Non-belief as self-deception?

Abstract

The suppression thesis is the theological claim that theistic non-belief results from culpable mistreatment of one's knowledge of God or one's evidence for God. The thesis is a traditional one but unpopular today. This article examines whether it can gain new credibility from the philosophy of self-deception and from the cognitive science of religion. The thesis is analysed in terms of the intentionalist and the non-intentionalist model of self-deception. The first proposed model views non-belief as intentional suppression of one's implicit knowledge of God. It is less feasible psychologically but has a good theological fit with Paul's and Calvin's versions of the thesis. This model also helps the argument for the culpability of non-belief. The second model views suppression as a process of subconscious motivated reasoning driven by a desire to avoid an uncomfortable truth. It fits Pascal's view that one's desire for or against God determines whether one sees general revelation as providing sufficient evidence for God. There is some empirical and anecdotal evidence for both models, but obvious cases of non-resistant non-belief present a major problem for the suppression thesis. Also, it is hard to see what might motivate anyone to deceive oneself about God's existence.

Introduction

Let ‘the suppression thesis’ be the theological claim that theistic non-belief results from culpable mistreatment of one's knowledge of God (or one's evidence for God).¹ The essential idea connecting various versions of the suppression thesis is that the lack or substantial distortion of theistic belief is not due to ignorance or shortage of evidence for the one true God, but the result of wilful and culpable resistance to God. The best way to understand the thesis is to consider its biblical basis in the first two chapters of Paul's Epistle to the Romans.

For the wrath of God is revealed from heaven against all ungodliness and injustice of those who by their injustice suppress the truth. For what can be known about God is plain to them, because God has made it plain to them. Ever since the creation of the world God's eternal power and divine nature, invisible though they are, have been seen and understood through the things God has made. So they are without excuse, for though they knew God, they did not honour him as God or give thanks to him, but they became futile in their thinking, and their senseless hearts were darkened. [. . .] When gentiles, who do not possess the law, by nature do what the law requires, these, though not having the law, are a law to themselves. They show that what the law requires is written on their hearts, as their own conscience also bears witness, and their conflicting thoughts will accuse or perhaps excuse them. (Rom. 1:18–21, 2:14–15, NRSV)

While in this text Paul is attacking polytheism rather than atheism, many Christian theologians have taken it as an account of all forms of theistic non-belief. These chapters serve as the locus classicus for four interconnected theological ideas:

• • General revelation: God has provided sufficient evidence of his existence and basic attributes in the natural world and of his moral law in human conscience (Demarest 1982; Sudduth 2016). While theologians have offered long lists of generally revealed divine attributes, Romans 1–2 depicts God revealing himself as personal, powerful, super-knowing, invisible, eternal, and morally good Creator who requires worship and obedience.

• • Natural knowledge of God: As a result of general revelation, all humans possess either sufficient evidence of God or have natural knowledge of God (Dowey 1994; Plantinga 2000; Oliphint 2019). However, by ‘all’ theologians typically seem to mean all cognitively mature and healthy humans (Launonen 2022a). This knowledge is considered primarily as implicit instead of explicit, as intuitive instead of reflective, and as immediate instead of mediate.

• • Noetic effects of sin: Original sin has cognitive consequences. The most drastic one is that people suppress their natural knowledge of God or their evidence for God so that they become blind to God's reality (Moroney 2000; Peels 2011). However, humans still retain implicit knowledge of God or of the evidence for God.

• • Inexcusability and culpability: Since all people know (or have once known) the one true God (or the evidence for God), theistic non-believers have no excuse for not believing in and worshipping God.

The suppression thesis is a traditional one in Christian theology and philosophy in general, and in historical and contemporary Reformed thought in particular.² For example, the Canons of Dort state the following:

There remain, however, in man since the fall, the glimmerings of natural light, whereby he retains some knowledge of God . . . But so far is this light of nature from being sufficient to bring him to a saving knowledge of God, and to true conversion, that he is incapable of using it aright even in things natural and civil. Nay further, this light, such as it is, man in various ways renders wholly polluted, and holds it in unrighteousness, by doing which he becomes inexcusable before God.³

In contemporary philosophy of religion, the suppression thesis is central to what Max Baker-Hytch (2023) has recently called ‘sin-based approaches’ to the problem of divine hiddenness (e.g. Wainwright 2002; Azadegan 2013; Wood 2013; Nemoianu 2015; Taber and McNabb 2018). J. L. Schellenberg's hiddenness argument is famously based on the incompatibility between the existence of a perfectly loving God and the existence of non-resistant non-belief (Schellenberg 2015; Howard-Snyder and Green 2022). Many responses to the argument claim that God may have good reasons to withhold sufficient evidence of his existence at times even from persons open to a relationship with God. God thus allows non-resistant non-belief at least temporarily. Sin-based approaches, however, argue either that non-belief is always resistant or that it is sometimes resistant. Baker-Hytch calls these universal and partial accounts, respectively. We may similarly speak of a universal and of a partial suppression thesis. The partial thesis is of little use in responding to Schellenberg's argument, while defending the universal thesis is a daunting task. However, theologically conservative thinkers may still have reasons to emphasize suppression as a cause of non-belief. For one, scripture and theological tradition teach it. Moreover, the idea that non-resistant non-belief is widespread in human history has serious theological implications. These will be discussed in the end of the article.

According to Baker-Hytch, sin-based responses to the hiddenness argument suffer from a serious deficiency: ‘Beyond providing a description of what a notable strand of the Christian tradition has historically asserted about fallen human nature, these authors don't obviously offer anything that could be regarded as positive evidence that this really is the way things are with humans’ (Baker-Hytch 2023, 3; italics mine). Where should we look for such positive evidence? This article points to the psychology and philosophy of self-deception as a promising source. While the contemporary philosophical debate on self-deception has roamed more than half a century (DeWeese-Boyd 2021), and while William Wood (2013, 2021) has applied it in his analysis of the nature of sin, it has not been used to support the suppression thesis. This is surprising, since the thesis is effectively a claim about self-deception: non-believers are said to lead themselves into the false belief about God's non-existence without fully realizing what they are doing. Another promising source is cognitive science. Research has shown that our belief-forming processes are largely hidden from our conscious awareness, and that our explicit beliefs often conflict with our implicit beliefs. Moreover, cognitive science of religion suggests that belief in gods and spiritual beings is cognitively natural (Barrett and Visala 2018). So, while most atheists and agnostics are obviously not lying about their explicit non-belief (they really believe they don't believe in God), cognitive science might offer evidence of implicit belief in God among them.

This article considers whether the suppression thesis can gain support from these sources. I discuss two approaches to self-deception, the intentionalist and the non-intentionalist model. Discussion of each model is followed by an analysis of theological compatibility, that is, whether the model of self-deception under discussion provides a theologically fitting analysis of the psychological process assumed by a certain version of the thesis. Since the thesis is first and foremost a theological claim, and most sin-based approaches to the hiddenness argument draw especially from the writings of Calvin and Pascal, it is important to show how the claims of influential theologians' map onto the contemporary debate on self-deception. Most importantly, I also consider empirical and anecdotal evidence for theistic non-belief as a form of self-deception. Finally, I discuss how non-believers might be morally responsible for deceiving themselves and what could possibly motivate them to do so (assuming they initially know God exists or have good evidence of God).

The intentionalist model of self-deception

Self-deception minimally ‘involves a person who seems to acquire and maintain some false belief in the teeth of evidence to the contrary as a consequence of some motivation, and who may display behaviour suggesting some awareness of the truth’ (DeWeese-Boyd 2021). Philosophers discuss two basic models of self-deception: the intentionalist and the non-intentionalist (or ‘deflationary’) model. The intentionalist model views self-deception as analogous to interpersonal deception. S1 deceives S2 only if S1 knows that p but intentionally leads S2 to believe that not-p. Similarly, self-deception occurs when S1 knows that p but intentionally leads S1 to believe that not-p.

Consider the television series Breaking Bad. Walter White decides to leave money behind for his family by cooking and selling methamphetamine before cancer kills him. To do so, he must deceive his wife Skyler. He intentionally leads Skyler to believe that he is not doing anything illegal. But Walter also deceives himself. He knows that what he is doing is not only illegal but also immoral. However, since he really desires to be a self-made man and earn money for his growing family, he intentionally leads himself to believe that what he is doing is morally justified – perhaps even virtuous. This might happen via self-persuasion.⁴ He reminds himself of the hard facts: he has a small income as a chemistry teacher, a cancer diagnosis, and a second child on the way. He has been a failure if he dies and leaves behind a penniless widow with two children. What would be so terribly immoral about breaking the law for this one time? The junkies are going to use someone's drugs anyway. Why not his clean, quality meth? So, Walter goes about starting a drug business, but continues to feel uneasy about it. This telling of the story satisfies the key criteria of the intentionalist model:

1. (1) intentionality

2. (2) contradictory beliefs

3. (3) a psychic and/or behavioural conflict

First, Walter deceives himself knowingly and intentionally. Second, Walter seems to entertain two mutually contradictory beliefs: that his behaviour is immoral and that his behaviour is not immoral. Walter's story also includes a third feature that is central to both intentionalist and non-intentionalist models of self-deception: a psychic and behavioural conflict. When Walter faces the terrible consequences of his actions – such as innocent deaths – the look on his face tells the viewer that he realizes the immorality of his actions. Still, he wants to continue to cook meth, and is divided within. He must harden himself and turn his gaze away from the true nature of his deeds.

