The Emergence of Religious Narrative

Abstract

This article examines the conceptual connections between simpler and more complex forms of religiosity, focusing on the transition from ritual-based practices to religious narratives and theological reflection. Drawing on Wittgenstein’s method of perspicuous representation (übersichtliche Darstellung), the authors propose a series of models that illuminate this spectrum. These models demonstrate how religious narratives achieve autonomy of a sort that challenges reductionist interpretations. Rituals, initially guided by primitive reactions, become structured through linguistic conceptualisation and are woven into cohesive narratives that, in turn, serve as internal justifications for ritual practices, creating a linguistic space that encourages reflection. The article contends that theological reflection emerges when narratives encounter discrepancies—whether from external challenges or internal inconsistencies—prompting a systematic re-evaluation of beliefs. By critiquing Wittgenstein’s own reductionist tendencies, along with the “Wittgensteinian fideism” that emerged in its wake, the authors seek to emphasise the importance of recognising disputes within and between religious narratives as being integral to human life.

1. Introduction

Discussions within the tradition of the philosophy of religion inspired by Ludwig Wittgenstein typically focus on the use of religious expressions. Critics of this approach tend to argue that the debate centres on the epistemic status of religious claims (cf. Pichler and Sunday Grève 2024). In contrast, proponents suggest that the issue lies in the autonomy of the meaning of religious linguistic practices as distinct from practices related to describing facts and justifying propositions (cf. Schönbaumsfeld 2023, pp. 44–46). This article, to some extent, situates itself within this discussion. Its aim is to expose certain philosophical misunderstandings that have arisen concerning the meaning of expressions used in religious contexts. To this end, we seek to clarify the relationship between simpler and more complex forms of religiosity—specifically, the relationship between forms limited to ritual behaviours and forms that include narratives or reflections relating to objects of religious veneration.

We will employ Wittgenstein’s method of perspicuous representation (übersichtliche Darstellung), an idea emphasised by Gordon Baker and further elaborated by Oskari Kuusela. Our implementation of this method involves presenting a series of models that illustrate a spectrum of conceivable forms of religiosity. These models serve as “objects of comparison” in that they help to explicate relationships between the meaning of religious ritual and religious experience, as well as the meaning of religious narratives and intellectual reflection within a religious framework.

The philosophical misunderstandings we aim to expose are rooted in a misrecognition of the nature of this relationship. More specifically, they stem from a tendency to deprive more complex forms of religiosity—such as narratives and forms of reflection—of their distinctiveness and autonomy, or even to reduce them entirely to the sphere of ritual and experience. It appears that even Wittgenstein himself was not immune to this tendency, as his remarks on religion can be read in a reductionist vein. This, in turn, has led some of his followers to adopt an inadequate understanding of the meaning of religious expressions.

Our inquiry, therefore, proceeds, to use Karl-Otto Apel’s well-known phrase, “with Wittgenstein against Wittgenstein”.1 Its goal is to demonstrate the autonomy of the meaning of religious narratives and theological thought—not autonomy in relation to epistemic practices, but rather in relation to the purely ritualistic and experiential layers of religious life.

2. The Misunderstanding and Its Source

Let us begin with a well-known general remark that Wittgenstein wrote in 1937:

The origin and the primitive form of the language game is a reaction; only from this can more complicated forms develop. Language—I want to say—is a refinement, “in the beginning was the deed”.

(Wittgenstein 1984, p. 31)

This expresses the view that language—both in its genetic and structural dimensions—is rooted in the realm of reflex actions and represents a subtle refinement of that realm. One can suppose that, according to Wittgenstein, the meaning and understanding of language are closely tied to practical activity. However, he clearly did not mean that every meaning is a construct made of simpler “building blocks” that can ultimately be traced back to reflexes. When he stated that language games, as socially established institutions determining meaning, are rooted in practices, he also emphasised that the range of these practices is very broad: from forms of behaviour that are almost reflexive, such as greetings or expletives, to highly complex and nuanced activities that rely almost entirely on the manipulation of symbols, such as translating poetry or solving mathematical problems (Wittgenstein 2009, §23).

However, as Rush Rhees observed, the metaphor of language games served different functions in Wittgenstein’s project at various stages of its development (Rhees 1969, p. vi). Consequently, the use of the terms “origin” and “primitive form”, along with the discussions of the builders’ language games in the early sections of Philosophical Investigations, can be seen as an echo of what might be called a reductionist conception. This conception is explicitly outlined in a prominent passage of the Blue Book, specifically in the explanation accompanying the first occurrence of the term “language game” in the text:

Language games are the forms of language with which a child begins to make use of words. The study of language games is the study of primitive forms of language or primitive languages… When we look at such simple forms of language the mental mist which seems to enshroud our ordinary use of language disappears. We see activities, reactions, which are clear-cut and transparent. On the other hand we recognize in these simple processes forms of language not separated by a break from our more complicated ones. We see that we can build up the complicated forms from the primitive ones by gradually adding new forms.

((Wittgenstein 1969, p. 17)—italics added by us)

In his introduction to the Blue and Brown Books, Rhees suggested that this passage is, to some extent, a remnant of the Tractatus idea of language analysis (Rhees 1969, p. vii), a claim that indeed seems convincing. He also speculated that a certain passage from Wittgenstein’s manuscripts on simple language games indicates that the author of the Philosophical Investigations soon abandoned this reductionism. That passage, translated into English by Rhees, reads as follows:

When I describe certain simple language games, this is not in order to construct from them gradually the processes of our developed language—or of thinking—which only leads to injustice (Nicod and Russell). I simply set forth the games as what they are, and let them shed their light on the particular problems.

