be incorporated in the process of discerning patterns and reaching ‘internal understanding’. The category of critical threshold is thus common ground for the two phenomena of visual apprehension and of the dawning of a meaning aspect not previously appreciated ([15], OC § 141)²². The possibility of a retrospective justification is also constitutive of a more complex process such as the acquisition of a new and shocking valuation and understanding in complex mazes of particularity.

In addition, this understanding may be considered a vision of sorts, since the eyes of the subject are opened wider and turned in new directions, and she also comes to see patterns and connections between facts that were in a good sense already well known to her²³. Crucially, the progress from blindness to vision is here presented in cognitive terms as a journey from incomprehension to a deeper understanding, with the latter words carrying their usual commitment to the truth of what is understood, recognized and perceived²⁴. Additionally, it is relevant that, like it happens in the case of simple gestalt shifts, one can achieve understanding without adding any new items to those data that are already universally available, without even noticing a hitherto unnoticed item, but just by recognizing for the first time the nature of the pattern that is to be seen in the items. The content of the subject’s previous experience is thus fully retrieved when a new understanding begins, but each time the same content is seen in a new perspective²⁵. Also, the phenomenon of aspect-blindness is common to the instruction in visual forms and the instruction in modes of understanding.

Finally, and most crucial for our purposes, gestalt shifts, whatever their kind, are essentially subjective, meaning that they cannot be the result of rational compulsion.

However, it would be rather rash to suppose that knowing what ‘subjective’ means in a perceptual context of gestaltic transformations, is easily going to be transposable into a hermeneutical context. This question might also be posed in terms of how an apprehension of meaning, unlike what happens in simple, visual shifts, becomes real for the subject.

The crucial point is tersely expressed in [15], OC, § 117, where Wittgenstein points to suppositions and possibilities which are “idle” because they “would not tie in with anything in my life”. Something similar happens with reversible images, which though perspective-dependent, are not interspersed with human life. As we read Wittgenstein, there is here a stark contrast with the process that leads to the appreciation of meaning, which is integrally related to the whole person. Wittgenstein usually gets at meaning as we are understanding it here, by asking what would be missing from the life of someone who is blind to such aspect ([19], PI II p. 214). The self-interested character of informal modes of reasoning, as they are embodied in ethical, aesthetic and religious ways of reasoning that lead to subjective understanding, comes through clearly in Wittgenstein’s emphasis on the human activity involved in internal understanding.

It is not only that subjective understanding cannot be directly passed on to another person, but rather, that the recipient must in some way think through the meaning of what is said concretely in relation to her life. The meaning must be reduplicated by the recipient by means of an active involvement and appropriation. The recipient must learn to care about what the instructor cares about, and so, she must grow and develop in ways that it would not otherwise when it is imaginatively tempted by a redescription of the internal character and relations of the thing or person, event or situation, whose meaning is in dispute. Internal understanding thus presupposes that reason is a characteristic of real persons, and that the reasoning of such persons is shaped by their actual character as well as by the attraction of a tempting self-realization.

Crucially, the presentation of alternative meanings is not a static presentation because the picture is not an abstract possibility. On the contrary, it is a possibility which is recognized as a demand or as a task of self-development. It is in this sense that for one to discover the ethical is to discover oneself as someone involved in the imaginative activity of seeing truth as something to be incorporated in the inner life of the individual. It is also in this sense that the imagination is the medium through which meaning and truth make contact with human life, and the faculty where we find the bindingness of meaning, reason, knowledge and value²⁶. This is what it means to say that a possibility has become real for us. Understanding may be thus considered as an intensified form of being oneself and becoming subjective, namely concrete. Again, to see a possibility as a demand is to go beyond the given in a way that only the imagination can; to see as a demand what could otherwise be seen as a notional, idle possibility is to explore it with the imagination, and thus to transform what Newman called ‘notional assent’ into ‘real assent’ which is fully anchored in a human life²⁷.

