Semi-secularity and simultaneity of interpretation

It is a well-established sociological fact that about 50 per cent of the population of Europe are neither practising and believing adherents of traditional religion nor fully secularized unbelieving atheists. Since it defies the traditional dichotomy between religious and non-religious, this kind of religious identity is often referred to as ‘fuzzy’, a term coined in an influential article by David Voas (Reference Voas2009). Burén, however, prefers the term ‘semi-secular’ and for the sake of convenience we adopt her terminology in what follows.Footnote ²

In Sweden, the semi-secular part of the population is larger than the European mean. Erika Willander estimates this group, which she fittingly calls the religious mainstream, to make up as much as three quarters of the population:

Given these figures, it is certainly a highly interesting question what people in the semi-secular religious mainstream of Sweden think about religion and related matters. What do their non-naturalistic worldviews look like? While the topic has long been neglected in favour of research on traditional religion, it has recently begun to receive the academic attention it rightfully deserves. The findings of these studies include widespread but vague and unarticulated beliefs in ‘something more’ beyond the mundane (Willander Reference Willander2013; Idem Reference Willander2020) and a similarly unarticulated folk spirituality centred on nature (Thurfjell Reference Thurfjell2020).

Burén's Living Simultaneity is part of the new literature on the semi-secular religious mainstream of Sweden. Burén stresses simultaneity and her main point is that a semi-secular person is in some sense both religious and secular at the same time. This basic approach differs from those of most other theorists, who regard semi-secularity as somewhere in-between religiosity and secularity.

The study builds on in-depth interviews with twenty-eight residents in a neighbourhood of central Stockholm, twelve of whom were interviewed twice. Most of the study is about how these semi-secular individuals understand the term ‘religion’ and about their religious identity. However, in the last part of the study Burén turns to investigate the participants' worldviews by discussing their interpretation of events which they themselves considered to be ‘out of the ordinary’ (Burén Reference Burén and af2015, 20).

Burén's fascinating finding is that when it comes to the interpretation of extraordinary events in their own lives, several of the semi-secular interviewees did not accept a naturalistic interpretation as the only explanation. Furthermore, they were either unwilling or unable to bring their non-naturalistic interpretations of these events into coherence with their basic, naturalistic picture of the world.

According to Cristine Legare et al. there are at least three ways naturalistic and supernatural explanations can be combined. In integrative thinking, the explanations are integrated into a single coherent view. In target-dependent thinking natural and supernatural explanations are invoked to address different aspects of the event which is to be explained. Only in synthetic thinking are they held together without any effort to make them rationally consistent (Legare et al. Reference Legare, Evans, Rosengren and Harris2012, 783).Footnote ³ Since Burén stresses the unmediated simultaneity of the different kinds of explanation, it is clear that the phenomenon Burén calls ‘simultaneity of interpretation’ corresponds to Legare's synthetic thinking (but for the sake of clarity, we continue to use Burén's term). Now, to introduce this simultaneity of interpretation more properly, we present in detail the three most salient cases in Burén's account: Göran, Victor, and Sara.Footnote ⁴

Göran is a middle-aged man with a secretive manner, who according to Burén has experienced much which falls outside the everyday norm. She relates a story from Göran's youth in which he crashed his motorcycle (which was completely destroyed) but managed to escape unharmed. He even landed on his feet after vaulting over the car he collided with (Burén Reference Burén and af2015, 181–183). What is particular about Göran is that he refuses to choose between a natural and supernatural explanation of the event:

Rather than choosing one of the two explanations, Göran entertains both simultaneously. He claims that in doing so he is being ‘nuanced’, and that this is in fact the natural working of the mind (ibid., 203).

The second example is Victor, a man in his forties. Like Göran, he has experienced things which are out of the ordinary and is reluctant to accept a purely naturalistic explanation of these events. Examples include a medium passing on a message from a dead friend, and a deep feeling of being in the right place after moving to a new neighbourhood. However, unlike the case of Göran, who accepts the ambiguity, Victor struggles with the validity of his experiences:

In his rational thinking, Victor is drawn towards a naturalistic picture of the world, but emotionally he longs for meaning and hopes for an afterlife. In his struggle, he actively tries not to think rationally about the extraordinary events which he finds deeply meaningful (ibid., 197–199).