Intentionalist models face two puzzles or ‘paradoxes’. The static paradox refers to the seeming impossibility of holding contradictory beliefs. On the one hand, it is impossible to have a belief that p and not-p; it is logically possible to believe that p and to believe that not-p. On the other hand, it seems psychologically implausible that people could explicitly hold contradictory beliefs simultaneously.⁵ However, the intentionalist need not claim that self-deceivers entertain the beliefs that p and not-p in the same way and at the same time. Self-deception is a psychologically complex process that usually takes place during an extended period of time. Many intentionalists refer to some kind of psychological partitioning or temporal partitioning. Consider the following example of the latter:

[A]n official involved in some illegal behaviour might destroy any records of this behaviour and create evidence that would cover it up (diary entries, emails and the like), knowing that she will likely forget having done these things over the next few months. When her activities are investigated a year later, she has forgotten her tampering efforts and based upon her falsified evidence comes to believe falsely that she was not involved in the illegal activities of which she is accused. Here the self-deceiver need never simultaneously hold contradictory beliefs even though she intends to bring it about that she believes that p, which she regards as false at the outset of the process of deceiving herself and true at its completion. (DeWeese-Boyd 2021)

The second paradox, called the dynamic or the strategic paradox, points to the fact that it seems impossible intentionally to get oneself to believe what one thinks is not true. We are sensitive to cues of deception and react with hostility when we feel someone may be intentionally misleading us. To deceive another person successfully, the deceiver must follow a strategy and be aware of whether the victim is swallowing the lie. Contrariwise, the person being deceived must not be aware that they are being led astray. As DeWeese-Boyd (2021) puts it, ‘for one to carry out an intention to deceive oneself one must know what one is doing, to succeed one must be ignorant of this same fact’.

Temporal partitioning may alleviate the strategic paradox as well. As Wood notes, ‘everyone accepts that immediate, decisional control over belief-formation is impossible, but everyone also accepts that indirect, long-term control over belief-formation is straightforwardly possible’ (Wood 2013, 169). He also appeals to psychological partitioning:

While the self-deceiver remains in the state of self-deception – while he continues to deceive himself – he does not ever attend to the fact that he believes not-p, and he does not ever attend to the fact that he is intentionally deceiving himself into believing p. But he is still able – he still has the capacity – to notice these facts about himself. Presumably, if he were to notice them, he would thereby cease deceiving himself, by definition. While it is impossible for him to notice his beliefs and intentions and still continue deceiving himself, it is perfectly possible that he does not notice them at any particular point during his self-deception. (Wood 2013, 172)

While such an argument may not solve the strategic paradox, the purpose of this article is not to engage in the debate of whether the intentionalist model can be made to work.⁶ We turn now to see if the model can help make sense of the suppression thesis.

Non-belief as intentional suppression of the knowledge of God

Could non-belief sometimes be due to self-deception – or perhaps always? The first thing to note is that the concept of non-belief covers a range of attitudes regarding God, such as atheism, agnosticism, and ignorance of the concept of God (‘natural non-belief’; Marsh 2013). Self-deception, however, is not merely about resisting a true belief that p but also about acquiring a false belief that not-p. Therefore, atheism may seem like the only form of non-belief that can be analysed in terms of self-deception, as it involves the positive belief that God does not exist. However, I see no reason why self-deception may not sometimes result in, say, self-induced ignorance or mere disregard about the truth of p. For this reason, while the discussion below is relevant mainly for atheism, it has implications for the debate on non-belief in general.

Consider first the theological fit of the intentionalist model. The model seems compatible with Paul's claims in Romans 1 about what goes on in the mind of those who first realize God's existence but then turn away.⁷ Calvin's Institutes contains another helpful example. According to Calvin, those who say in their hearts ‘there is no God’ (Ps. 14:1)

stifle the light of nature, and intentionally stupefy themselves. We see many, after they have become hardened in a daring course of sin, madly banishing all remembrance of God, though spontaneously suggested to them from within, by natural sense . . . Thus although they are forced to acknowledge that there is some God, they, however, rob him of his glory by denying his power . . . [T]hough they struggle with their own convictions, and would fain not only banish God from their minds, but from heaven also, their stupefaction is never so complete as to secure them from being occasionally dragged before the divine tribunal. Still, as no fear restrains them from rushing violently in the face of God, so long as they are hurried on by that blind impulse, it cannot be denied that their prevailing state of mind in regard to him is brutish oblivion. (Calvin 1997, 47)

Calvin's version of the suppression thesis seems to satisfy all three features of the intentionalist model. Non-believers ‘intentionally stupefy themselves’ (intentionality). Whether their ‘prevailing state of mind’ regarding God is disbelief (‘there is no God’) or ‘brutish oblivion’, they are nonetheless ‘forced to acknowledge’ God (contradictory beliefs).⁸ They also ‘struggle with their own convictions’ (psychic-behavioural conflict). The process of suppression happens through temporal partitioning: by hardening their hearts, non-believers gradually banish ‘all remembrance of God’. Calvin also employs psychological partitioning. He divides between implicit and explicit beliefs by saying that awareness of God is nonetheless ‘spontaneously suggested to them from within’.