Additionally, we must pay attention to another aspect of the imagination which is crucial to distinguish standard examples of gestalt shift from painstaking work on oneself and one’s way of seeing things ([25], CV p. 24)—imaginative understanding involves a simultaneous, dynamic holding of ourselves in two perspectives regarding the same object of experience; it is a way of holding oppositions together, and thus, a way of retaining together the less lively way of looking at things and the landscape of symbol and metaphor that human experience becomes through the imagination’s own active reaching out into the world²⁸. Not only does the imagination present our mind with images, it also produces, through its active perception of the world, the fundamental matrix of associations and identifications that is the necessary precursor to all internal understanding and unconditional assent²⁹.

The point is, however, that the change undergone in the appreciation of meaning is that the subject sees and thinks of a thing, a person, or a situation, in terms suggestive of another description, in another semantic domain, with another set of associations, and imbued with what Wittgenstein calls a “subjective certainty” ([15], OC § 194). While in visual gestalt shifts the prior figure is fully replaced by a new one, what happens in imaginative understanding is that the imagination is holding in tension a prior characterization of a situation in terms of traditional associations and a new one just being born in which the situation is redescribed. There is a suspension (but not the annihilation) of the previous perspective. Alternatively, and in respect to philosophical contexts, in imaginative activity what it is held in a state of suspension is the objective perspective of pure thinking that precedes the subjective perspective of care and interest—an objective perspective that ends up in scepticism with respect to beliefs and nihilism with respect to values. In any case, the imagination creates a new periphery. The object is re-created and reorganized around a new meaning; it acquires a different significance which, nonetheless, is related to the old perspective. Only someone for whom the situation has the old, standard associations will be inclined to redescribe it in novel ways.

Again, the main disanalogy between visual shifts and changes in understanding lies in that in the former there are two distinct, unrelated, and balanced objects to be seen—the duck and the rabbit figures—while in genuine imaginative reconceptualization there is only one object. That object undergoes change, it is spoken in terms suggestive of another—but the redescription does not contain two actual objects: it rather contains two distinct perspectives on the same object, and those perspectives are held in paradoxical tension. Similarly, when a person adopts a particular religious picture of the world, the world itself does not have to be replaced by another world. What changes is the network of associations, commitments, and possibilities through which the person understands it. The transition must therefore be described in relation to the particular religious and non-religious pictures involved, rather than as a universal passage between two homogeneous worldviews. There remains one world to be understood before and after conversion, although its significance may be reorganized through the transformation of the subject’s orientation.

The fact is that the hermeneutical and imaginative model that we find in the appreciation of meaning complements the perspective of gestalt shift, and brings to the fore the decisive non-symmetrical character of this kind of transition and its cognitive contribution. If the instructor and the recipient (before the shift) look at the same data, there is a clear sense in which they are both looking at the same thing. But there is an equally clear and equally important sense in which they see quite different things. The fact is that the actual thing and the ‘imaginative’ thing are not equivalent entities, for one comes to see the former in terms suggestive of the latter, to see in it a depth and a significance of which one was previously unaware. The decisiveness of the imaginative revisioning and its character as a cognitive advance are thus a function of this reconceptualization of information that is already available to us. The reconceptualization is not before the transition; the transition is not before the reconceptualization—the reconceptualization is the transition.

The upshot is that Wittgenstein maintains not only that there are modes of reasoning and apprehension which are neither logical nor grounded in neutral evidence, but also, and more radically, that objective modes of reasoning themselves ultimately depend on subjective modes of reasoning and apprehension for us to have a world and to relate ourselves to the truth by means of making it ‘real’. Like for one to discover the ethical is to discover oneself as something more than a moral ensemble, for one to have a worldview is to discover that the world is more than a list of decontextualized facts and their logical and causal connections³⁰.

For Wittgenstein, the person who chooses the pure objectivity of rationalist epistemology loses the truth both in life and in belief. Thus, Wittgenstein rejects that our relation to our systems of reference is epistemic all the way down. At this point of the discussion, let us recall that reason in its normative sense is defined in terms of the search for truth—and also, in terms of the search for truth regarding the thick nature of normal experience.

4. Paths to Conversion

It is clear even from this brief discussion on ‘internal understanding’ and its characteristics, as well as from the fact that religious transitions fit perfectly well into the hermeneutical and imaginative model of reasoning presented above, how great an advance Wittgenstein’s account represents over the epistemic accounts that it was designed to supersede (and which, ironically, hinge epistemologists usually project to Wittgenstein in their epistemic readings of OC). It shows that disputes about the existence of God are neither empirical nor formal nor unintelligible. It makes clear that the unbeliever and the believer are engaged in a genuine conflict, and that the question concerning which they are in dispute has nonetheless a logic. And Wittgenstein rightly represents controversies where different pictures of the world are involved as having a logic without wrongly representing them as having the abstract, decontextualized logic of empirical beliefs and deductive inferences.

Examples of internal understanding would be valuable even if Wittgenstein’s purpose were no more than to produce counterexamples to rationalist restrictions on the range of reason and rationality. But those examples have a more direct bearing on questions about the nature of the controversy between the non-religious and the religious views of the world. It shows that the believer and the unbeliever may be disagreeing about a matter of fact, about what in fact the character of the world is, even when there is no evidence available to the one that is not available to the other, and when there is no expectation of future evidence that could be epistemically disclosed to anyone so as to solve the question. It shows also that dispute about the meaning of the world is like a dispute about the meaning of a poem or about the character of a person, a dispute which may persist even when each party to the dispute has before her all the data that are before the other. Crucially, it shows indirectly how there is something imperialist in epistemic modes of reasoning, which always seem to be striving to master and to make universally available (namely ‘gained for free’ and ‘requiring no subjective re-appropriation’) the recalcitrant elements of the human condition.

There is the need, however, of a further investigation on the fateful attraction of objective reasoning for hinge epistemologists and of the ways to dispel the illusion that there must be something (a gap-filler) that bridges the gap between evidence and assurance to provide legitimation for religious conviction—an illusion that also expresses itself in Pritchard’s characterization of faith as ‘arational’, meaning that faith is ‘above’ rationality³¹. This further investigation comes from the demand for a more detailed exploration, against the background of our foregoing discussion on transition, of how faith may in some sense be a natural development of reason, while in a different sense it may be conceived both as the fulfilment of reason’s normative aims, and as a fulfilment that baffles reason.

Let us start with the main features of the position that Pritchard maintains, as he has explored this problem in detail.

The first aspect to notice is that Pritchard’s terminology (‘quasi-fideism’, ‘arational’, etc.) invites the misunderstanding that the religious person does not care about truth, since reason, as it was noted earlier, can be defined as a set of processes that are aimed at truth, and therefore, as ‘the human organ of truth’. However, believers think otherwise—they think of faith as a way of relating themselves to the truth.

But his framework seems to invite a further difficulty: it appears to treat the realm of facts as divorced from the realm of values.

Pritchard is right in noting that religious hinge commitments are not to be understood as theoretical claims about empirical matters (see, for example, Pritchard [40]). He is also right in thinking that there is an axiological dimension to religious conviction in that to have religious faith is to see the world as imbued with value and meaning in a very specific way (see Pritchard [41]). However, we need to be wary of supposing, as his distinction between ‘mundane’ and axiological hinges may suggest, that there is a hard-and-fast divide between those factual hinge commitments which constitute what he calls the ‘mundane’ view of the world and the thicker worldview of human experience that also incorporates moral, political, aesthetic, and religious values. On this reading, it would be as if those ideals were superimposed on value-free reality in an autonomous way, as if the empirical self and the moral community in which it is placed were the only source of axiological authority and validity, and as if, divorced from the realm of truth, our value choices were arbitrary and groundless.

The concern here is not that Pritchard is committed to metaphysical materialism. The distinction between factual and axiological hinge commitments does not, by itself, establish such a commitment. Rather, our concern is whether the distinction leaves the significance of values insufficiently integrated into the account of how reality is understood. Pritchard’s discussions of axiological and religious hinge commitments ([40,41]) make this a relevant interpretive question, but a stronger metaphysical conclusion would require textual evidence that the distinction is intended to reflect a separation between facts and values in reality itself. In the absence of such evidence, the argument should be directed not against a materialist position attributed to Pritchard, but against the possibility that a sharp separation between factual truth and axiological significance would inadequately account for the meaningful character of human experience³². Additionally, when we said above that certainty and assent contain a reference to a norm, all that we meant is that in all certainty (including religious certainty), as distinguished from opinion and surmise, we have the sense of there being no alternative—one cannot view the matter in any other way than this. The imagination is now bound by the norm (and value) of truth, as it always is when we come into contact with reality. Thus, by a quite different path, we are led once again to the conception of constraint as an important element in certainty. It is constraint (binding) what militates against Pritchard’s view of religious commitment (and conversion).

The crucial factor in this controversy is that hinge epistemologists in general are taking the same kind of ‘analytical’ attitude regarding reasoning and belief that empiricist psychologists were prone to adopt regarding perception³³—an attitude that Wittgenstein, following Gestalt psychologists, clearly contrasted with the ‘normal’ attitude in perception, and towards which he was harshly critical ([17], RPP I §§ 110-12). When we take the analytical attitude recommended by epistemological conceptions of rationality, we form beliefs in exactly the same abstract (deductive or inductive) way epistemologists say we strictly speaking do when we are reasoning and forming ordinary beliefs. In this sense, epistemology may be considered as (i) a ‘purified’ version of our ordinary rational practices, and (ii) a technique of sorts which, through decontextualization and abstraction from subjective, ‘impure’ factors in belief formation, get us into this attitude that opposes the ‘normal’ and subjective attitude in our ordinary lives. The hinge epistemologist unjustifiably reads the experience of reasoning we have while adopting the analytical attitude—one which is highly arbitrary—into that which we have when taking the normal attitude in reasoning and believing.

It is, therefore, as if hinge epistemologists were blind to subjective modes of apprehension. It is small wonder that on this view, for which reason is radically reduced to objective reasoning, there is room neither for the Wittgensteinian insight that certainty is possible only in so far as we are moral beings, nor for the thought that there is a generic faith which is in some sense a natural idea, as well as the bottom rock, of reason. Following Wittgenstein, the right standard for measuring the objective perspective that hinge epistemologists usually take is asking what would be missing from the life of one who is blind to axiological, meaningful aspects of the world.

As was argued in the previous section, the ethical dimension is intrinsically involved in the imaginative activity of incorporating truth to the life of the individual. Our present concern, however, is with the normative aspect of reason (our moral obligation to the truth) and its foundations, namely with the question as to the meaning of the insight that that without which we could have no moral obligation must be true. Consider the general principle that whatever is necessary in order that life may have moral significance (‘meaning’, as Wittgenstein understood it) is true. For it is this which really constitutes the heart of Wittgenstein’s version of the Kantian doctrine of the primacy of practical reason, a doctrine that binds together ethics and epistemology and which is tersely expressed in what Wittgenstein calls the primacy of acting, “which lies at the bottom of the language-game” ([15], OC § 204)³⁴. This one cannot doubt, because to doubt it would be to reject the foundations of normative reason and the very possibility of relating ourselves to the truth.

But to say that our moral obligation to the truth must be true, is only to say that we live in a moral universe. The one thing that one cannot doubt is that this is a moral universe, that we have demands and the ability to perform them. This ‘belief’ in the significance of our sense of moral obligation is the fundamental act of faith on which reason is grounded—what we previously called ‘generic faith’.

