Search and Hit Enter

The ‘Magritte effect’ in the study of religion, Part I and II

In this theoretically rich piece, Jean-Pierre Warnier discusses the entanglement of ‘things’ and their representations.In most religious traditions, this topic plays an important historical role in determining how devotees respond to imagery and materiality, especially as these media convey or embody their most important religious concepts. Cycles of iconophilia and iconoclasm relating to this issue form a central thread in the Abrahamic faiths, for instance. Warnier insists that scholars of religion need to be more circumspect regarding the ‘cognitive gap’ that exists between the praxeological ‘things’ of a religious tradition and the representations of those things.


MLA citation format:
Warnier, Jean-Pierre “The ‘Magritte effect’ in the study of religion, Part I and II”
Web blog post. Jugaad: A Material Religions Project. 12 November 2014. Web. [date of access]

Part I: The ‘Magritte effect’ (or the gap between practice and representation)

What is the relevance of the representations of the ‘religious’ to the practice of religion? Why should we as anthropologists, historians, theologians etc., care about the difference between practice and representation? What are its epistemological and methodological implications?


The surrealist painter Magritte once produced a series of paintings representing a smoking pipe. Under each representation of the pipe, he wrote: “this is not a pipe”. This is true and untrue at the same time. Looking at a picture or statue of Buddha, the devotee will say: “This is Buddha”. He will behave accordingly and bow to the image/artefact, whereas it is, to an onlooker, a representation of Buddha. Following Magritte’s logic, under each statue and each image, we should read: “this is not Buddha”. But some will say that, to a certain extent, it is Buddha. Conversely, Muslims are against representations of Allah. They argue that actualizing the presence of the deity by an image or a statue would be misleading for the devotee and this is why there should be no object or thing. The devotee might easily believe that “this is Allah” whereas “this is not Allah”. It is just a representation. In the case of both the Buddhists and Muslims, the dynamic of a ‘thing’ and its ‘representation’ are integral to the conceptualisation of belief.
The Treachery of Images (This is Not a Pipe) (La trahison des images [Ceci n’est pas une pipe]) René Magritte (Belgium, 1898-1967), Belgium, 1929, Oil on canvas. The work is now owned by and exhibited at LACMA.
I have previously used the term “Magritte effect” (Warnier 2007: 6-13) to refer to the confusion between a thing and its representation, between the smoking pipe and the painting, between the territory and the map. It is important that we do not collapse the thing and its representation. They have different affordances. You cannot stuff tobacco in a painting and light it to inhale the smoke. What you can do is to hang it on a wall. If you want to smoke tobacco, you need a real smoking pipe, not its representation.
The collapse between thing and representation becomes more problematic, for example as anthropologists, when we describe and analyse other cultures. Consider the list of questions that you would raise if you were dealing with a religious practice as a praxeology and not a representation. (The practice need not necessarily be in a church, temple or mosque; routine activities such as eating, cooking, hunting, etc. are considered spiritual pathways in many cultures. Indeed, a recent conference that I organised dealt with contexts ranging from shamanism to waste.) What light would this shed on the worldview of a group or people? Would one be dealing with abstract concepts of what the world is or the manifold processes by which worlds are created (or destroyed)?


To consider an extreme example, if one studies warfare and deals with its practice rather than its representation, the questions might be: What about the way combatants move. How do they fight and yell? How do you kill them, with what weapons? In a previous publication (Warnier 2011), I addressed this question as regards the fighter’s bodily techniques and subjectivity during WWI. What about blood on your hands or on your clothes? How do you analyse and theorise the real bodily and material cultures instead of their representations?


One might write about war and yet not consider the ways in which one actually becomes a warrior. This is not to say that such studies are irrelevant. Far from it! In the human species, speech, representations, symbols, notions, worldviews are most important. They deserve a study for their own sake. But when talking of bodily and material cultures, I propose that we need to consider the pipe together with the picture of the pipe; the bodily and material cultures of religious practice together with their representations.


As an anthropologist, why should I take the Magritte effect into account? It is a problem in epistemology and methodology because there is a cognitive gap between the pipe and its representation and we cannot assume that there is any clear degree of fit between the two. Up to a point, the pipe and its image fit more or less together when you take them as signs in a coded system of connotation or communication. Both connote the action of smoking tobacco. One could say that their sign value is more or less the same (although not quite). But from the praxeological point of view and for their affordance value in a system of agency, they do not fit together. So we are talking about two kinds of values here – sign or representational value and praxeological (bodily and material) value. If I take the difference between these two kinds of values into account, what are the methods/theories I should implement to grasp everything that pertains to the material thing, the way it is perceived through the various sensory canals, the way it triggers emotions or moods, etc.? And what are the methods/theories I should use to grasp everything that pertains to the representation? The representation may use the medium of speech, images, sounds etc. In each case, I will have to implement different methods.