Is there empirical support for the claim that non-belief results from intentional self-deception? Cognitive science of religion might offer some evidence of contradictory beliefs among non-believers. Many scholars in this field suggest that even self-proclaimed atheists often have implicit beliefs of supernatural agents (Uhlmann et al. 2008; Bering 2010). The reasoning behind this claim is that supernatural agent concepts (gods, ghosts, and goblins) are intuitive and attention-grabbing and require merely minimal cultural scaffolding (Barrett and Visala 2018). Atheism, however, usually requires non-universal conditions such as significant cultural scaffolding (e.g. a WEIRD⁹ society that provides safety and comfort), effortful analytic reasoning (e.g. higher education), or a non-neurotypical (autistic) cognition (Norenzayan and Gervais 2013).¹⁰ Religious belief is thus the default cognitive setting of most human minds, and supernatural intuitions are hard to extinguish completely.¹¹

Indeed, there is some evidence of implicit theism among atheists. In one study, even non-religious individuals were less likely to judge religious concepts (e.g. God or Angel) as imaginary after writing about their own deaths than after writing about watching television (Jong et al. 2012). Reminders of one's own death are thought to bring to the surface the implicit beliefs that provide us with a sense of meaning and control. In another study, non-believers were shown pictures of natural phenomena (e.g. a mountain or a hurricane) (Järnefelt et al. 2015). When responding under time pressure, the participants often judged such phenomena to be intentionally created. However, the majority (84–62%) of the phenomena were not seen as purposeful even under speeded conditions. Many studies have also failed to find evidence of implicit supernatural beliefs among non-believers (Lindeman et al. 2016). Now, perhaps a defender of Calvin's suppression thesis could still use the positive findings to argue as follows: perhaps over time atheists succeed in banishing ‘all remembrance of God’ and their minds become ‘darkened’; thus, their theistic intuitions are buried too deep in the subconscious layers of the mind to be unearthed by psychological experiments.

A psychic/behavioural conflict could be suggested by experiments where non-believers have been found to behave as if God were more than an imaginary being. For example, one study found that atheists become emotionally aroused when daring God to do terrible things (Lindeman et al. 2014). Another study showed that activating thoughts about God influenced self-regulation such as temptation resistance independently of pre-existing religiosity (Laurin et al. 2012, 14). Several experiments have also investigated the effects of religious priming on prosocial behaviour. For example, Shariff and Norenzayan (2007) had the participants play an economic game and gave them opportunities to earn more money by acting selfishly. Activating thoughts about God made both believing and non-believing participants act more prosocially. However, a meta-analysis of over ninety studies showed that while religious priming has strong effects on a variety of prosociality measures, religious priming does not reliably affect non-religious participants (Shariff et al. 2016). Again, perhaps a defender of the suppression thesis could argue that while over time most non-believers may become insensitive to religious priming, the finding that reminders of God make some non-believers act prosocially serves as evidence of an inner conflict.

How about the intentionality requirement? Do non-believers intentionally deceive themselves into denying what they know to be true? Atheist philosopher Thomas Nagel's famous confession about his fear of religion may serve as anecdotal evidence:

In speaking of the fear of religion, I don't mean to refer to the entirely reasonable hostility toward certain established religions and religious institutions, in virtue of their objectionable moral doctrines, social policies, and political influence. Nor am I referring to the association of many religious beliefs with superstition and the acceptance of evident empirical falsehoods. I am talking about something much deeper – namely, the fear of religion itself. I speak from experience, being strongly subject to this fear myself: I want atheism to be true and am made uneasy by the fact that some of the most intelligent and well-informed people I know are religious believers. It isn't just that I don't believe in God and, naturally, hope that I'm right in my belief. It's that I hope there is no God! I don't want there to be a God; I don't want the universe to be like that. (Nagel 1997, 130–131)

Nagel desires that God not exist, and he fears that theism is true. Such a desire could easily lead one to intentionally deceive oneself about God's existence. Nagel's confession may also suggest an implicit realization of the truth of theism. But obviously we cannot assume that all or most non-believers experience such a desire or such a fear. Presumably, Nagel's confession is cited often because it is so exceptional. Atheist philosophers rarely seem to report such a fear in their personal narratives (see e.g. Antony 2007). Of course, the suppressionist could point out that this is to be expected if non-belief really is the result of self-deception.

Overall, my few examples of what seems like relevant evidence for the intentionalist suppression thesis (the claim that non-belief in general and atheism in particular results from intentional suppression of the knowledge of God) offer some support to the thesis. The idea that some individuals trick themselves into non-belief seems psychologically feasible. Of course, we are still far from any suggestion that theistic non-belief is always or even typically caused by such a process. Moreover, as we shall see, the intentionalist model itself is considered as psychologically implausible by many philosophers.

The non-intentionalist model of self-deception

Non-intentionalist or ‘deflationary’ accounts of self-deception reject both the intentionality requirement and the contradictory belief requirement. Alfred Mele is perhaps the leading defender of such accounts. He thinks the intentional model (or what he calls the agency view) is fraught with problems. In addition to appealing to ‘mental exotica’, the model does not apply to typical cases of self-deception.