However, the declaration that the world is ethically significant, if it represents a real belief, is not so much a theoretical as a practical attitude. The assertion of this ethical dimension is absolutely necessary, if one presupposes the resolve to obey the mandate of truth. This, then, is what Wittgenstein understands by the primacy of acting, and also what justifies the claim that faith is rational because it is linked to the practical life both in its moral and in its cognitive dimensions. Wittgenstein’s perspective does not mean that the ethical demand can establish for us theoretical propositions which objective reasoning is unable to establish. It means that certain beliefs and commitments which philosophers are wont to call ‘theoretical’ are not theoretical after all—that doubt in regard to them is a disease of the will. The individual who doubts reason’s faith in the meaning of the world is one whose attitude to life is in need of healing. Wittgenstein writes that we might picture her as “behaving more ‘like an automaton’” ([17], RPP I §198), “as it were sleep-walking” ([17], RPP I § 178). To be blind to subjective understanding is thus to lose that dimension of experience which makes it possible incorporating truth to our experience. Abstract reasoning is surely a case of world-blindness, a detached, etiolated perspective where the world is flat, prosaic, faceless³⁵.

However, this generic faith is only loosely related with religious faith. It is true that interested thinking as Wittgenstein conceives it opposes pure thinking; but religious conversion is a transition to a way of looking at the world that plays no positive role, whether constitutive or regulative, in grounding reason. What is more, with regard to Christian faith, the contents of religious experience are intrinsically paradoxical, since a revelation that takes the incarnation as its object represents a break with all immanent human thinking. Our concern now is with the question as to whether religious conversion may be somewhat related to a broadened, loosely Kantian conception of rationality (one which includes the practical dimension at its heart), and how.

As to this problem, the first thing to note is that reason in its normative sense is defined in terms of truth, and therefore, that reason, in its own terms, leaves room for the possibility to fulfil its goal by means of ceasing to make rational intelligibility its primary end, namely by means of setting itself aside in order to achieve its ultimate aim. Just as changes in one’s conception of the world require an encounter with someone else who thinks differently, reason may be challenged by something that thought itself cannot think. In this sense, we could interpret normative rationality as seeking its own limits, or better said, as being traversed by paradox, namely by two opposite tendencies—its tendency to judge religious faith irrational, and its tendency of transcending itself for the sake of itself (of its goal). One might thus say that in conversion reason founders but is not (teleologically) annihilated. One might also say that it is only when reason is acquainted with unintelligible contents that the possibility of recognizing its own untruth is given, and with it, one acquires an understanding of why faith cannot be understood by autonomous reason in its two interpretations: as external, algorithmic understanding, and as practical reason and the ethical. This understanding is not yet conversion; but it is the necessary path to conversion. As we read it, this is what Wittgenstein means when he writes: “If Christianity is the truth, then all the philosophy about it is false” ([25], CV p. 89).

The crucial point lies, however, in the transition to faith. Again, the question is not only whether rationality on its own leaves some room for religious conversion, but rather, whether conversion is tied somehow to human reason, and so, whether there is some way to contend that religious faith makes sense, that it may be appropriated (though not mastered) by rationality. The question is, what are the credentials, if any, for religious faith?

The way out of the quandary consists in heeding Wittgenstein’s warning that it is against the backdrop of the gradual development of the human life that transcontextual criteria of meaning and understanding can be found. The point is that religious conversion is, for Wittgenstein, intrinsically related both to hermeneutical modes of understanding and to the practical re-orientation which is inherent to them. It is, again, by means of going back to the rough soil of human life that religious conversions might be redeemed. Our convictions are anchored in a commonality—which does not license every picture of the world: pictures that cannot remain answerable to shared human experience, or that cannot withstand critical comparison, may be legitimately rejected—of experiences which are the substantial and transcontextual criteria for the evaluation of worldviews, including the religious pictures of the world and the forms of life in which they are embedded. They are standards of intelligibility and truth that make contact with human life, which both believers and unbelievers share, and which leave room for an internal, subjective understanding of religious faith as well as for the possibility of criticism³⁶. Crucially, those phenomena that are not in dispute mainly refer to facts that happen in the human soul—something that Wittgenstein emphasized when saying that “Christianity is not a doctrine”, but “a description of something that actually takes place in human life” ([25], CV p. 32).