The second problem concerns the wo/man in the street, not as an anthropologist but as a devotee (or a tobacco smoker). For example, I recognise that an image, picture or a film is not the ‘real’ thing. I can see the film of the Urkupina Pilgrimage. I know it is a representation. I am sitting here in Europe. I cannot travel to Bolivia. I cannot go and participate in the pilgrimage. Yet, the film has an effect on me. If I am a Catholic devotee, I may be moved and reinforced in my devotions to the Virgin Mary by watching the film. Same thing with any kind of medium: sound recording, still images, digital images on the web, etc. There will be a gap between the real thing and the representation provided by the medium. Medias give a different kind of experience than the ‘real’ thing. Identifying what is ‘real’ in this case is not an abstract exercise in philosophy. The wo/man in the street makes a clear difference between the two- the real and the representation. If the devotee can go on the pilgrimage, he/she will go, because it is the real thing. If they cannot, they will be content with images and small statues, or a bottle of holy water. But in their view, it is not quite the same thing. It is a substitute. There may be a notion of efficacy, of success or failure in this difference and in the different media. I may, for example, be moved by an image and not by the sounds. If I hear a recording of a Catholic ceremony in Notre Dame, with the organs, the choirs, etc. on my player, I may be immensely moved by the sounds, whereas a photograph of the bishop in the choir surrounded by all the clergy etc. may fail to affect me.


What I want to tackle, when dealing with the Magritte effect, is the failure to distinguish between these two kinds of experiences—of the real and the substitute, and to point out the significance of the differences for both scholars and subjects, respectively. Anthropology and the social sciences are generally speaking imbued with speech, notions, and forms of verbalized knowledge. They are logocentric sciences. For many anthropologists, it is enough to have access to the representations, to the verbal descriptions, and to the associated images. It is enough because s/he is not aware that there is a cognitive gap between the thing and its representation, and that we are coping with two different kinds of data: the equivalent of the smoking pipe itself (warfare) and the painting of the smoking pipe (the verbal comments by the warrior). From a praxeological point of view, those things are incommensurate and quite different. In anthropological analysis too they have to be kept apart. They do not constitute the same kind of data, the tools for their analysis are not the same and they give access to different conclusions.

Part II: Marching Towards the Religious ‘Real’

So far I have described the academic and subjective confusion between the ‘thing’ (or the bodily and material cultures) on the one hand, and their representations on the other hand. In order to illustrate this confusion or ‘Magritte effect’ as a cognitive gap with socio-cultural effects, I will use the Foucauldian concept of subjectivation. It refers to the way subjects produce themselves and are produced, shaped and subjected to something else that can be called a sovereignty. It rests on the implementation of a number of technologies of the subject that are as many technologies of power. They produce a compliance of sorts that is also found in the religious subject. Thus, I would argue that the very process of religious subjectivation is based on a cognitive gap and the collapse and contradiction between the thing and its representations.
The sacred king smears camwood on his wives and children right after a sacrifice to his ancestors (Awing, Cameroon, 1973). Image: J. P. Warnier.


An example of such a cognitive gap can be found in my research on the kingdom of Mankon (Warnier 2007: 287-292). If you listen to the doxa expressed by repetitive verbalizations by the king and his subjects, you are made to believe that all the subjects of the kingdom share equally in the benefits provided by the king and the monarchy. They keep claiming that they “speak with one mouth”. They pretend to live in agreement with each other and with the king, to share in the unity of the kingdom. They claim that they owe their life and prosperity to the king as the trustee and channel of ancestral life substances he bestows on them. It is a Rousseauist, egalitarian discourse. Yet the bodily practices and the monopoly exercised by the king over ancestral material substances and containers built up a strong inegalitarian hierarchy. In the past, the king monopolized 5% of the female population of the kingdom. 50% of the male subjects remained bachelors for life and more or less barred from genital sex (Warnier 2007: 233-266). The subjects kept bringing to the palace not only livestock and foodstuffs that fetched a given price on the marketplace, but also servants and wives for the king. The king paid them back by spraying raffia wine from his mouth on them or smearing red pigments of camwood on their skin. These substances had a huge symbolic value but no exchange value on the marketplace. They built up an extremely strong hierarchy, the weight of which was shouldered by the women, but even more by the unmarried, low-ranking male cadets. The pressure on the different levels of the hierarchy translated into fierce competition over successions to titled offices, over marriage, and also by witchcraft accusations followed by potentially lethal ordeals. How is one to understand and analyse this difference between the discursive claim that the people “speak with one mouth” and the fierce and often violent competition for resources and privileges?