Stock examples of self-deception, both in popular thought and in the literature, feature people who falsely believe – in the face of strong evidence to the contrary – that their spouses are not having affairs, or that their children are not using illicit drugs, or that they themselves are not seriously ill. Is it a plausible diagnosis of what happens in such cases that these people start by knowing or believing the truth, p, and intentionally cause themselves to believe that ~p? (Mele 2001, 9)

Mele thinks the answer is no. Self-deception typically does not involve a conscious intention – a ‘trying’ or an ‘effort’ – to get oneself to believe that not-p. Instead, self-deception is a process of motivated reasoning driven by a desire that not-p. While also in the intentionalist model self-deception starts with a desire that not-p, negative emotions have a more thoroughgoing role in the non-intentionalist model. The desire gains strength from emotions such as fear and anxiety that lead to biased interpretations of the evidence for p and not-p. Biases such as vividness of information, the availability heuristic, and confirmation bias interfere with our appraisal of the evidence (Mele 2001, 28–29). Mele lists four ways in which a desire that p may encourage motivated reasoning and biased treatment of the evidence:

1. 1. Negative misinterpretation (e.g. Don thinks the referee's comments do not justify the rejection of his article by a journal).

2. 2. Positive misinterpretation (e.g. Sid interprets Roz's friendly behaviour as indicating she loves him).

3. 3. Selective focusing/attending (e.g. Beth comes to believe she was her deceased father's favourite child by focusing on photos with only her and her father).

4. 4. Selective evidence-gathering (e.g. political campaign staffer Betty seeks primarily evidence against the claim that her boss is sexist). (Mele 2001, 26–27)

According to Mele's non-intentionalist model, then, S deceives himself if the body of data possessed by S at the time provides greater warrant for not-p than for p, and if S's biased treatment of the data is a non-deviant cause of his belief that p (Mele 2001, 50–51). Consider Breaking Bad again. Walter's desire to be a self-made man is driven by his fear of dying as a loser and leaving behind a poor widow with two children. He reflects on his situation by focusing selectively on points such that he has always been a hard-working, law-abiding citizen, that his friends and the society have treated him unjustly, and that manufacturing and selling drugs is an easy and a relatively harmless way to make money. Walter does not intentionally seek to acquire the false belief that doing so is immoral in his situation, but his fears and desires lead him to believe that his case is an exception to the rule. He persuades himself to believe a lie, but there is no reason to suppose he consciously intends to do so. Anxiety takes over, and no ‘trying’ or ‘effort’ is needed. Walter also holds no simultaneous contradictory beliefs.¹²

The usual complaint against non-intentionalist accounts is that they are unable to clearly distinguish self-deception ‘from other kinds of false believing, such as error, prejudice, wishful thinking, stubbornness, etc.’ (Wood 2013, 159). It is good to acknowledge such viewpoints of the self-deception debate as we are considering which model might prove more helpful in constructing an argument of non-belief as self-deception. But again, the purpose of this article is not to engage in that debate. Next, we shall examine whether the non-intentionalist approach might provide a theologically sound and empirically credible analysis of Pascal's version of the suppression thesis.

Non-belief as non-intentional suppression of the evidence for God

Mele's approach is less compatible with Paul's or Calvin's versions of the suppression thesis than the intentionalist account. Both Paul and Calvin assume that every non-theist either has or has once had knowledge of God. Some other notable theologians, however, seem to suggest that what non-believers reject or suppress is not their knowledge of God as such but rather their evidence for God. Consider Pascal's famous words regarding the ambiguity of general revelation.

[W]ishing to appear openly to those who seek him with all their heart and hidden from those who shun him with all their heart, [God] has qualified our knowledge by giving signs which can be seen by those who seek him but not by those who do not. There is enough light for those who desire only to see, and enough darkness for those of a contrary disposition.¹³

Wood (2013) argues that Pascal's theological anthropology has a good fit with the intentionalist model of self-deception. This is partly true regarding this quote as well since Pascal thinks non-belief results from the wilful rejection of God. However, this quote does not suggest that non-believers have (had) knowledge of God. Instead, Pascal seems to be saying that while God has provided sufficient evidence of his existence to all in the natural world, the evidence is correctly perceived as good evidence for God only by those who desire to know God. So, the quote has a better fit with the non-intentionalist model where a self-deceiver acquires a false belief (e.g. that there is no God) despite possessing good evidence against it (e.g. that there is a God).

Another contemporary thinker inspired by Pascal, C. Stephen Evans, posits two principles regarding God's self-revelation: a good God can be expected to provide evidence of his existence that is widely accessible but also easy to resist (Evans 2010, 12–17). In this scenario, God's revelation in nature and conscience does not automatically bring about universal knowledge of God. Instead, general revelation is evidentially and causally sufficient for all ‘biased’ towards finding God and it generates natural knowledge of God in them, but susceptible to mistreatment by anyone not interested in loving and worshipping God.¹⁴ The rejection of the evidence for God is, in this scenario, morally equivalent to rejecting God himself.

Cognitive science might support this idea. In their book on the cognitive underpinnings of theistic arguments, Helen de Cruz and Johan de Smedt argue that design arguments rely on two cognitive inference mechanisms: ‘the design stance, which leads us to treat complex and purposive structures as the product of design, and intuitive teleology, the propensity to discern purpose in nature’ (De Cruz and De Smedt 2015, 84). Regarding teleological thinking, Deborah Kelemen and her colleagues have conducted numerous studies showing that both children and adults in different countries and even professional scientists display implicit teleological and purpose-based reasoning (Kelemen 2004; Kelemen and Rosset 2009; Kelemen et al. 2013). As previously mentioned, many non-believers also judge natural phenomena to be intentionally created especially when responding in speeded conditions (Järnefelt et al. 2015). But while the tendency to perceive nature as designed seems to be a universal feature of human cognition, De Cruz and De Smedt argue that such mechanisms in themselves do not lead anyone to believe in a Designer. Nature does not force itself upon our minds as God's handiwork. Something more is needed to bridge the cognitive gap between the perception of design and the belief in a Designer. Pascal and Evans might suggest that bridge is the desire to know God. Importantly, when it comes to the question of whether God exists, Pascal thinks there are no disinterested observers. We either desire to know God or we desire not to know God.

Consider, then, four possible ways in which an anti-theistic desire might cause mistreatment of the evidence for God revealed in the natural world and in human conscience.

1. 1. Negative misinterpretation (e.g. Don thinks Kelemen's findings on intuitive teleology show that random natural selection has given humans a tendency for attributing design to non-designed objects, and thus the appearance of design in nature is bad evidence for God).

2. 2. Positive misinterpretation (e.g. Sid interprets the pleasure he gets from having an affair as evidence against the thought that his nagging conscience has to do with a God who judges infidelity).

3. 3. Selective focusing/attending (e.g. Beth focuses only on animal suffering in the natural world and bypasses animal pleasure).

4. 4. Selective evidence-gathering (e.g. Betty only reads books against theistic arguments from design and morality).

Let us take stock. The non-intentionalist model of self-deception is considered more psychologically feasible than the intentionalist model by many philosophers. The idea that non-belief results from non-intentional suppression (of the evidence for God) likewise seems more feasible than the idea that non-belief results from intentional suppression (of the knowledge of God). The evidence we have surveyed can support both claims. For example, Nagelian anti-theistic desire could similarly lead to intentional suppression of one's knowledge of God or to non-intentional suppression of one's evidence for God. Evidence of atheists' theistic intuitions and of a psychic/behavioural conflict among non-believers can likewise serve as evidence for both versions of the thesis.

However, scripture and Christian theological tradition (especially the Reformed tradition) seem to favour the intentionalist view of suppression, not least because it provides a better basis for moral culpability. The theological function of the suppression thesis is, after all, to show that non-belief is not due to ignorance or lack of evidence but is the result of wilful and culpable resistance. But as we will see, non-intentional models can also ground moral responsibility.

Are non-believers morally culpable for non-belief?

What could be morally wrong about self-deception? Some philosophers say self-deception exhibits the vice of epistemic cowardice (e.g. Levy 2004, 15). A self-deceiver fears the truth. Now, some cases of self-deception seem less morally problematic than others. Few people would blame a mother who non-intentionally deceives herself into thinking her son is not using drugs or a husband who convinces himself that his wife is not having an affair. Deceiving oneself about the existence and nature of God, however, is not only cowardly but, from a theological perspective at least, also sinful (Wood 2013, 122). The Reformed philosopher-theologian K. Scott Oliphint puts the charge as follows: ‘The cause of God's wrath towards us is our unrighteous suppression of the truth. In other words, God's wrath is revealed from heaven because, in our sinful wickedness and unrighteousness (in Adam), we hold down (in our souls) that which we know to be the case’ (Oliphint 2019, 156).¹⁵

Philosophers tend to agree that if intentionalism offers the correct diagnosis of self-deception, then self-deceivers are morally responsible for their action. Responsibility is more difficult to argue for under non-intentionalism. After all, ‘[i]f my anxiety, fear, or desire triggers a process that ineluctably leads me to hold the self-deceptive belief, I cannot be held responsible for holding that belief. How can I be held responsible for processes that operate without my knowledge and which are set in motion without my intention?’ (DeWeese-Boyd 2021). The rest of this section thus focuses on whether a defender of the suppression thesis can make a convincing case that non-believers may be culpable for suppression without a conscious intention to mistreat the evidence of God they have.

Debates on moral responsibility often focus on the issue of control. Under the non-intentionalist model, the question of moral responsibility for self-deception seems to become a question of whether one is able to control the biases that lead one to mistreat the evidence for p and not-p. On the one hand, we clearly hold some people responsible for not controlling the influence of their biases. We expect professionals like judges and doctors to keep watch for biases that might bring about an unjust verdict or a misdiagnosis. We rightly blame a judge who, because of his racist bias, tends to over- or underestimate the weight of the evidence depending on the skin colour of the accused. It is part of the job to watch out for those kinds of epistemic bugs. On the other hand, such introspective hunting for biases requires skills not everyone has, and even judges and doctors cannot always be held culpable for biased thinking in their personal lives.

Neil Levy thinks that we are mostly unaware of the biases that interfere with our assessment of the evidence, and thus we are typically not responsible for falling victim to bias-driven false beliefs. However, he also argues that culpability depends on how knowingly people mistreat the evidence they have.

[B]lame for false beliefs is appropriate when it can be traced back to an act of knowing epistemic negligence (the agent knowingly consulted an out-of-date source, or failed to attend a meeting at which he knew the topic would be discussed). Hence, self-deceivers are responsible for their state so long as it is a result of a knowing act or omission. No doubt, sometimes this condition is satisfied by self-deceivers. For instance, self-deception is sometimes traceable to an intentional act of refusing to investigate a subject any further, because contemplation of it provokes anxiety. (Levy 2004, 304)

All people sometimes face situations that make them look for evidence for or against something that may have important consequences for their lives. According to Levy, the situations in which we can expect people to undergo such a reflective process involve two conditions. ‘It must be the case that (1) the subject matter of the belief is important (whether morally or in some other manner), and (2) that we are in some doubt about it[s] truth (call these the importance and the doubt condition)’ (Levy 2004, 304–305).

The question of whether God exists clearly seems to satisfy both conditions. Most non-believers probably think the question is an important one, and most have some doubts about their non-belief. As evidence for the latter claim, a recent global study of atheism found that atheists generally have a low confidence in the truth of their position. In responding to the question ‘I feel confident that my beliefs about God's existence are the right ones’ on a scale from −2 (strongly disagree) to +2 (strongly agree), the average atheist responses from six countries were evenly divided between 0 (neither agree nor disagree) and 1 (somewhat agree) (Bullivant et al. 2019). Therefore, on Levy's criteria, most non-believers arguably have the responsibility to carefully consider the evidence they have for and against God's existence. This evidence could be arguments for and against God's existence, but also one's own religious experiences and those of others, or the lack of such experiences after a time of sincere God-seeking – whatever seemingly relevant evidence the person has access to. Therefore, it seems possible to build a decent case that non-believers are generally culpable for their non-belief even if self-deception is caused by processes that operate below awareness and which are set in motion without any conscious intention.

However, recall that, according to Pascal and Evans at least, considering the evidence is not enough to bring a person to believe. One must also have the desire to know God (and not have the desire not to know God). So, the question whether non-believers may be morally culpable or blameworthy for non-intentional suppression is not solely about whether people can be held morally responsible for controlling their biases. It is also about whether people can be held morally responsible for their desires. On the one hand, it seems that people cannot always be held responsible for having the desire that God not exist. Such a desire may be caused, say, by bad experiences of religious people. On the other hand, our morally relevant desires are dependent on our moral character, and some philosophers argue that we can be responsible for forming a character that is susceptible to self-deception (e.g. Johnston 1995). But this suggestion is problematic in the theological case of non-belief. Christian theology tells us that our sinful natures are not simply formed by our own free choices. We inherit our morally corrupt desires from our parents, and, ultimately, according to theological tradition, from Adam.¹⁶ Therefore, the question whether non-believers are culpable for not desiring God (and, as a result, for misinterpreting the evidence for God) falls back on the question of whether humans are responsible for the morally bad character with which they are born. Recent debates on these issues suggest that showing that we do have such responsibility is not an easy task (see e.g. Visala and Vainio 2024).

What might motivate suppression?

Whichever model of self-deception we opt for, there is an even more difficult question facing defenders of the suppression thesis: what exactly might motivate anyone to deceive oneself about God's existence? We may understand why Walter White would want cooking and selling meth to be morally justified in his case. But why might Nagel desire that atheism be true? Referring to Paul's list of sins (Rom. 1:29–31), preachers and theologians often claim that people want to live without a moral straitjacket and do as they please. Because of the noetic effects of sin, humans desire to live autonomous lives and be their own masters. Theologian Michael Bird puts the idea as follows: ‘[T]he presence of sin in the human heart means that this knowledge [of God] becomes traumatic because it implies God's authority over them and their accountability to God. People pretend not to hear this natural revelation, and they deactivate their inward mechanism for receiving the message’ (Bird 2013, 176).

One problem with such suggestions is that non-belief is unnecessary for avoiding the uncomfortable implications of divine authority. One can simply adopt a concept of God as non-personal or as having looser moral standards. As Alvin Plantinga has pointed out, the noetic effects of sin may lead one to view God as ‘an impersonal abstract object (“the ground of being”) rather than as a living person who judges me’ or as ‘an indulgent grandparent who smiles at the childish peccadilloes of her grandchildren’ rather than ‘as a holy God who hates sin’ (Plantinga 2000, 179). Such self-deception is arguably a cleverer strategy than the attempt to convince oneself of God's non-existence. Moreover, even though Christians believe that God is a moral judge who hates sin, the moral life of an average believer (at least in WEIRD societies) does not seem very different from the moral life of an average non-believer (e.g. Launonen 2022b). Since also many believers leave their moral straitjacket in the closet, you clearly do not need to reject God's existence to enjoy the pleasures of sin.

Although it may be true of human nature that we want to be morally autonomous, it is also true that religious belief is very beneficial for personal well-being and for experiencing life as meaningful (Koenig 2012). Moreover, many non-believers think that, since there is no God, our lives lack transcendent meaning (which would be valuable), and, since there is no afterlife, millions of people who suffer evil receive no justice or compensation (which also would be very valuable). In fact, there may be better reasons for suspecting that belief in God is a form of self-deception than for thinking that non-belief is. But this obviously depends on one's background beliefs.

Real-life examples of non-resistant non-belief present the most serious problem for the universal suppression thesis, that is, the idea that non-belief always results from self-deceptive suppression. As examples of non-resistant non-believers, Schellenberg mentions people like former believers (who ‘grieve what they have lost and seek to regain it’) and lifelong seekers (‘individuals who seek [God] but do not find’) (Schellenberg 2007, 228–233). A defender of the suppression thesis could always argue that perhaps such people are self-deceived also about the integrity of their grief or about their quest for God. But for most Christian theologians and philosophers who personally know many non-believers such claims are probably hard to swallow.

Implications, philosophical and theological

Where does our analysis leave us? The theological claim about suppression can indeed be understood as a claim about self-deception. However, the evidence we have found merely serves to support the partial thesis (that some non-belief is due to suppression), not the universal thesis (that all non-belief is). Such a result may seem uninteresting since both sides of the hiddenness debate tend to agree that some instances of non-belief may be due to resistance. But as Baker-Hytch has pointed out, showing that a ‘non-negligible portion of the non-belief we initially thought was non-resistant is actually due to sinful resistance’ could be relevant in responding to probabilistic hiddenness arguments claiming that ‘God would not permit as much non-resistant non-belief as we seem to find in our world’ (Baker-Hytch 2023, 3). While I have reviewed only some evidence in support of the suppression thesis, there could be resources for building a more convincing case that a significant portion of non-belief is indeed due to resistance/suppression.

The argument that suppression is widespread is also theologically important. The primary theological function of the suppression thesis is arguably to ground the claim about inexcusability. This way the thesis safeguards the traditional Christian claim that people need to believe, if not in Christ specifically, then at least in God to be saved. But if God allows non-resistant non-belief, then some non-believers – including some militant atheists, perhaps – are not culpable for not believing in and worshipping God, and a just God would have to save them despite their non-belief. Many conservative theologians are likely to find this consequence unappealing, as it downplays the role of personal belief and undermines the traditional idea that non-belief/unbelief is a sin. However, in some inclusivist accounts of salvation, belief is not necessary. One of the dogmatic documents of the Roman Catholic Church says the following:

Those also can attain to salvation who through no fault of their own do not know the Gospel of Christ or His Church, yet sincerely seek God and moved by grace strive by their deeds to do His will as it is known to them through the dictates of conscience. Nor does Divine Providence deny the helps necessary for salvation to those who, without blame on their part, have not yet arrived at an explicit knowledge of God . . . (Lumen Gentium 16)

So, some theological systems can clearly leave room for some amount of non-resistant non-belief. At the same time, it would seem problematic if such non-belief was more common than suppression. Both scripture and tradition emphasize the latter. For this reason, Christian scholars who hold scripture and tradition in high regard while acknowledging the reality of non-resistant non-belief may find my arguments in this article appealing.

Conclusion

The suppression thesis is the theological claim that theistic non-belief results from the wilful and culpable mistreatment of one's knowledge of God or one's evidence for God. I have argued that the thesis is effectively a claim about self-deception. Philosophers view self-deception minimally as a process where, because of some motivation or desire, a person acquires and maintains a false belief in the teeth of evidence to the contrary. Such a person may also display behaviour suggesting some awareness of the truth. While atheism is the only form of non-belief that involves the positive belief that there is no God, I have suggested (but not argued at length) that accounts of self-deception may be applicable also to other forms of non-belief.

Paul's and Calvin's versions of the thesis can be naturally interpreted through the intentionalist model: theistic non-believers know or have known God at least implicitly but have intentionally led themselves to reject or distort this belief. While the intentionalist model may be less feasible psychologically, it is better for making the case that self-deceivers are morally responsible for their false belief, including the atheistic belief that God does not exist. The non-intentional model coheres more with Pascal's famous words about light and darkness. On this model, suppression is viewed as mistreatment of one's evidence for God instead of one's knowledge of God. The non-intentional view attributes less agency to the self-deceiving person than the intentionalist view. It views self-deception as a more subconscious process and puts more emphasis on negative emotions and the desire to alleviate those emotions which leads to biased reasoning. However, even on this model, atheists could be held accountable for their false belief. People can to some extent control the influence of the biases that interfere with their assessment of the evidence. If prospective non-believers recognize the question of God as an important one, and if they are uncertain of what to think of it, they arguably have some responsibility to control the influence of their biases on their investigation of the subject. However, if careful consideration of the evidence is not sufficient to bring one to believe, but one must also desire to know God (as Pascal argues), the question of culpability falls back on the question of whether humans are responsible for the corrupted desires with which they are born.

We found some empirical evidence from cognitive science and anecdotal evidence from Nagel that can be used to support the suppression thesis. However, I also argued that it is not clear why many non-believers would be motivated (by a desire for moral autonomy, for example) to suppress their knowledge of God or the evidence they have for God. Moreover, the non-belief of people like lifelong seekers and former believers is very hard to explain away as the result of self-deceptive suppression. But while the universal suppression thesis seems hopeless, strengthening the partial thesis may help respond to claims that God would not allow as much of non-resistant non-belief as there seems to be in the world. As suppression is taught by scripture and tradition, arguing that suppression is rather common and non-resistant non-belief is rather rare also has theological importance.