One form of rational engagement across pictures of the world deserves explicit mention. Noaparast, in a different context, observes that criticism between frameworks need not rely on the critic’s own commitments; it may instead turn the interlocutor’s own commitments against the framework she inhabits, for instance by showing that it fails to satisfy standards of coherence that she herself accepts ([36]; see [15], OC § 609). Such immanent criticism does not constitute a neutral proof, and we do not suggest that it exhausts the reasons that may bear on a transition. It shows, however, that considerations can cross the boundary between two pictures without presupposing the critic’s picture. This is compatible with the account of informal rationality defended here: reasons of this kind may disclose tensions within an orientation and so contribute to making an alternative available for critical consideration, without entailing its adoption. Likewise, Noaparast describes the personal ‘tests’ that suffering may impose on religious commitment as involving both rational and emotional elements, rather than as merely arational ([36]). We take this as a point of convergence with our claim that the transition is neither a formal inference nor an arbitrary choice, while noting that our own account locates the intelligibility of the transition in imaginative apprehension rather than in coherence considerations alone.

Finally, it is important to note that while ‘generic faith’ involves an awareness of God that is linked to our ethical nature, religious faith is precipitated by a heightened sensitivity to our fragility as moral agents, namely, to the unfeasibility of moral demands and the radical contingency of the ethical. It is this truth, which selfish reason is unable to fully grasp and to appease, the main aspect of human experience to which we would be blind without faith, and which, brought in conversion to the surface, provides its ‘post hoc’ legitimation.

5. Concluding Remarks

There is, however, an apparent gap between reconceptualization and conversion that it is particularly noticeable in the religious sphere. It seems as if religious reconceptualization is not itself the transition because, as was previously noted, there is the possibility of coming to appreciate the religious vision for it to attract us without coming to a real, transformative acceptance of it. In fact, the paradoxical nature of reason makes this possibility particularly alive—in terms of Kierkegaard, reason may take offense of the contents of faith, it may refuse to step aside and to accept its own untruth. The problem lies in that the choice between offence and acceptance seems to require a pure act of the will, a wilful decision between two arbitrary options: offended reason and reason that has been transformed by faith. On this view, the will becomes the ultimate gap-filler on which transition hangs.

It is true that resistance to faith is natural to reason, and therefore, that for religious transformation to occur it is in a certain sense required that the will (through an effort) would be able to remain interested in the new vision, and so, to counter-balance this tendency. However, it is also relevant to note that in most reports of conversions (Ferreira, [42], pp. 105–113), it is stated that one finds herself already converted, with no explicit ‘act of the will’ required. Our view is that what it is called ‘an act of the will’ is the realization of how real the religious vision is for us, namely an expression of the way in which a tempting reconceptualization has decisively become real for us. The transition is thus a decisive engagement, rather than an arbitrary choice between alternatives.

The idea is that in conversion one lets oneself to be transformed by the new vision by means of keeping oneself interested and engaged. The ‘act of the will’ does not describe what happens at the ‘leap’; it rather refers to our free, active engagement with reconceptualization, and to the transformation that results from it. At most, it also refers to the explicit effort to focus our attention—one can decide to pay attention to the new view, but one cannot decide to be transformed by it, even though transformation is also essentially active. In short, acts of the will find their place among the conditions for conversion, which is, as it were, the crystallization of imaginative activity. They are not, however, the ultimate cause that precipitates conversion.

It is useful to flag a consequence of this claim, which is that if appreciation does not lead to conversion, it is not because one decides to resist conversion. On the contrary, what one decides is to remain detached, uninterested, and abstract (objective)—a decision that indirectly prevents reconceptualization to be subjectively actualized.

To summarize, we have argued for a re-evaluation of Wittgenstein’s account of transitions to a new way of looking at things, with particular attention to his view of the transition to faith. Both implicitly and explicitly, we suggested, Wittgenstein calls attention to the importance of the imagination in such a way as to challenge all those epistemic readings which exclude or ignore imaginative activity, as well as those which limit imaginative activity to ‘wishful thinking’. Wittgenstein calls attention, in particular, to the ways in which transformative changes can be understood as an imaginative, reorienting shift in perspective. It is our view that Wittgenstein’s logic of subjectivity clearly challenges some widely assumed representations of Wittgenstein’s last notes.