The third example is Sara, a young woman torn between different interpretations of reality. Thinking about religion in terms of ‘ready-made packages you either embrace or reject’, Sara is disturbed by what philosophers call the problem of religious diversity (i.e. how can one believe in the truth of one religion while there are other competing religions with contrasting claims about reality). She repeatedly confesses that she does not know what to believe. Sara's struggle is tied to the fact that her grandmother has recently passed away, which has brought to the fore the issue of whether there is an afterlife (ibid., 200–202). Here is a telling quote related by Burén:

Unlike Göran, who does not feel the need to choose, and Victor, who struggles not to choose, Sara is simply unable to make a choice between competing interpretations of the world.

We will show that from a non-doxastic perspective Göran is in fact fully rational in living with simultaneity the way he does, and there is nothing preventing either Victor or Sara from embracing the possibility of an afterlife even though they do not really believe in it. First, however, we will take a closer look at Burén's own interpretation of this simultaneity of interpretation. Unlike our non-doxastic interpretation, Burén's reading presumes the irrationality of the subjects involved.Footnote ⁵

The incongruent interpretation

According to prominent sociologist of religion Mark Chaves (Reference Chaves2009), it is a fallacy to expect congruence in religion, either between beliefs or between beliefs and actions. While not denying that religious congruence exists, he claims that coherence is the exception and incoherence the norm. Chaves calls for a sociology of religion which does not presuppose congruence in belief or any straightforward connection between religious belief and action. Burén's analysis of semi-secular simultaneity is very much in line with the interpretative thrust suggested by Chaves.

To understand simultaneity of interpretation, Burén mainly turns to the work of Alfred Schütz, a twentieth-century phenomenologist working in the tradition of Husserl.Footnote ⁶ Schütz's work is extensive, and Burén's presentation of Schütz is explicitly simplified. While perhaps a Schützian scholar might have had something to say about Burén's interpretation, we will take her reading of Schütz's theories at face value. It is, after all, Burén's reading of the simultaneity of interpretation that interests us, and if she has in any way misinterpreted Schütz, that is irrelevant for our present purposes.

According to Schütz, we all live in multiple realities or provinces of meaning,Footnote ⁷ with distinct cognitive styles (like the world of everyday life, the world of dreams, the world of religious experience, of scientific reason, of drugs, of insanity, and so on). Burén stresses the incoherence between these finite provinces of meaning, and that a ‘shock’ or ‘Kierkegaardian leap’Footnote ⁸ is required to move between them:

The stress on incoherence is perhaps even stronger in Schütz's original account:

Another feature of Schütz's theory emphasized by Burén is the role played by ‘epoché’. Epoché is a method of principled suspension of belief central to phenomenology. According to Schütz, there is also everyday epoché and every reality has its own variety. It basically means that when we are in a specific province of meaning we suspend doubt that what we experience in this realm is real; any such existential doubt only hits us afterwards when we have shifted to another, contrasting reality. According to Burén, it is through epoché that the semi-secular navigates their multiple, incoherent beliefs:

Schütz intended his theory to be universal in scope, but Burén seems somewhat ambivalent regarding its applicability. At one point she explicitly states that she only accepts it as a local theory for explaining her empirical findings and not as a global theory of human cognition (ibid., 186). At another point (see quote below) she seems to accept it on quite general terms. Regardless, it is clear that she regards it as a fitting description of her semi-secular respondents' views of the world:

It seems an inescapable consequence of Schütz's theory that subjects are in a substantial sense irrational. There are strong intuitions in favour of regarding inter-province coherence as a requirement of rationality, and few epistemologists would deny the basic assumption that incoherence requires belief-revision if the subject is to adhere to the demands of epistemic rationality.Footnote ⁹ Compartmentalization is rationally unacceptable. For example, a person who is normally an atheist but who believes in God on Sundays is clearly blameworthy in an epistemic sense. Likewise, there would be something wrong with a person who believes in scientific explanations when at work while explaining things in terms of spirits and magic in her free time.

However, a qualification is needed. It is an empirical fact that our cognitive systems operate with some degree of compartmentalization, since it is not in our power instantly to assess the coherence of every new belief we acquire with every previously held belief. Therefore, new beliefs are often at least partly compartmentalized until further reflection occurs.Footnote ¹⁰ This weak sense of compartmentalization is essentially harmless. The contrasting problematic, strong sense of compartmentalization is when a subject is either unwilling or unable to work out a coherent view when confronted with incongruences in her belief-system or when she fails to notice pressing and salient inconsistencies (as in the examples above). In such cases, the subject has failed to meet an acceptable epistemic standard and is therefore epistemically blameworthy.

Schütz's theory, where ‘experiences and interpretations that belong to a certain province may look contra intuitive or even absurd from the viewpoint of another’, certainly suggests compartmentalization in the strong sense. This picture is embraced by Burén, who dedicates much space to explaining how, by different processes of suspension of belief and disbelief, the semi-seculars can move between ‘different realities where they [the different interpretations] fit into the overarching logic of the particular realm.’ (ibid., 187, 193).

Against the incongruent interpretation

We think there are strong reasons to resist a Schützian reading of the simultaneity phenomenon. In philosophy, it is standardly considered a theoretical virtue to follow the principle of charity. According to this principle, we should always attribute rationality to others and interpret what others say with a high degree of charity. If at first a theory sounds irrational or ludicrous, we should search for other interpretations on which it is coherent and makes sense.

Philosophers like Donald Davidson (Reference Davidson and Davidson2009) have argued that the principle of charity is applicable not only in philosophy, but to all communication. He also holds that it is much more than a mere courtesy: it is a basic requirement without which human communication would not be possible. Attributing rationality to others is a basic requirement for making others intelligible and therefore also a basic requirement for understanding. As explained by Jeff Malpas, the principle means that we should assume both that others are coherent in their beliefs and that their beliefs correspond to a shared reality: ‘In fact the principle can be seen as combining two notions: a holistic assumption of rationality in belief (“coherence”) and an assumption of causal relatedness between beliefs – especially perceptual beliefs – and the objects of belief (“correspondence”)’ (Malpas Reference Malpas2019). Of course, it might be questioned if the requirement always to ascribe rationality to others really should be taken at face value. Are there not instances where the irrationality of the other is so obvious that ascribing rationality actually distorts understanding? Granting the possibility that this might indeed be the case, we settle for a weaker version of the principle of charity:

This principle only claims that we should prefer a rational interpretation over an irrational one as long as the irrational interpretation is not better supported by the data. It is thereby far less controversial than Davidson's idea that we should always ascribe rationality whenever such a reading is possible. Indeed, it seems so intuitive that it borders on the trivial. If an equally plausible rational reading is available, what reasons could there possibly be to prefer an irrational reading?

The weak principle of charity suggests that we should reject Burén's Schützian reading of the simultaneity phenomenon if there is an alternative reading which ascribes rationality to the subjects involved and which is at least as compatible with the data as Schütz's theory. In the rest of this section, we will demonstrate that Schütz's theory does not fit the data (we will even claim that it is logically inconsistent with it). In the next section, we will introduce a non-doxastic reading of the simultaneity phenomenon, which ascribes rationality to the subjects involved and which fits the data to a significantly higher degree than Schütz's theory.

Now, why do we say that Schütz's theory does not fit Burén's empirical findings? First, it is a salient feature of Schütz's theory that the different provinces of meaning are incoherent, and that they might even look ‘absurd’ when viewed from the outside. If a Schützian reading were correct, we would expect the subjects of Burén's study to find at least one of their contrasting interpretations of their experiences to be absurd or counterintuitive. However, this does not seem to be the case for any of the main subjects.Footnote ¹¹ On the contrary, what seems to give rise to the simultaneity is the existence of competing explanations which all seem somewhat likely. The most salient feature of Göran's stance is that he does not want to choose between two explanations he finds equally probable. Sara is caught up in the problem of religious diversity exactly because all religious options seem equally plausible to her. A straightforward way to understand why Victor struggles not to scrutinize his special experiences is that he fears that such scrutiny would reveal that a non-naturalistic explanation is untenable. Contrary to Burén's reading, this seems to suggest that Victor would have abandoned the non-naturalistic interpretation if he had found it too unbelievable or absurd. The upshot is that even though the different explanations are mutually exclusive they are still plausible at the same time to the same subject.

Also, in Schütz's theory a subject moves between inconsistent provinces of meaning by making voluntary Kierkegaardian leaps, or by being ‘shocked’. Granted, these shocks might be very minor (like falling asleep), but they are nonetheless required. While it might be true in some cases that their original extraordinary experiences shocked the interviewees, there is nothing indicating that they had to make any leap of faith at the time of the study. To the contrary, it seems that all contrasting interpretations were rationally available at the same time during the interviews. How else to make sense of the idea that they could be entertained simultaneously (Göran) or that a subject could struggle to make a choice between them (Sara)?

Likewise, the notion of ‘epoché’ seems ill-suited to express the respondents' relation to the contrasting interpretations. The respondents do not seem to suspend belief or disbelief to move between interpretations. If one suspends belief in contrasting interpretations (or suspends doubt that one's current interpretation is the correct one), one cannot very well at the same time struggle to make everything fit. Victor's and Sara's struggles seem the antithesis of epoché. While Göran does not struggle, it is not because he suspends belief to move between interpretations. It is because he has suspended judgement permanently – a situation in which no epoché is required.

Nothing seems further from Burén's empirical findings regarding simultaneity than a theory which demands Kierkegaardian leaps or moves between provinces of meaning by the use of ‘epoché’. It is hard to see how the Schützian picture, where the subject only visits a single province at any given time, could ever allow for simultaneity of the kind encountered by Burén. Not only does simultaneity seem to be something altogether different from compartmentalization, the two stances even seem incompatible at a conceptual level. Clearly, one cannot entertain multiple views simultaneously and be compartmentalized regarding the same views at the same time.

According to the weak principle of charity, we should always prefer an interpretation where the subjects appear rational, given that no theory ascribing irrationality is better supported by the evidence. We think that it is safe to say that Schütz's theory receives little if any support from the evidence. We now purport to show how a non-doxastic reading allows us to regard the subjects as rational, and that such a reading fits well with Burén's empirical findings.

A non-doxastic approach

According to religious non-doxasticism, it is possible to make a positive engagement with religion and lead a religious life even in the absence of outright religious belief. It builds on the assumption that religious belief can be substituted with some weaker cognitive attitude, such as hope (Pojman Reference Pojman1986), assumption (Howard-Snyder Reference Howard-Snyder2017), acceptance (Alston Reference Alston, Jordan and Howard-Snyder1996), voluntary assent (Schellenberg Reference Schellenberg2005) or the like, and that this is enough to make substantial religiosity possible.

We believe that the semi-secular Swedes in Burén's study can be fruitfully thought of as having a non-doxastic stance towards some religious or non-naturalistic propositions. According to this interpretation, we do not ascribe non-naturalistic beliefs to the participants in the study. Instead, we regard them as entertaining epistemic possibilities while lacking outright belief. Rather than saying that Victor believes he got a message from a dead friend through a medium, we would say that Victor regards it as an epistemic possibility that he has received such a message. The difference compared with an interpretation in terms of belief is perhaps clearest when considering the case of Göran. While it would be blatantly irrational simultaneously to believe in both a supernatural and natural explanation of the same event,Footnote ¹² there is nothing irrational in entertaining two contrasting epistemic possibilities at the same time. A non-doxastic reading thereby allows us to understand the subjects of Burén's study as adhering to the standards of epistemic rationality.

Furthermore, we should not assume that people like Göran or Victor relate non-doxastically to a non-naturalistic worldview which goes beyond their interpretations of extraordinary events. While philosophers addressing non-doxasticism often assume that subjects take a non-doxastic stance towards a full-blown worldview, there are no theoretical obstacles to having a non-doxastic pro-attitude towards some vaguer view, or against applying it in a more piecemeal fashion.

While the details can vary quite substantially between accounts, the common core of non-doxasticism can be summarized in two necessary conditions. In short, S has a non-doxastic pro-attitude towards some proposition p if the following two conditions are met:

ND1 relates mainly to the motivation needed to take a non-doxastic stance. It is important to see that unlike belief-based stances, taking a non-doxastic stance is standardly voluntary.Footnote ¹⁴ If you believe p, your belief will dictate what you should do or think, and you really do not have much choice in the matter. But if you only consider p to be an epistemic possibility you are free to act but in no way bound to do so. A positive non-doxastic engagement with p thus requires that the possibility represented by p is deemed good or desirable.

Consider the following example: if you believe it will be sunny tomorrow, you will behave, think, and act accordingly. You really do not have much choice in the matter if you want to remain rational. However, if you lack outright belief but consider it possible that it will be sunny tomorrow you can still act on that possibility if you have the right motivation. You need to desire the truth of p – ‘it will be sunny’. If you do, you can, for example, go on a hike in the hope that it will be sunny. But if you are indifferent you cannot rationally take a non-doxastic stance.

ND2 is the most important condition, since it defines the subject's epistemic stance towards p. It requires that the subject neither believes nor disbelieves p but regards the truth of p as an epistemic possibility. As pointed out by Eklund (Reference Eklund2018), the requirement that the subject must believe that p is an epistemic possibility means that she is not actually belief-less in a strict sense. Since the term ‘non-doxasticism’ might be read to imply otherwise, it would perhaps be more correct to talk about ‘sub-doxasticism’.Footnote ¹⁵ However, ‘non-doxasticism’ is the term commonly used in the literature and to avoid confusion we stick with the established terminology.

Proponents of non-doxasticism have offered many different interpretations of the notion of epistemic possibility involved. Some are content with using epistemic possibility in the widest sense possible, as that which is neither known nor justifiably believed to be false (Schellenberg Reference Schellenberg2005, 8). Others opt for a more restricted sense and require that p have a ‘non-negligible chance of being true’ (McKaughan Reference McKaughan2013, 113) or that p be a ‘live hypothesis’ in the Jamesian sense (Palmqvist Reference Palmqvist2021).

The difference between a wide and narrow sense of epistemic possibility can be illustrated as follows. Imagine that you do not know where your teenage son is at the moment, that it is evening and that you have not seen him since breakfast. It is surely an epistemic possibility in the widest sense that he can be almost anywhere. He could uncharacteristically have acted on a whim and gone on a trip to the countryside (even though he usually is very reluctant to leave the city) or he could have been kidnapped by modern slavers and be halfway around the globe. Or perhaps he was abducted by aliens and taken through a wormhole to the other side of the galaxy. However unlikely these scenarios, they are all epistemically possible in the widest sense (you do not really know that there are no aliens, do you?). Contrasting epistemic possibilities in the narrow sense include the possibilities you would actually consider in this scenario: that he hangs out with his friends at their usual place, or that you simply failed to notice when he returned home from school.

We find it reasonable to suggest that a non-doxastic engagement must be built on an epistemic possibility in the narrower sense if it is to be rational. This allows us to claim that while it is rational to hang on to the live possibility of an afterlife (like Victor) it is not rational to have non-doxastic faith in the flying spaghetti monster.

Before continuing to examine just how well the simultaneity of interpretation fits the non-doxastic framework, some words about the central attitudes of non-doxasticism. Most accounts of non-doxasticism come in terms of propositional faith,Footnote ¹⁶ which is analysed as a composite attitude. In this analysis, the belief-component of traditional faith is supplemented by one of the weaker cognitive attitudes mentioned earlier (such as hope, acceptance, or assumption, etc.). Since it would not be very appropriate to claim that the subjects of Burén's study have faith (with the possible exception of Göran), it is important to see that it is fully possible to spell out the details of non-doxasticism without invoking the ‘faith’ label for its central attitude.

James Muyskens (Reference Muyskens1979) is the first contemporary proponent of non-doxasticism of whom we are aware, and he held that a religious commitment might be based on hope alone. After identifying the common denominator ND1 and ND2 with the standard analysis of hope, Palmqvist (Reference Palmqvist2019; Idem Reference Palmqvist2020) reaches a similar conclusion. Eklund (Reference Eklund2018) goes a step further and claims that all that is required is really ND2, that the subject regards p as an epistemic possibility. While we do not want to go as far as Eklund, we believe that ND1 + ND2, at least in combination with some weak cognitive attitude such as hope,Footnote ¹⁷ assumption, or acceptance, is sufficient for non-doxastic religiosity. We think this ‘faithless’ notion of non-doxasticism can be used to make sense of the simultaneity of interpretation.

Let us now return to the three most prominent examples in Burén's study and take a detailed look at how they fit the non-doxastic pattern. Let us start with Victor, whose stance is a textbook case of a non-doxastic commitment. Remember that Victor has had some extraordinary experiences, the most salient concerning a mediumistic message from a dead friend. The most striking feature of Victor's case is how he struggles with the validity of his experiences, trying not to scrutinize them too much out of fear that they will not stand up to rational analysis. Given that extreme level of doubt, we should be hesitant to ascribe belief to Victor. A more charitable interpretation is that Victor regards the validity of the mediumistic message as a live possibility. He tries to accept that his dead friend gave him a message through a medium and he seems very much to hope that this is the case, which are exactly the kind of attitudes we suggest are sufficient for non-doxastic religion.

It is obvious that the required positive evaluative component required by ND1 is very much present in the case of Victor. Victor is unhappy with the lack of objective personal meaning in the naturalistic picture of the universe, and he struggles to hang on to experiences which, contrary to that picture, suggest that there is meaning in the world. Claiming that Victor ‘desires’ or ‘evaluates positively’ the validity of his extraordinary experiences seems almost an understatement, and it also seems very clear that it is this strong desire which makes him commit non-doxastically to the live possibility that his dead friend really relayed a message to him.

The key characteristic of Göran is that he refuses to choose between a natural and supernatural interpretation of the extraordinary events of his life. This is easily interpreted in terms of epistemic possibility: in Göran's view, both a natural and a supernatural interpretation represent epistemic possibilities. His refusal to commit strongly indicates that he does not really believe either of the explanations, and perhaps one could even suggest that it is the existence of an undefeated alternative explanation which prevents outright belief in this case. Far from being irrational or ‘compartmentalized’, it seems that Göran is only doing what any epistemically responsible person faced with contradictory but equally likely theoretical options should do – he suspends judgement.

In fact, Göran does more than suspend judgement; he seems to accept both interpretations simultaneously. This differs from the standard picture, in which a non-doxastic commitment is directed at a single desired possibility that is too improbable to cause outright belief. However, there is nothing inherently irrational in having a non-doxastic commitment to two incoherent views. A good example of this is a scientist who accepts and is working on two incoherent theories because the evidence is inconclusive (given the apparent inconsistency between quantum physics and macrophysics, this is presumably something all physicists do on some level). It only becomes irrational if the subject has convincing evidence that one of the views is true or much more likely to be true than the other, and this does not seem to be the case with Göran.

What about the emotional component? It seems clear that Göran is emotionally drawn to both possible explanations of his experiences. It is surely somewhat unusual not only to accept but also to desire the truth of two competing theories. Here, Göran again differs from the standard picture. However, there is nothing strange in having conflicting desires, and as long as the evidence remains inconclusive there is nothing forcing Göran to choose between the two interpretations. Here, too, we may compare his situation with that of the scientist, who presumably hangs on to both theories because they are both fruitful and because it would be good if either of them turned out to be true.

Sara, finally, fits the non-doxastic pattern nicely. Having a simplified view of religions as ‘ready-made packages’, she does not believe in any of them, but neither does she express outright disbelief. To her, the different religious ‘packages’ represent different epistemic possibilities, but she cannot choose between them. It also seems clear that the idea of an afterlife and the thought that her dead grandmother can see her from the other side represents an epistemic possibility she desires to be true but does not fully believe in.

Unlike Victor and Göran, Sara has not made any clear non-doxastic commitment. But if she truly desires the truth of the proposition ‘my grandmother watches over me’, or if she determines that one of the ‘religious packages’ presents a desirable picture of the world, she is certainly in a position to make such a commitment.

According to the weak principle of charity, we should always prefer an interpretation ascribing rationality as long as it has at least the same evidential support as any irrational reading. In the previous section we saw how poorly the data fits a Schützian, irrational reading. Since we can now also claim that our non-doxastic, rational interpretation fits the data well, it seems safe to conclude that it should be preferred all things considered.

Concluding remarks

Philosophy of religion is standardly conducted in relation to traditional religion, and religious non-doxasticism is no exception. In this article we have demonstrated how the non-doxastic approach can be fruitfully applied to semi-secular outlooks, thereby suggesting how both the field of non-doxasticism and philosophy of religion in general can be widened.

The exact limits of the applicability of the non-doxastic framework remains to be examined. Recall Chaves's claim that incongruence is the norm in religion while congruence is the exception. In this article we have shown how one such apparent incongruence can be reinterpreted along more charitable lines and given a non-doxastic understanding. It is certainly a live possibility that much of the apparent incongruence mentioned by Chaves could be reinterpreted in this way. It could even prove that having a non-doxastic attitude like hope or acceptance is the norm in religion, and that a belief-based stance is the exception. Of course, this is just speculation, and much work needs to be done in order to explore whether this suggestion has any gravity.

Our discussion has centred on Burén's findings that the worldviews of semi-secular Swedes are multifaceted, containing supernatural elements in combination with a basic naturalistic picture of the world. We have rejected the Schützian interpretation according to which these people have incoherent beliefs and are cognitively compartmentalized. Such an interpretation seems hard to reconcile with the data, since compartmentalization is conceptually inconsistent with the notion of simultaneity itself. In its place, we have offered a non-doxastic reading, in which we do not ascribe supernatural beliefs to the semi-secular subjects involved but regard them as relating to different non-naturalistic epistemic possibilities of high personal significance. Unlike Schütz's theory, the non-doxastic framework both fits the data well and allows us to ascribe rationality to the subjects involved.

Of course, it remains an open question how common views of this kind are among the hundreds of millions who make up Europe's semi-secular ‘religious mainstream’. A handful of interviews from Stockholm is clearly not enough to answer that question. Sociologists are still only scratching the surface when it comes to investigating the worldviews of this important group. The philosophical work of analysing and interpreting these semi-secular views and assessing their rationality has barely started.