(Rhees 1969, p. viii)

These sentences, in precisely the form cited by Rhees in German as well, appear for the first time on page 81 of manuscript Ms-115, which was written between the 14th and 31st of December 1933. Thus, they could indeed have been written after the earlier statement by Wittgenstein during the dictation of the Blue Book to his students.2 The issue, however, is that a very similar passage in terms of its overall meaning (differing primarily in respect of the absence of references to Nicod and Russell) is to be found on the reverse side of page 45 of manuscript Ms-113, dated 1 March 1932—that is, several months before the aforementioned statement.

This could indicate that the trajectory of the reductionist conception in Wittgenstein’s project was far more complex than Rhees’ reconstruction suggests: it was not a linear development but rather an oscillation between the spectre of constructing more complex forms from simpler elements and the rejection of this idea. We refer to this as a “spectre” or “echo” because, generally speaking, Wittgenstein’s later philosophy is contextualist in its spirit. The meaning of a specific linguistic activity was always to be considered in the context of the broader whole—namely, the language game within which that activity occurred—and, in a wider sense, in light of the relevant form of life: that is, the particular cultural and biological framework shaping human functioning within a given community and historical period.

A trace of the reductionist conception can still be observed in Wittgenstein’s thinking about religion. In the notes written between 1930 and 1931, which form the first part of the Remarks on Frazer’s ‘The Golden Bough’, the phenomenon of religion is presented, in practice, almost as an expression of instinct. The content layer—not to mention theological constructions—is entirely subordinated to the ritualistic sphere. Let us consider two characteristic passages:

Was Augustine mistaken, then, when he called on God on every page of the Confessions?

But—one might say—if he was not in error, then surely was the Buddhist saint—or whoever else—whose religion expresses entirely different notions. But none of them was in error except where he was putting forth a theory.

(Wittgenstein 2018, p. 32)

[W]hat is characteristic of ritual action is not at all any view, opinion, be it right or wrong, although an opinion—a belief—can itself be of ritual nature, or belong to a rite.

(Wittgenstein 2018, p. 44)

The first passage conveys the belief that, despite differences in linguistic expressions (“notions”), the statements of saints from various religions hold value and do not contradict one another, as long as they do not descend into intellectual speculation (“theory”). The second passage suggests that religious belief finds its proper place within the framework of ritual, thereby rendering questions about its correctness or incorrectness meaningless. Indeed, the entire structure of Wittgenstein’s argument across these dozen or so pages of notes is based on a contrastive juxtaposition of terms such as “explanation”, “opinion”, and “theory” with terms like “piety”, “ritual”, and “ceremony”, and a corresponding discrediting of the former as having nothing to do with religion.

It can be observed that the emphasis on practice and action is a characteristic feature of the entire project of the author of the Philosophical Investigations, and thus the first part of the Remarks on Frazer’s ‘The Golden Bough’ seemingly aligns well with this focus. However, whereas in the Investigations, where intellectual activities, such as the previously mentioned translation of poetry or mathematical operations, are considered autonomous and irreducible to simpler activities, in the Remarks, intellectual activity within religion is presented solely as a component of ritual. From the perspective of approaches that seek to consider the cognitive content of religious statements, the concept of religion outlined in this text clearly falls within a strong version of non-cognitivism (cf. Pichler and Sunday Grève 2024). From our perspective, it also expresses a reductionism of sorts: the meaning of the entire content of the complex phenomenon of religion is reduced to ritual activities motivated by the impulse to express reverence.

In Wittgenstein’s later remarks on religion, that kind of ritualistically oriented reductionism almost entirely disappears. Reflections on celebration give way to a strong emphasis on the role of personal commitment. The author of the Philosophical Investigations focuses on the role of religion in human life, effectively adopting an approach inspired by the existentialist concept of authenticity—probably influenced by his fascination with Kierkegaard (cf. Schönbaumsfeld 2007; Fox-Muraton 2022). Wittgenstein explains that religious narrative functions as a call that demands a clear “yes” or “no” in response, while a lukewarm reaction such as “maybe” constitutes a complete misunderstanding (cf. Wittgenstein 1972, p. 53).

This imperative dimension of the call entirely dominates the meaning of the narrative, rendering any different individual receptions incommensurable. For this reason, Wittgenstein was able to say during his lectures in 1938 that his lack of belief in the Last Judgement did not mean he was contradicting someone who did hold such a belief (cf. Wittgenstein 1972, p. 55). Similarly, his references to religion in the late 1930s and 1940s should be interpreted in this light: for example, the well-known and frequently cited remark from 1937 about the irrelevance of historical findings to the significance of the Gospel’s message. In this remark, Wittgenstein stated that “historical proof (the historical proof-game) is irrelevant to belief” (Wittgenstein 1984, p. 32). This statement was preceded by an imperative clarifying the specificity of the religious call: “Here you have a message!—don’t treat it as you would another historical message! Make a quite different place for it in your life” (Wittgenstein 1984, p. 32).

Despite certain hesitations,3 Wittgenstein maintained a distinctly existentialist approach to religious content virtually until the end of his life, as evidenced by a well-known remark dating from 1948:

Religious faith and superstition are quite different. One of them results from fear and is a sort of false science. The other is a trusting.

(Wittgenstein 1984, p. 72)

This remark can be interpreted as a radical contrast between two approaches to religious messages: the positively valued acceptance of the call (“trusting”) is identified with faith in the strict sense, while an attempt to interpret the message in an objectivist manner is discredited, not only through unequivocally pejorative epithets such as “superstition” and “false science”, but also by pointing to its unworthy psychological foundation—namely, “fear”.

The movement formed after his death, commonly referred to as “Wittgensteinian fideism”, interpreted this essentially marginal theme in Wittgenstein’s reflections as the application of his method of philosophical therapy to the specific domain of religious faith, and sought to continue this approach (cf. Holland 1956). In the writings of the “Wittgensteinian fideists”, most notably in the works of Dewi Z. Phillips,4 popular religion (what Wittgenstein might call superstition) was exposed as being based on a misreading of the religious message in terms of factual reporting (causal discourse) or super-facts (metaphysical discourse). It was also reiterated, following Wittgenstein, that the psychological motivation for this interpretation is fear, particularly the fear of death. At the same time, it was argued that true faith is grounded in an attitude of radical selflessness, self-sacrifice, and love of one’s neighbour.

Philosophers of religion inspired by Wittgenstein did not continue the reductionist approach present in the Remarks on Frazer’s ‘The Golden Bough’. On the contrary, they emphasised the autonomy of religious meaning and built their critique of more traditional views—such as those of John Hick, Richard Swinburne, and Alvin Plantinga—on this foundation. These traditional approaches treated the message of the Gospel as a narrative about what actually happened and God as an omnipotent, conscious being (cf. Hick 1977, p. 122).

The “Wittgensteinian fideists” themselves acknowledged that God is an independent reality (Rhees 1997, p. 9; Winch 1964, pp. 308–9; Phillips 1988, p. 106). However, due to the aforementioned autonomy of meaning and the radically pluralistic understanding of the term “reality” associated with it, the consequences of such a declaration were far from intuitive. Among other things, it became clear that this reality was still not something open to dispute (cf. Phillips and Jones 1993, pp. 125–26).

From the perspective of the position adopted in this article, both Wittgenstein’s later position and that of the “Wittgensteinian fideists” themselves bear certain traces of reductionist thinking. As we have noted, in the early 1930s, Wittgenstein wrote that Augustine would not be able to contradict a Buddhist saint unless they began to theorise. He never abandoned this conviction, although over time he changed the strategy by which he justified it. This strategy, along with the conviction itself, was also adopted by the philosophers of religion inspired by him.

We do not intend to deny Wittgenstein and his followers a certain validity in respect of their position: a religion not built on commitment becomes empty speculation, raising the question of whether it can still be called religion at all. However, we take the claim that the existence of genuine religious arguments is a confusion rooted in superstition to itself be a component of a rather peculiar reductive religio-theological doctrine, rather than a conclusion arising from philosophical activity in the Wittgensteinian sense.

In the subsequent sections of this article, we will provide a detailed explanation of what such philosophical activity entails and then present an example of it. However, before doing so, we must explicate the perspective we have adopted on the problem of religion.

3. The Question of Emergence

In the introduction, we stated our intention to address the relationship between simpler and more complex forms of religiosity. Specifically, we are concerned with how ritual-based forms relate to religious narratives or, alternatively, how the latter emerge from the former.

Achieving a degree of clarity on this matter, however, requires specifying several issues. First, we need to determine the type of relationship in question—whether it is a causal relationship between different forms of religiosity or a conceptual one. This distinction is closely tied to Wittgenstein’s distinction between reasons and causes.5 Second, we must decide in what terms the most primitive forms of religiosity should be understood. Third, we need to identify the components characteristic of the more complex forms of religiosity.

The first of these issues is particularly significant because a lack of clarity here can lead to conflating philosophical explanations of the connections between different forms of religiosity with scientific explanations of the evolution of those forms. One manifestation of such conflation is the mistaken interpreting of philosophical accounts of relationships between forms of religiosity as amounting to quasi-empirical theories explaining the evolution of religion. The other two issues are important because answering our primary question requires specifying what belongs to the most primitive forms of religiosity and what belongs to the more complex ones.

To address all these questions, we will employ the method of presenting certain objects of comparison, namely, models. The models we introduce below are forms of religiosity. On the one hand, they are not exhaustive descriptions of actually existing forms of religiosity, while on the other, they are not merely descriptions of possible but unrealised forms of religiosity. Instead, they serve as objects of comparison designed to highlight the particular aspects of real forms of religiosity that are of interest to us. Our comparison of these models aims to clarify the ways in which more complex forms of religiosity are related to its more primitive forms.

It seems that the question of the relationship between the most primitive and more complex forms of religiosity can be understood as a question about how the latter developed on the basis of the former. Answers to the question thus construed will therefore involve explanations referring to the causes and actual conditions underlying different forms of religiosity. Such explanations are formulated within the framework of the empirical sciences. Therefore, in our view, questions concerning the relationship between primitive and more complex forms of religiosity, understood in this way, can be addressed through research in anthropology, psychology, sociology, and history in the broader sense.

If, however, the question of the relationship between primitive and complex forms of religiosity prompts philosophical rather than empirical reflection, we should bear in mind that the aim of such reflection cannot be any kind of theory or hypothesis; in general, it cannot involve any form of explanation. As Wittgenstein wrote,

it is, rather, essential to [philosophical] investigation that we do not seek to learn anything new by it. We want to understand something that is already in plain view”.

(Wittgenstein 2009, §89)

The motivation for asking philosophical questions lies in the existence of philosophical problems, that is, situations where “I don’t know my way about” (Wittgenstein 2009, §123). These “are solved through an insight into the workings of our language, and that in such a way that these workings are recognized—despite an urge to misunderstand them” (Wittgenstein 2009, §109). In our case, this lack of clarity arises from our tendency to misunderstand the relationship between rituals and religious narratives.

Let us return to the beginning. What exactly are we asking when we pose the philosophical question about the relationship between primitive forms of religiosity and its more complex forms? This matter requires clarification. Essentially, the question concerns the nature of the relationships between the meanings that can be attributed to the most primitive forms and the meanings expressed within the more complex forms.

What we seek to understand includes, among other things, both how the meanings of more complex forms of religiosity are enriched compared to those of the more primitive forms, and what differences exist in the ways meanings are expressed within various forms of religiosity. Clarifying the latter issue is particularly important, as a certain naïve understanding of the differences between more primitive and more complex forms of religiosity reduces these differences to the idea that the more complex forms simply express a greater number of meanings than the primitive ones.

In our view, however, these differences also manifest themselves in the ways in which religious meanings are expressed. In the case of the most primitive forms of religiosity, meanings are expressed solely through rituals; in somewhat more complex forms, they may also be expressed through specific formulations accompanying the rituals; and in even more complex forms, they are expressed through narratives and even theological statements. The detailed characterisation of the nature of these relationships and their components will be addressed in one of the sections below.

4. The Method of Objects of Comparison

In relation to the aims of our inquiry as outlined above, the following question arises: by what means can these goals be achieved? As we have already indicated, our investigation is not historical, anthropological, or psychological in nature, but rather philosophical in the Wittgensteinian sense of the term. Consequently, our reflections will neither present the history of the development of religion nor examine the psychological or sociological mechanisms that might underlie the emergence of new forms of religiosity.

Instead, we intend to employ a method used by Wittgenstein in his inquiries. Its goal is the perspicuous representation of a given phenomenon. It involves introducing certain objects of comparison—also referred to as models—which are then used as reference points for analysing the phenomenon under investigation. By the use of these points of reference, we aim to draw attention to aspects and features of the phenomenon that might have been previously overlooked.

In other words, as Gordon Baker—who regarded the idea of perspicuous representation as being central to Wittgenstein’s thought—explained, such comparisons are intended to “liberate our thinking from enslavement to particular analogies by bringing to light other analogies which are equally well supported as the ones of which we unconsciously make use” (Baker 2004, p. 34). The choice of a given model (or series of models) and the emphasis on certain aspects of the phenomenon are guided by the nature of the philosophical problem we are seeking to dissolve. It follows that the objects of comparison should serve exclusively to resolve specific existing misunderstandings and must never be employed to construct any general theory of language functioning (Baker 2004, p. 27).

The most well-known example of Wittgenstein’s use of this method is his description of various language games in order to understand the functioning of different aspects of language and avoid confusion regarding its use. For instance, in the language game of Philosophical Investigations §8, he demonstrates that the conception of meaning, according to which all words are names of objects and all sentences are statements, amounts to a misunderstanding. However, a perspicuous representation can just as well consist of entirely fictional examples, unrelated to “our grammar” (cf. Baker 2004, p. 42).

Oskari Kuusela also regards the method of resolving philosophical problems by comparing the phenomena under consideration to different models as one of the most distinctive features of Wittgenstein’s later philosophy. Kuusela emphasises that, through the application of this method, Wittgenstein’s philosophy avoids dogmatism, which arises from advancing theses about the phenomena being considered based on models that are, by their nature, simplificatory:

Insofar as the discussion of the proper role of the ideal—and of turning the examination around—concerns the question of how to avoid metaphysical projections and dogmatism, an answer is readily available. Presumably Wittgenstein’s method of using simple language games for purposes of clarification is also meant to accord with the aim of avoiding dogmatism. But then the remark on models as objects of comparison could very well be presented as concerning at one time the proper role of the ideal and at another time the role of language games. In both cases the problem is the same, i.e., that of coming to project a mode of presentation onto the object of investigation in the form of a thesis. Similarly, the solution in both cases is the same. We are to take the model—the ideal or the language-game; the term “model” applies to both—as an object of comparison.

(Kuusela 2008, pp. 128–29)

Kuusela is undoubtedly correct in asserting that when Wittgenstein puts forward objects of comparison, he does not treat them simply as exhaustive descriptions of real phenomena. First, such objects can represent unrealised possibilities or even abstract models so idealised that they could never be actualised (e.g., the aforementioned builders’ language game in Philosophical Investigations §2). Second, objects of comparison, by their very function, relate to the phenomena they aim to illuminate only through a limited set of features (whether consistent with or in contrast to those of the phenomena in question).

Nevertheless, in our view, contrary to Kuusela’s position, this does not exclude the possibility of treating certain models as adequate descriptions of aspects of the reality under consideration. Properly understanding this issue requires recognising that whether a given description is adequate or not depends on context; in one context it may be adequate, while in another it may not. As Peter Geach (Geach 1957, pp. 33–44) and later John McDowell (McDowell 1996, pp. 6–7) argued, the richness of a fact is inexhaustible, and its specific meaning depends on the perspective we adopt—in other words, on the purpose we are pursuing. For Wittgenstein, the purpose of defining the context upon which the adequacy of a description depends is always the resolution of a specific philosophical problem at hand. Thus, when discussing the adequacy of a description of a phenomenon in the context of Wittgenstein’s idea of objects of comparison, we are by no means referring to the illusion of context-independent completeness on the part of that description.

Thus, while we agree with Kuusela that Wittgenstein does not aim to provide maximally exhaustive descriptions of various real phenomena, we disagree with his assertion that Wittgenstein does not aim to provide adequate descriptions of such phenomena. Wittgenstein does not engage in the former task because it belongs to the domain of the empirical sciences, not philosophy, which examines phenomena only insofar as they are sources of conceptual problems. However, the latter task—providing an adequate description of certain real phenomena—can contribute to solving various philosophical problems. For instance, an adequate description of how we actually use the word “meaning” can free us from the illusion that meaning is some kind of abstract object.

From the perspective of philosophy, an adequate description of the actual use of the word “meaning” does not necessarily need to rely on empirical studies, such as how frequently the word “meaning” is used in different contexts. Instead, it seems sufficient in this case to provide a few standard ways in which we paraphrase sentences containing the word “meaning”.

It is also worth noting that Wittgenstein does not aim merely to present various possible languages but seeks to clarify how our language functions—a goal that is impossible without providing adequate descriptions of its functioning. Once again, however, let us emphasise, following Baker, that adequacy here does not mean completeness; it can be achieved equally well by invoking counterfactual models. Wittgenstein’s aim is not to develop a general theory of language but rather to resolve the specific philosophical problem troubling him at a given moment. He does so by providing a perspicuous (übersichtliche) description that adequately represents the problematic aspects of the functioning of the particular fragment of language under consideration.

5. Models of Religious Forms

As we have stated, our main goal is to clarify the conceptual relationship between ritual and narrative—metaphorically speaking—to chart a path through the philosophical labyrinth from one to the other, thereby severing the roots of grammatical misunderstandings. We will now attempt to achieve this goal by presenting a series of increasingly complex models of religiosity. However, before we proceed to describe the individual models, we must explain their role, as well as the function and form of the sequence as a whole.

First, the models we describe do not necessarily need to have been exemplified in any historical community. Let us again emphasise that Wittgenstein’s method does not consist solely in juxtaposing facts, but in comparing facts with counterfactual possibilities determined by our concepts. Second, the classification generated by the concepts we use to describe the models is significant in relation to the problem we are considering; a different problem would likely require a different classification—one that, for example, distinguishes more or fewer stages. Third, the sequence of stages is structural rather than temporal: they can be arranged from simpler to increasingly complex forms, but this order does not necessarily reflect the historical evolution of religious beliefs. Fourth, the description of the features of the models should not be surprising to us; it is not meant to inform us of something we do not already know. Once again, let us stress that their perspicuous representation is intended to enable a proper understanding of the conceptual relationships obtained within the phenomena under consideration and to eliminate philosophical illusions surrounding them.

5.1. The Model Reduced to Reactions and Rituals

We refer to this model as reduced because its cultivation by an actual human community is highly unlikely, except perhaps under artificial conditions imposed by cruel experimenters. Language is an inherent part of natural social life, and it seems strange to leave some aspects of it outside the practices of naming and describing. However, it is conceptually possible: one can imagine that certain things, perhaps due to their traumatic nature, might remain permanently in the realm of what is unspoken.

Wittgenstein referred to the phenomenon of hesitating to pierce the eyes in a photograph of a loved one with a pin; we recoil from doing so, even though we know, in theory, that we are merely destroying a piece of paper. Such primitive reactions can remain entirely outside any discourse. Imagine a member of a tribe discovering a cave in the forest that, for some reason, evokes in them an undefined sense of fear. Despite its potential utility, the cave becomes a forbidden space, one they do not enter. Now imagine that this cave becomes forbidden to the entire tribe. Its members know where it is and recognise that it could serve as a shelter at night, yet they avoid it—initially due to a primitive reaction of undefined fear, perhaps triggered by the shape of the entrance, the smell of a wild animal, or some other factor. Over time, avoiding the cave becomes a silent custom of the tribe, and crossing its threshold is subject to a prohibition that is neither articulated nor spoken about. Similarly, one might imagine a prohibition arising from a primitive reaction of awe: members of the tribe encounter an extraordinarily beautiful clearing or a peculiar tree, and this place becomes forbidden to them.

There are theories suggesting that the phenomenon of religious awe or dread is intrinsically linked to a relationship with a certain—perhaps only imagined—person (cf. Marion 1995). It is also worth noting that all of Wittgenstein’s examples of primitive reactions are somehow connected to interpersonal relationships. However, this is an issue we do not need to settle definitively: it is (conceptually) possible that a form of worship exists in which even the vague awareness of a personal presence does not arise. In the examples we have discussed, though, it is conceivable that the members of the community sense an undefined personal presence in the cave or the clearing. Yet, they lack the linguistic means to refer to this presence.

The example of a forbidden place can be further developed as follows. Let us assume that the tribe reacts in a specific way to the breaking of the prohibition: anyone entering the forbidden place is killed. Furthermore, we can imagine that this killing takes on a ritualistic character. The transgressor is captured, bound, and then brought to a special location, such as a large flat stone, where a designated member of the tribe executes them by slitting their throat with an obsidian knife. Of course, for this act to be called a ritual, it must constitute an established practice of the community; in other words, it must occur with some regularity. It becomes difficult to imagine that the development or execution of all these actions could take place entirely without words. However, we might allow those performing the ritual killing to speak to one another, exchanging remarks such as “Tie his hands tighter” or “Lay her further to the right”. We might also imagine that certain words are part of the ritual itself. The executioner, at the climax of the ritual, might solemnly describe each action in detail: “I now raise the knife”, “I now slit their throat”, and so on. The key point, however, is that these expressions and words must function in normal, non-ritual contexts as well. The assumption of the reduced model is the absence of a specific vocabulary that the community uses to speak about the ritual. While the distinction between the sacred and the profane exists in the tribe’s practice, it cannot be explicitly articulated.6

5.2. The Model of Conceptualised Rituals Without a Narrative

The conceptualisation of a ritual does not consist merely of the appearance of certain utterances within its framework. Rather, it involves the development of a specific vocabulary associated with the ritual. In the case of the forbidden cave or clearing, this might entail the concept of a special type of place, which could mean that the place is “different from all others”, “set apart from use”, or indeed “forbidden”. In the case of ritual sacrifice, a concept of a particular kind of wrongful act might emerge, along with a corresponding notion of a specific type of wrongdoer.

How can one recognise that a special vocabulary has been developed? This does not necessarily require the creation of new words or expressions. It is possible that a religious concept could be expressed using an existing word, but in a way that differs from its everyday usage. This issue can be elucidated by referring to the previously presented example of the forbidden cave. We can imagine the scenario evolving such that a tribe using the word “cave” in their daily language also develops a distinct, specialised usage of the word. This specialised usage might be marked by a change in intonation—for instance, lowering the voice or slowing articulation—when referring to the specific forbidden place. In its religious sense, the word would have different permissible collocations than in its ordinary meaning. For example, it might not be acceptable to use it in conjunction with the adjective “musty”, which would constitute a grammatical prohibition in Wittgenstein’s sense of the term.

It should be noted that a group of people who speak about their surroundings using, for example, common nouns,7 will naturally employ these nouns to name special places, such as a forbidden cave or a forbidden clearing. Indeed, the special way of pronouncing these common nouns may belong to the complex of primitive reactions associated with these places: fear or awe is often accompanied by changes in articulation.

In our imagined scenario, the word “cave”, when spoken in the context of a forbidden place, must detach from its common-noun function and become a proper noun with a specific usage distinct from that of a general term. Until this occurs, we remain in a situation of unconceptualised reactions and rituals. One could say that the transition from speaking an ordinary word with solemnity to the creation of a new concept is analogous to the transition from an individual primitive reaction (even if exemplified by many people) to a ritual. In this transition, an individual negative reaction to the statement “this cave is musty” transforms into a socially sanctioned condemnation of blasphemy.

A sacred conceptual structure can be minimal, consisting of just one special word, or it can be more elaborate, with specific terms expressing various aspects of worship. It does not, of course, constitute a separate “religious language”, but is instead part of ordinary language, fully integrated with non-religious concepts. This integration, incidentally, allows for the mutual enrichment of both conceptual domains through the mechanism of metaphorisation (see Lakoff and Johnson 2003). We can imagine that this mechanism might operate even without any religious narrative: with a few interrelated concepts such as “special cave”, “special prohibition”, and “violation of the special prohibition”, relationships within this domain could be mapped onto another domain, thereby illuminating some other, non-religious phenomena. Similarly, certain conceptual structures from, say, the domain of social relations might be mapped onto the religious domain, giving rise to new concepts (still without necessarily developing a narrative).

We noted earlier that, even within a non-conceptual model, it is possible to imagine associating a forbidden place with someone’s presence. However, only in a conceptualised one is it possible to assign a name to this mysterious presence. For example, the statement “Hyzi lives here”—spoken in the context of a forbidden place that is not home to any visible or audible being—indicates that the language contains at least one religious concept. We might be tempted to say that this word is the name of a spirit, demon, or deity, but for now, we are speaking only of one mysterious cave or clearing where someone invisible resides. In such a situation, general terms are not yet applicable.

However, we can imagine that the tribe identifies not just one but several special places—some primarily evoking fear and others awe. The presences associated with these places might elicit different sentiments, which would naturally lead, through naming, to the emergence of, say, two categories of invisible beings. These would be general terms functioning as religious concepts.

5.3. The Model of Rituals with a Narrative

The recognition of deities or spirits with specific attributes (possibly distinctive of some religious context or other), and belonging to particular types, does not necessarily require any religious narrative. Conversely, one can imagine a religious narrative that concerns something other than personal beings. What is such a narrative? It is a story about an event or process that either occurred once, will occur at some point, or recurs repeatedly (e.g., cyclically).

Humans live deeply immersed in social relationships, and some researchers suggest that they project their knowledge of these relationships onto the sacred realm, which, as a result, is typically populated by personal beings. Furthermore, interactions occur both between these beings and between them and humans, giving rise to a narrative. Before exploring this path further, however, let us consider how a narrative might exist without involving persons.

Once again, let us use the example of the cave. Suppose it is associated with a sacrificial ritual in which a special individual (a priest) kills a goat on a special table (an altar). This ritual takes place annually at the Winter Solstice. We are not concerned here with explaining the origin of this ritual, but with imagining the following stages of this model: if none of the elements of the ritual—the cave, the special individual, the special table, the sacrificial goat, or the solstice day—has a name specific to the context of these practices, we are dealing with a reduced model; if such special names—or other parts of speech—emerge in the language, but there is no story associated with them, we are dealing with a conceptual model without a narrative; and if a story also emerges, then we have a model with a narrative.