In order to analyze this paradox, I will follow three lines of argument.
  1. The “Bateson argument”: Subjectivation is achieved through the conjunction of two different channels or media that are used at the same time between interrelated subjects: verbalized, propositional knowledge (words, speech, discourse, text, anything that can be expressed in so many words, including systems of signs that can be decoded and translated into words such as, for example, polluting substances as ‘matter out of place’ according to Mary Douglas); and, on the other hand, procedural knowledge that is incorporated in bodily conducts and material culture, such as riding a bicycle or non-verbal bodily communication between two subjects (eg. mother-child interaction as photographed by Bateson and Mead 1942, or interactions in the Naven ritual as described by Bateson 1972: 178-183). There is no automatic agreement between what is conveyed or achieved by the two media in any given situation, or between the same persons or groups. This is exactly what Bateson postulated when he expressed the ‘non-agreement’ hypothesis. He stressed the cognitive gaps and contradictions between the two media. For example, the rejecting attitude of Balinese mothers while handling their children and talking affectionately to them. Hence, the Palo Alto school and the theory of schismogenesis. In ordinary life, says Bateson, such a cognitive gap may trigger a process of schismogenesis in the person (1936: 195-196), family (207-208), groups (183-185), politics (186), etc.
  2. The Foucauldian argument: Subjectivation is a double process of a) production of given subjectivities, lifestyles, repertoires of practice, and b) a process of subjection to a sovereignty. Or else, said in a different way, governmentalities are systems of action on other people’s action. One cannot assume that acting on other people is devoid of tensions, contradictions and conflicts. Yet, governmentalities act on the subject precisely at the point where the subject governs itself, and takes him/herself as the object of his/her own actions. Subjection/subjectivation is geared to an ambiguous exercise of power; power as a means of producing subjects and shaping them, and power as a means to control, exercise domination, and even punish. Power addresses the body of the subject through material “dispositifs”. At the same time, power proposes a discourse that may cover up the material contraptions used to subject people. Subjectivation is not a smooth, peaceful and entirely positive process.
  3. The neuro-cognitive sciences argument: The body and the cognitive unconscious. Whereas any subject has a reflexive awareness of his speech and what it means, this is not the case with bodily conducts. Bodily conducts and the sensorium are buried in the cognitive unconscious (see Berthoz 2000 and Buser 2005). We are not conscious of the billions of nervous impulses needed to peddle efficiently while keeping one’s balance on a bicycle. We are not conscious of all the stimuli that we receive through the seven sensory channels (and many more). If we were conscious of them, we would be overwhelmed by the flood of sensations and incapable of performing any coordinated bodily conduct. We would be incapable of keeping control over such a flood. In ordinary life, our nervous/cognitive system selects and brings to our awareness only those stimuli that are relevant to our present situation, that is, a minute fraction of what our seven senses actually catch. It inhibits the perception of irrelevant stimuli. Consequently, there is a fundamental imbalance between, say, verbal self-awareness (and intentionality) in speech on the one hand, and the bodily/sensory unconscious on the other hand. Experiments with top athletes show that the description they give of their movements when speaking are mistaken up to 50%. This peculiarity of our cognitive system provides the conditions for cognitive gaps, contradictions, conflicts between conscious verbalized knowledge, and mostly unconscious embodied procedural knowledge and the sensorium over which we have very little conscious control.

Coming back to my example of the Mankon kingdom, the contradiction between verbalized discourses on the one hand, and bodily and material technologies of power on the other hand, was essential to the technologies of royal power and to the process of monarchic subjectivation. The cognitive gap and the contradiction ‘marched’ the subjects (and the king himself). You can challenge a discourse. You cannot challenge the fact that the king’s body is imbued by ancestral substances and, consequently, can spray his saliva on you to impart some ancestral life essence. You just accept it as a fact of life, because it is real. This kind of cognitive gap is precisely what creates compliance to the religious ‘real’ and makes the faithful march forward. It is not something accidental or an indication that something has failed to function. It belongs with the very mechanism of subjectivation. It means that the mental image of the king is something different from the king, just as the painting of the pipe “is not a pipe” as Magritte aptly demonstrated.To conclude, unless we as anthropologists, make a clear distinction between bodily and material cultures of religion on the one hand and their representations on the other hand, between the pipe and the representation of the pipe, and between their diverse affordances, we miss something that is essential to religious compliance. Unless we contrive to have specific conceptual and theoretical tools to study bodily and material cultures (gestures, motions, perceptions, material characteristics of objects, the sensorium etc.) on the one hand, and their representations on the other hand, we miss something essential. Unless we go one step further and analyse the articulation between the experience of the pipe to its representation for given subjects (as did Bateson), we miss something essential. We fall victims to the Magritte effect.


  • Bateson, Gregory and Margaret Mead (1942). Balinese Character: A PhotographicAnalysis. New York, NY; New York Academy of Sciences.Bateson, Gregory (1972). Steps to an Ecology of Mind: Collected Essays in Anthropology,Psychiatry, Evolution, and Epistemology. London; Chandler Publishing Company.
  • _____(1936). Naven: A Survey of the Problems Suggested by a Composite Picture of the Culture of a New Guinea Tribe Drawn from Three Points of View. Stanford, CA; Stanford University Press.
  • Berthoz, Alain (2000). The Brain’s Sense of Movement. Cambridge, MA; Harvard University Press.
  • Buser, Pierre (2005). L’Inconscient aux mille visages. Paris; Odile Jacob.
  • Warnier, Jean-Pierre (2007). The Pot-king: The Body and Technologies of Power. Leiden; Brill.
  • _____(2011). “Bodily/material culture and the fighter’s subjectivity”, Journal of Material Culture, 16(4): 359-375.



1 Comment

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.

%d bloggers like this: