How Can a Painting Make One Lose One’s Faith?

In 1867, the deeply religious Fyodor Dostoyevsky visited the Basel Art Museum and saw for the first time the original of Hans Holbein’s painting The Body of the Dead Christ in the Tomb. His wife later reported in her memoirs that the painting had such a powerful emotional effect on the writer that, in violation of the museum’s rules, he stepped on a chair to take a closer look. His face turned white, she recalled, and she had to drag him away from the painting fearing he would have an epileptic fit.  “Such a picture might make one lose one’s faith,” Dostoyevsky later told her.

The writer’s strong reaction to the image appears paradoxical: Christian Orthodox tradition, of which Dostoyevsky was a follower, does not deny Christ’s human nature or the fact of his death on the cross. After all, Christ’s sacrifice and his subsequent resurrection are the cornerstone of the Orthodox theology of salvation. Why would then an overly naturalistic depiction of the dead body of the Savior, marked by the early signs of decay, challenge an Orthodox believer’s faith in the resurrection of Christ? The answer, I suggest, lies in naturalistic imagery’s capacity to make intuitive biological knowledge which contradicts the doctrinal belief in the possibility of resurrection more salient.

The Body of the Dead Christ in the Tomb by Hans Holbein the Younger, 1520–22.

Anthropological and cognitive research on the acquisition of supernatural concepts shows that the latter do not displace basic biological and physical knowledge, but rather expand it, as it is only in specific contexts that religious concepts are deployed (Astuti 2011; Barrett 1999; Boyer 1994; Evans-Pritchard 1976). In contrast to the folk understandings of physics and biology that are easily evoked, certain religious concepts take time to master and thinking with them can sometimes require a continued effort on behalf of a believer who must temporarily suspend her intuitive knowledge of reality (Luhrmann 2012). As Justin Barrett (1999) demonstrates, people’s theological concepts are subject to cognitive constraints and often vulnerable to distortion by resurfacing intuitive knowledge that is projected on them. While Barrett focuses on the role the cognitive demandingness of a task (attention overload, a need to provide a quick reply etc.) can play in disrupting the process of making inferences with religious concepts, I suggest that certain visual cues and the sensory evidence they provide can similarly have such a disruptive effect. Dostoyevsky’s unnerving reaction to Holbein’s painting presents one example of such a process.

The apophatic ethos of the Orthodox tradition postulates the impossibility of cognizing or understanding the holy mysteries and encourages believers to instead “experience” the divine with their hearts through private prayers and participation in rituals (Naumescu 2018). Icons in the Orthodox context are conceived as sacral objects that are meant to serve as mediums for the transcendent world which through their materiality help believers to experience direct contact with the divine (Herzfeld 1990). Orthodox (Byzantine) iconography does not aim to depict historical events with any degree of detail and accuracy, but seeks to transcend time and space and convey the meaning of the event. In this context, historical details and naturalism in the depiction of Christ or saints are thought of as redundant elements that would bring the believers’ minds back to the earthly world instead of helping them elevate to the spiritual one.

Orthodox icon of Crucifixion by Dionisii, 1499 – 1500.

The Byzantine canon of depicting the scene of the crucifixion differs significantly from the practices of Western religious art. While Western iconography starting from the High Medieval period often portrays the suffering of Christ rather graphically, accentuating the gruesome details, Orthodox icons, which tend to focus more on the theme of triumph and resurrection, rather than suffering and death, usually present a highly abstract depiction of Christ’s body. His suffering is alluded to, but never foregrounded. This aesthetic difference, I argue, is not merely theological, but is determined by the properties of the images and the functional role they play in sustaining the two traditions.

In their analysis of the coexistence of conflicting biological and religious conceptions of death among rural residents of Madagascar, Rita Astuti and Paul Harris (2008) demonstrate how different contexts activate different types of knowledge about death. When presented with the narratives that focus on the corpse, their Madagascan informants were more likely to deploy their biological knowledge when answering questions about a deceased’s capacity to feel or think. In contrast, when primed with a ritualistic context of the ancestral practices associated with the afterlife, the respondents resorted to their religious conceptions and insisted on the continuity of mental processes after death. Building on this insight, I suggest that the different iconographic styles can function as visual primers that appeal to the different conceptions of death in believers.