What a narrative adds is a form of internal justification for the ritual within the community of believers. For instance, the cave in question might be linked to the daily and annual journey of the Sun—it might, for example, lead to an inaccessible place where the Sun rests—and the ritual could serve to “feed” the Sun at its weakest moment, giving it the strength to climb higher across the sky once more.8 The narrative connects all the elements of the ritual: the cave becomes the Sun’s resting place, the priest its servant, the sacrificial table its place of nourishment, and so on. These conceptual elements, previously connected through the practice of the ritual, are now interconnected within the story.

There is no quasi-causal relationship here: the ritual sense does not translate into the narrative sense by any rule. When the narrative emerges, it may encompass the entire sacred sphere, seeking to sanction the presence of each of its elements. However, some aspects of the community’s ritual life might remain on the margins if their inherent meaning is too distant from that of the narrative.

Narratives play another crucial role in religion: they facilitate the introduction of children into ritual practices. During childhood, there is a stage when children begin to ask their caregivers countless questions about the surrounding reality. These questions—though they may seem trivial or irritating from an adult perspective—serve an important function in a child’s intellectual development: they help children learn to distinguish between valid questions and those that are silly or unnecessary. Naturally, children who participate in or witness religious rituals will also ask questions about them: what their purpose is, why certain actions are performed in a particular way, and so on. It could even be argued that children’s questions are a fundamental reason why models without narratives feel somewhat artificial: people would need to invent some kind of story to explain ritual actions to children.

There is yet another reason why the transition from a conceptual model to one supplemented by a narrative seems natural. As Paul Ricoeur (Ricoeur 1965) observed, the dimension of linguistic meaning allows individuals to transcend the limitations of a specific point of view,9 introducing a distance between the person and reality and thereby enabling reflection. In the context of rituals, this means that naming elements of the sacred reality removes their immediacy, rendering them mediated by language and creating a space between the sacred and the community of believers. This space invites the establishment of a new, linguistic relationship with the sacred, and this is precisely what a religious narrative represents.

5.4. The Question of the Role of Theology

It seems worthwhile to complement our reflections on religious narratives by addressing the issue of the contribution of theology. It is often said that mature religious systems include a theological layer. But what is theology? In its broadest sense, it is conceptual work on the religious narrative—a kind of more or less systematic reflection upon it. Engaging in such reflection requires a degree of distance from the narrative itself: a minimal distance is necessary for conceptual work to even begin, and the process of reflection—making the narrative an object of contemplation—further increases that distance.

An individual deeply immersed in a religious narrative may not even realise that they hold certain beliefs related to that narrative, as long as they exist within a community of similarly engaged individuals. The transition from an awareness of the religious narrative to an awareness of holding religious beliefs10 is the first step of theological reflection and, simultaneously, the first sign of separation from the direct experience of the narrative.

It is possible that reflection might arise spontaneously, without external provocation: a person might ponder various aspects of their life, including their religious life. However, certain situations may also prompt this reflection. One such situation is the experience of glaring injustice—something happening that should not happen, such as an individual or collective misfortune that is perceived (at least subjectively) as undeserved. For instance, a sacrifice made to the god of rain does not bring rain, and the drought persists. Despite performing the most sacred rituals, a plague devastating the tribe does not relent. A tribunal of priests delivers an egregiously unjust verdict.

Another situation that may provoke reflection is encountering followers of a different religious narrative or religious sceptics. Yet another is encountering discrepancies within the narrative itself, such as the circulation of different versions of the story within the community of believers. Naturally, such experiences might lead to the complete abandonment of one’s previous faith. Alternatively, they may lead to its reconstruction or systematisation, with the first step in this direction being reflection on one’s religious beliefs.

Conceptual work on a religious narrative can take various forms and serve different purposes. It may involve codifying one version as canonical or establishing that there is a shared, immutable foundation for many acceptable alternative variations within the same belief system (already recognised as belonging to a broader community). It may serve an apologetic purpose, involving the creation of a set of beliefs marked by internal logical consistency or the search for arguments, independent of the original narrative, to support certain core beliefs derived from it. It may also involve internal critique leading to a reform of the faith: modifying certain beliefs, questioning the legitimacy of parts of the narrative, or even completely discrediting and rejecting them. In extreme cases, it may result in a critique of the religious belief system as a whole and its ultimate abandonment.

The intellectual reworking of a religious narrative can be an individual endeavour: someone may reach certain conclusions for themselves and adopt a particular worldview based on them. However, the results of individually conducted reflection can be shared—through conversations or publications—and may thus become part of a communal process of determining a religious (or non-religious) stance. Reflection can also be conducted in dialogue, which might take on an institutionalised form (e.g., a council or synod in the Christian tradition). In such cases, a collective body may engage in codificatory or revisionary reflection on the religious narrative, develop an associated system of beliefs, and expand or modify it.

Let us return to the moment that prompts theological reflection. It seems there is an analogy here with the moment that leads to the creation of a religious narrative. While a narrative is an effort directed at strengthening the relationship with the sacred in a new context, one shaped by the distance introduced by linguistic conceptualisation, theological reflection seeks something similar in the context of a new kind of distance arising from encountering otherness. This otherness is not so much ritualistic—in this model, it no longer holds such significance—but rather narrative in character.

Encountering a foreign religious narrative opens up new conceptual possibilities and, moreover, implants a sense of “otherness within ourselves”, creating a greater distance between us and what is sacred to us. Thus, there arises once again a tendency to bridge this distance, but this time in the form of a universally recognised discourse. The goal of this discourse is to lead “unbelievers” to submit to our deities, thereby neutralising the otherness within ourselves. This is precisely what rational theological discourse represents: an attempt to draw on justifications external to a given system of narratives, appealing to outside authorities.