Christ on the Cross (Isenheim Altarpiece) by Matthias Grünewald, 1512 – 1516.

The graphic portrayal of suffering and death in the context of Western Christianity engages the audience’s biological conception of death and its embodied knowledge, as it seeks to elicit a strong emotional response and invites the believers to ponder the greatness of Christ’s sacrifice and to participate in the Biblical drama through co-suffering with him (Carlson 2010). The visual mediation of the gruesome corporeal reality of death also aims to remind believers of their mortality, making them more aware of the imminent decay of the flesh, and instilling them with a fear of final judgement to deter them from engaging in sinful practices.

A more other-worldly directed Orthodox tradition, by contrast, encourages the believers to cultivate spirituality through detaching from material reality and directing their minds to the divine. Orthodox tradition does not require its practitioners to meditate on or try to vicariously experience the corporeal horrors of Christ’s passion. Even with the spread of the Western iconographic influences in Russia and several other Orthodox countries starting from the 17th century onwards, Orthodox icons of Crucifixion and Deposition of Christ still focused on conveying the beauty of Christ, rather than portraying the tragedy of his death. Orthodox tradition presents the concept of Christ’s death as a paradox (i.e. a divine eternal God cannot die) – a mystery that believers are asked to accept as such. The acceptance of the mystery of death and resurrection requires believers to suspend their biological knowledge and turn instead to a learnt religious conception. Byzantine iconography, which directs attention to the transcendent world and conveys the essence of Christ and the saints, rather than their physical form, provides crucial visual support for the exercise. It is in this context, that a naturalistic painting’s capacity to induce a momentary crisis of faith in an Orthodox person needs to be understood.

Dostoyevsky never wrote anything on Holbein’s painting in his personal notes. However, he did return to the morbid impression the image left on him in his novel The Idiot, in which one of his characters, Ippolit, makes the following comment on the painting:

It is strange to look on this dreadful picture of the mangled corpse of the Savior, and to put this question to oneself: ‘Supposing that the disciples, the future apostles, the women who had followed Him and stood by the cross, all of whom believed in and worshipped Him— supposing that they saw this tortured body, this face so mangled and bleeding and bruised (and they MUST have so seen it)—how could they have gazed upon the dreadful sight and yet have believed that He would rise again?’ The thought steps in, whether one likes it or no, that death is so terrible and so powerful, that even He … was unable to triumph over it at the last (p. 750).

The question posed by Ippolit (and which may have haunted Dostoyevsky himself), nicely illustrates the tension between the intuitive biological knowledge inevitably triggered by the sight of a naturalistically depicted corpse and doctrinal beliefs in the miracle of resurrection. This episode, even if fictional, speaks to the importance of visual mediums in the transmission of religious concepts. It also points to the potential challenge naturalistic imagery poses to Orthodox modes of belief.  Deeply dependent on the mediating role of icons for enabling a direct contact with the holy figures, the transmission of Orthodox spirituality also relies on a specific mode of representing the mystical world.

The differences between iconographic traditions certainly should not be reified, as historically there has been a significant amount of variation in style and canon within them (Belting 1994). Neither should the challenge naturalistic depiction of death might pose to one’s faith be overstated. Christianity presents a complex system of beliefs and propositions and while naturalistic imagery can challenge some of them, it can also make others more salient. In fact, it can have both a disruptive and an affirmative effect on religious conviction, depending on the type of spirituality one is trying to cultivate. Although a naturalistic depiction of Christ’s dead body might not to be a suitable visual medium for transmitting beliefs about resurrection, it can play a reaffirming role in the mode of religiosity that focuses on Christ’s suffering and death as evidence of God’s boundless love for humanity. Whether exposure to different types of imagery can indeed cause substantial changes in believers’ religious sensibilities and their ways of understanding divine mysteries and relating to them, is an interesting empirical question that has yet to be pursued.


Astuti, Rita and Harris, Paul. 2008. “Understanding Mortality and the Life of the Ancestors in Rural Madagascar.” Cognitive Science 32: 713-740.

Astuti, Rita. 2011. “Death, Ancestors and The Living Dead: Learning Without Teaching in Madagascar.” Pp. 1-18 in Children’s Understanding of Death: From Biological to

Religious Conceptions edited by V. Talwar, P. Harris and M. Schleifer. Cambridge: Cambridge University Press.

Belting, Hans. 1994. Likeness and Presence: A history of Image Before the Era of Art. Chicago and London: University of Chicago Press.

Boyer, Pascal. 1994. The Naturalness of Religious Ideas: A Cognitive Theory of Religion. Berkley: University of California Press.

Carlson, Marla. 2010. Performing Bodies in Pain. Medieval and Post-Modern Martyrs, Mystics, and Artists. New York: Palgrave Macmillan.

Dostoyevky, Fyodor. 1968. The Idiot, translated by Eva Martin (2008).  Stilwell, KS: Publishing.

Dostoyevskaia, A.G. 1971, Vospominaniia, edited by V.V. Grigorenko, S. A. Makashina, S. I, Mashinskii and V. N. Orlov. Moscow: Khudozhrstvennaia literatura.

Evans-Pritchard, Evan.1976. Witchcraft, Oracles and Magic Among Azande. Oxford: Clarendon Press.

Herzfeld, Michael. 1990. “Icons and Identity” Religious Orthodoxy and Social Practice in Rural   Crete.” Anthropological Quarterly 63 (3): 109-121.

Luhrmann, Tanya. 2012. When God Talks Back: Understanding the American Evangelical Relationship with God. New York: Knopf Doubleday Publishing Group

Naumescu, Vlad. 2018. “Becoming Orthodox. The Mystery and Mastery of a Christian Tradition.” Pp. 29-53 in Praying with the Senses. Contemporary Orthodox Christian Spirituality in Practice edited by S. Luehrmann. Bloomington, Indiana: Indiana University Press.


  • comment-avatar
    Dan Sperber 5 March 2018 (16:45)

    A possible test
    This excellent anthropological post suggests that cognitive considerations might be relevant to understanding the fit (1) between Eastern Orthodox Christian negative (or “apophatic”) theology (with its stress on what God is not and on mystical experience as the only way to gain some theological knowledge) and Orthodox iconography which “seeks to transcend time and space and convey the meaning of the event,” and (2) between Western Christian positive (or “cataphatic”) theology with its more descriptive approach of the divine and Western more realistic iconography of religious themes.
    Here is a way in which this suggestion (inspired by Rita Astuti’s work that Victoria cites) might be made experimentally testable. With a morphing program, one might produce several pictures of the crucifixion on a continuum between Orthodox depictions highlighting the splendour of Jesus and Western depictions highlighting his bodily pain. First, would people’s preference between adjacent pictures in this series reflect their theological commitments? Second, could theological statement or religious narratives prime different preferences? Conversely, could different pictures prime agreement with different statements?

  • comment-avatar
    Pascal Boyer 7 March 2018 (15:42)

    A clash of intuitions
    Thanks, Victoria, for this beautiful illustration of how we can use psychology to re-consider important anthropological or historical questions. I cannot comment on the difference between Christian traditions, but would add a footnote to your comments on the cognitive processes engaged.
    The treatment of deceased bodies is a great example of a difficult and important “cognition and culture” problem. We know that in most human societies, people will not leave the dead alone, so to speak. They have to do something about it. And that has been going on for a long time (McCorkle, 2010). This is a puzzle because a) from an evolutionary standpoint, handling corpses is not directly adaptive and may be dangerous, b) cultural norms and representations do not tell us why people do it, only how people best explain the fact that they are motivated to do it.
    Empirical studies like Astuti and Harris’s investigation of explicit representations of death in Madagascar show us that people can readily switch between the expression of biological intuitions (the dead person is inanimate) and meta-represented beliefs (about the dead as potential ancestors), even though the two sets of representations are not entirely compatible.
    But the presence of dead bodies in all likelihood triggers a more direct conflict between two sets of intuitions, biological (about animacy cues, e.g., the likelihood that the body will move of its own accord) and social-psychological (about the deceased as a particular person, their particular ties to each individual around, their past interaction with different people, etc.). It seems that our biological systems easily re-categorize the dead as dead. But our intuitive psychology and sociology systems seem to carry on, so that people carry on having feelings about the deceased, e.g., being proud of some behavior that the deceased would have approved of (Boyer, 2001, pp. 210-222).
    Claire White has conducted several fascinating studies of this phenomenon. She and Dan Fessler showed that, across cultures, extensive physical contact with the dead, e.g., preparations of the corpse and other ritual contact, provides extensive inanimacy cues, reinforcing the biological intuitions triggered by the corpse (White, Marin, & Fessler, 2017). Conversely, seeing pictures or photographs of the dead tends to strengthen the psychological and social intuitions, leading to more frequent illusory encounters – people misidentify a stranger as the deceased person, or they heard their voice, etc. (White & Fessler, 2013).
    As Victoria comments, some versions of Christian tradition make this conflict particularly acute (even accounting for the fact that Dostoevsky felt all tensions and conflicts as acute and painful…) Many Christians represent Christ as a person with whom they have constant social interaction. But the official tradition, especially in Western churches, also represents him as suffering and dying. The emphasis on physicality and biological death may well be at odds with people’s social intuitions. To understand how that works, we need more anthropologists!

    Boyer, P. (2001). Religion Explained. Evolutionary Origins of Religious Thought. New York: Basic Books.
    McCorkle, W. W. (2010). Ritualizing the disposal of the deceased : from corpse to concept. New York: Peter Lang.
    White, C., & Fessler, D. M. T. (2013). Evolutionizing grief: Viewing photographs of the deceased predicts the misattribution of ambiguous stimuli by the bereaved. Evolutionary Psychology, 11(5), 147470491301100513.
    White, C., Marin, M., & Fessler, D. M. T. (2017). Not just dead meat: An evolutionary account of corpse treatment in mortuary rituals. Journal of Cognition and Culture, 17(1-2), 146-168.

  • comment-avatar
    Radu Umbres 9 March 2018 (23:30)

    Institutions and cognitive mechanisms
    Victoria, thanks for this very thought provoking text, which I really think should be explored more – it really asks some fascinating questions about the relationship between theology and imagistic representations. Your text really fleshes out the uncanny sensation I’ve always had that imagistic Orthodox Jesus and Catholic Jesus were different actors of the same play.

    As Dan and Pascal, you propose an cognitive mechanism between images and narratives / abstract representations. Let us suppose for a moment that such a causal mechanism exists (in one or both directions, as Dan suggested we can test). But how does psychology enter into the story of religious imagery? If we only think about the receptor, we forget about the producer of image as communication.

    I think we ought to pay attention to the way religious imagery was regulated in the two Christian traditions. In the East, content has clear constraints from Church directives, Dionysius of Fourna’s Hermeneia being one of the later strict codifications (it was the canon for Romanian Orthodox imagery). In the West, for all I know, content was controlled less, with many degrees of freedom for the painter as long as respecting Biblical accuracy.

    So could there be a story of different forms of cultural transmission? One of them institutionalised from above, conveys meanings controlled by church authority. The other one, more spontaneous and diverse, comes out of ground-up, multi-actor driven actions. Was Christ Pantocrator the preffered message impossed by Eastern Caesaropapism, while humanly-suffering Christ was the cultural attractor of independent creators and/or tradition schools in decentralised medieval Europe?

    Sorry for my impressionistic contribution, but thanks again!

  • comment-avatar
    Victoria Fomina 12 March 2018 (19:47)

    further research directions
    Thank you Dan, Pascal, and Radu for your thought-provoking comments that open a lot of interesting avenues for further research. Dan proposes a series of experimental tests that would shed light on the relationship between the images and narratives / beliefs in the Christian context. When it comes to regular believers (as opposed to the clergy), I doubt that their choice between adjacent images would be guided by their theological commitments. Rather, I would expect it to reflect their aesthetic preferences, which, one can argue, are shaped by theology, as it often determines the type of religious imagery believers are exposed to. In this context it would be interesting to see if the participants would easily depart from their aesthetic preferences when primed with different narratives in the second experiment.

    The third question Dan suggests is even more interesting, but also trickier to test: “could different pictures prime an agreement with different statements?” If the believers are presented with the statements corresponding to their doctrinal beliefs, I would expect them to embrace these statements regardless of the content of the images they are primed with. The difficulty here lies in capturing an oscillation in the strength of commitment to a specific belief that can be prompted by certain images. Since the acceptance of religious dogma is usually understood by believers as a matter of either or, rather than degree, trying to ask them to rate the strength of their agreement with a statement like “Christ has risen” would hardly yield any meaningful results. The challenge is then to work out an experimental framework that would somehow be able to register the various manifestations of clashing intuitions in the believers and the effect it has on their beliefs without explicitly asking them to express their attitude towards a dogmatic statement.

    I am grateful to Pascal for elaborating on the cognitive mechanisms underlying the conflicting understandings of death. I also welcome his suggestion for further anthropological enquiry into how people negotiate the conflict between their biological and social intuitions about the dead and how they switch between them. One place to start, perhaps, would be to identify the moments of rupture (akin to the episode with Dostoyevsky in Basel), in which the conflict between the clashing intuitions becomes apparent, and to explore ethnographically when such conflicts emerge and how they are being resolved.

    One good ethnographic example that comes to mind is the anti-religious campaign launched by the Bolsheviks during the Civil War that was explored at length by Robert Green in his book “Bodies Like Bright Stars: Saints and Relics in Orthodox Russia” (2010). To crush the peasants’ beliefs in the powers of the saints the Bolsheviks carried out a series of exhumations of saints’ bodies, presenting to the dismayed public the decaying remains in order to dispel the popular myth of the incorruptibility of saints’ bodies. While many of the peasants were shocked to find the dry bones in the saints’ coffins, Green reports, the campaign, nonetheless, had little effect on the believers’ practices of saints’ veneration. This episode in conjunction with other etnhoraphic evidence raises an interesting question as to why when the biological intuitions come to clash with the social ones the latter tend to win in the long run?

    Radu raises an important question of the role of image producers in the development of iconography and points out the differences in the modes of cultural transmission between the institutionalized Eastern and more decentralized Western Christian traditions. Based on the limited encounters I had with the theologians and icon-painters working in the Byzantine technique in Cyprus and in Russia, my intuition is to say that the actors regulating iconographic canon as well as the icon-painters are keenly aware of and intentionally exploit the capacity of the images to appeal to different intuitive systems in the audience. The icons are always created with an intention to produce a particular effect on believers and it is the theological discourses that determine what this desired effect is to be. I would speculate that whether the iconographic canon is regulated by the central authority of the Church or not, icon painters are still inevitably influenced by and reflect in their work the dominant theological discourses of their time. My interpretation is that the affinity between Orthodox mysticism and a more abstract depiction of human body and between passion-centered Catholicism and naturalistic imagery is the result of the experts’ (be it authoritative churchly figures or individual icon-painters) conscious efforts to produce images that would be most conducive towards turning the believers’ minds and hearts in the “right” direction.

    This being said, there are, of course, other significant factors at play in determining the cultural success of a particular iconographic style, like politics, to name just one. For instance, the powerful resurgence of the Byzantine canon in the Greek Orthodox world that started in the second half of the 20th century was legitimized through theological discourses of returning to the correct practice. At the same time, this revival was in many ways triggered by the political process of reinventing Greek national identity that was accompanied by a quest for rediscovering the “authentic” Byzantine tradition, construed as a repository of national wisdom, and an impulse to “purify” Orthodox practices from “corrupting” Western influences.