6. Conclusions

The set of models presented above primarily serves to illuminate the rationale behind the existence of religious vocabulary, narrative, and, ultimately, reflection on that narrative. Let us once again emphasise the necessity of distinguishing between the question of reasons and that of causes: the latter involves establishing facts through empirical investigation, whereas the former pertains to relationships between concepts and is typically posed in situations of conceptual confusion, which can only be clarified through philosophical inquiry.

The reason for the emergence of specific religious vocabulary lies in the desire to linguistically demarcate the sacred sphere. Theoretically, it might be possible to set it apart by treating the sacred as something about which one does not speak but remains silent. However, in the practical life of a community, any institutionalised religiosity inevitably encourages some degree of discourse.

The reason for constructing a religious narrative arises from the questions posed about sacred activities. A community provides itself with answers to why it performs a given ritual and what change that ritual brings to reality. As we have noted, it is very difficult to imagine communal ritual activities without such answers, as these questions are often asked by children, who, through such inquiries, come to understand the world.

The reason for engaging in theological reflection lies in disagreements over the interpretation of religious narratives, as well as confrontations that may involve their various versions or even entirely different narratives passed down in different cultures. The outcomes of such reflection can include identifying reasons to support or reject a particular narrative, reconciling different narratives with one another, or transforming a narrative into a set of beliefs.

Our models here are objects of comparison in the sense introduced by Wittgenstein. We have explicated this sense by referencing—and partially critiquing—the interpretation proposed by Kuusela. As stated at the outset, these objects were intended to dispel certain philosophical misunderstandings, primarily concerning the role of religious narratives. We identified two forms of such misunderstanding, both proposed by Wittgenstein at different stages of his project. The first was a full-fledged reductionism, most likely rooted in the Tractarian idea of analysis. According to this view, no authentic religious dispute was possible, as the meaning of all religious statements was inherently rooted in ritual activities (cf. Wittgenstein 2018, pp. 32, 44). The second carried only a trace of reductionism and was later adopted and developed after Wittgenstein’s death by the “Wittgensteinian fideists”, who, furthermore, presented it as the result of applying the critical method of the Philosophical Investigations to the specific domain of religion. In this view, theology was merely the grammar of religious expressions, incapable of resolving any substantive issue (cf. Rhees 1997, p. 44).

With respect to the first, strongly reductionist form, our set of models demonstrates that a religious narrative, though connected to the ritual activities that serve as its reason (at least as the reason for posing questions), is not subsumed within the ritual. On the contrary, it requires achieving a certain distance from the ritual. For this reason, its meaning is distinct from the meaning of the ritual itself, and these two orders of meaning, coexisting within a single tradition, may not be entirely consistent with one another.

As for the second form, its most evident problem—as we have already suggested during our discussion of it—is its characteristic of being closed to the possibility of disputes between different religious narratives. We have attempted to show that such disputes are precisely what creates a certain distance from the religious narrative, enabling theological reflection. Pichler and Sunday Grève (2024) point out the reluctance with which Wittgenstein, late in his life, conceded the existence of “honest religious thinkers”. His perspective on religion generally did not consider the possibility of reflection on religious beliefs.

The deeper rationale behind the impossibility of disputing beliefs lies in the existentialist and strictly individualistic conception of the religious message as a call addressed to the individual, who may accept or reject it. We would like to emphasise once more that Wittgenstein—and following him, the “Wittgensteinian fideists”—points to a genuinely significant aspect of the phenomenon of religion here. Without commitment and practice, it is difficult to claim that anyone truly adheres to any religious faith. It can indeed be said of theologians who focus solely on their systems of beliefs that they “gesticulate with words” (Wittgenstein 1984, p. 85). However, an exclusive focus on the immediacy of receiving the message also presents an incomplete picture of the phenomenon of religion.

A possible line of defence for proponents of “Wittgensteinian fideism” against these conclusions might be to attack our objects of comparison. The conceptual models we have presented could be regarded as distorted due to what might be claimed to be our flawed philosophical (in a pejorative sense) conception of religion. In other words, someone might argue that these models are not examples of perspicuous representation because they were constructed to fit a predetermined thesis.

In response to this critique, it can be noted that, as Wittgenstein observed, there is no single order, but rather many possible orders (Wittgenstein 2009, §132). Thus, it can be said that our approach represents only one possible set of models of religiosity. Another order might, for example, depict a series of models divided into two parallel developmental lines, one culminating in a “transactional” understanding of the relationship with a deity, so that it is effectively a “superstitious model” in Wittgenstein’s sense, and the other leading to a “model of trust”. While not denying that this second set of models—broadly aligned with Wittgenstein’s intentions in his remarks on religion—offers some illumination of the phenomenon, we argue that treating this order as the only possible one presents the phenomenon in a one-sided manner.

What our proposed order highlights is the possibility of engaging in disputes about religion. Although such disputes can be uncomfortable or even dangerous at times, they are an intrinsic part of human communities—one might even say, part of the “human form of life”. There is no reason to believe that the religious sphere should be exempt from this dynamic, and historical evidence, as well as observations of contemporary social reality, clearly show that it is not and never has been.

Wittgenstein, in his remarks, declared the irrelevance of historical proof games where the message of the Gospel is concerned. Pichler and Sunday Grève (2024) comment that even if this is so, truth games have a role to play in religion. We might add that why games also play a role, in which religious narratives are conveyed as intersubjectively communicable stories, the meaning of which we occasionally find ourselves disputing.