I. What exactly is consumption?
A chimpanzee eats a banana. A woman gathers mushrooms in a forest and cooks them for dinner. A guest at a party carries an expensive handbag and makes sure it is seen. A man listens to his favourite song for the hundredth time. A patient takes a prescribed antidepressant. In contemporary academic language, every one of these activities can be classified as consumption. What, precisely, do they have in common?
The chimpanzee is obtaining and metabolizing nourishment. The mushroom gatherer is moving through a forest, recognizing what is edible, carrying it home and transforming it over a fire. The guest with the handbag may be displaying wealth or seeking recognition. The listener is engaged with music, with memory and with a mood. The patient is attempting to change bodily states, and perhaps much more of how her life is going, through a pharmaceutical intervention. Material things are involved in all five scenes, and that is almost the only respect in which the category of consumption compares them. The participants are not performing the same activity, and they are not pursuing the same stakes.
The anthropology of consumption usually begins after the decisive classificatory step has been taken. Once something has been designated consumption, the task becomes explaining its meanings, its social organization, its cultural variations or its historical transformations. Far less attention goes to whether that comparison is relevant to what the researcher wants to understand. We can almost always describe an activity as consumption, and what needs to be shown is why doing so should help us understand what matters about it.
Consumption is an unusually concentrated case of a wider problem in social theory, because several errors that are usually encountered separately converge in it. It carries the transactional assumptions of economics, the symbolic priorities of interpretive anthropology, the explanatory hierarchies of class theory, and a tendency towards indefinitely expandable categories. This makes it a particularly useful case for a meta-analysis, one that examines a stabilized classification as a classification rather than working inside it.
Six problems run through the literature. Consumption is implicitly modelled on the modern relationship between producers, sellers and consumers, although much of what is called consumption involves no transaction at all. The occurrence of purchase, use, ingestion or depletion is taken to establish that consumption is central to understanding an activity. The category is defined so broadly that almost any engagement with objects, experiences, relationships or symbols can be described as consumption. Symbolic meaning, identity and distinction are given explanatory priority over the embodied, environmental, material and relational processes through which living is sustained. Distinct phenomena, ostentation, status competition, sensory pleasure, nourishment, care and maintenance among them, are treated as varieties of consumption rather than investigated in their own right. And once people are institutionally constituted as consumers, commercial and administrative arrangements reorganize their living around consumer categories, making the original classification look increasingly self-evident.
Rather than offering a better general theory of consumption, the article therefore challenges the assumption that there is a single phenomenon to which such a theory could apply. Consumption remains a useful comparison for some questions, and the task is to establish for which questions it is relevant and what it hides when it is applied to others.
II. How consumption became an apparently universal human activity
The verb to consume designates relatively specific operations: eating, using up, exhausting, destroying. Its Latin root, consumere, meant to take up completely, and fire consumes a house as surely as a guest consumes a meal. The noun consumer has acquired a much broader significance. In modern political economy a consumer is the counterpart of a producer or seller, someone who acquires goods or services for use rather than for resale or further production. This classification belongs to a historically particular architecture of economic activity, and nothing about its usefulness within that architecture shows that it identifies the fundamental organization of living. Historians have shown how recent the consumer is as a social identity. Frank Trentmann traces its emergence around 1900, when users of piped water and gas in British cities organized as consumers to contest prices and supply, and when consumer leagues in Europe and the United States began to claim a political voice in that name (Trentmann 2016). People had always eaten, worn and used things; being a consumer, as a way of standing towards the world and towards the state, was a new arrangement with a datable history.
Political economy paired production and consumption as complementary moments: production creates commodities, consumption is their destination. The pairing obscures much of what people actually do with things. They maintain them, repair them, store them, share them, adapt them, inherit them and keep using them for years. A coat is manufactured once, worn hundreds of times, mended, altered, handed down and finally cut up for rags. Which part of that history is consumption, and what is gained by treating the whole of it as a consumer practice?
Over the second half of the twentieth century, the economically defined category became an anthropological object. Consumer behaviour, consumer culture, taste, lifestyle and identity became fields of research, and in much cultural theory consumption became the general domain in which goods acquire meanings and individuals construct identities. Mary Douglas and Baron Isherwood defined consumption as "a use of material possessions that is beyond commerce and free within the law" (Douglas and Isherwood 1979). Daniel Miller made consumption the process through which people appropriate mass-produced goods and make them their own (Miller 1987). The category expanded from particular processes involving commodities into an apparently general account of how human beings relate to objects and experiences. The expansion was easy to overlook because consumption is an unmarked concept. A theorist who invokes biopower announces a theoretical choice; a theorist who invokes consumption appears merely to describe what people obviously do. When the broad use is questioned, the narrow one supplies reassurance, since nobody denies that people consume food, and the self-evidence of eating shelters the theoretical ambitions of a category that now covers identity, attention and relationships.
James Carrier's discussion of the concept shows the result with unusual candour. Surveying the field, he notes that his own working definition of consumption was simply "not production", that Douglas and Isherwood's definition is too vague and too narrowly tied to material possessions for many of its users, and that the term has come to cover so much that some regard it as effectively meaningless (Carrier 2021). A category defined by negation, as whatever is not production, can absorb nearly everything people do with things. Once contemplation, enjoyment and relationships can count as consumption, the category no longer requires acquisition, expenditure, incorporation or destruction, and it risks becoming another name for meaningful human activity.
A classification that emerged within particular political-economic circumstances has thus been extended far beyond them. The historical study of consumer capitalism, its department stores, credit, advertising and mass markets, is entirely legitimate, and consumer societies are real and consequential. What is illegitimate is turning their historically particular categories into universal analytical terms, as though all living were fundamentally organized around consumption.
III. First delimitation: consumption is not necessarily transactional
The first problem is the implicit identification of consumption with acquisition through a transaction. In a market economy the consumer typically appears as one party to a transaction: the consumer buys, the seller is paid. This corresponds to the architecture of transactive dualism, the value cosmology of separable actors, alienable goods, reciprocal transfers and transactions that can be identified as completed. Yet consuming a substance or using an object is a different operation from the transaction through which it may have been acquired. Buying food and eating food are two operations, and the same person need not perform both: a parent buys, a child eats.
The animal test makes the point plain. A bird eats a worm. A deer eats leaves. A hunter kills an animal and eats its flesh. There may be profound ecological relationships here, competition, appropriation, injury and death, but an ecological transfer of matter is no transaction. Predation involves no reciprocal exchange between contracting parties, and ingestion requires no seller, buyer or agreed transfer. Metabolic incorporation is as old as life, and human market arrangements are a very late addition to it. To take the consumer transaction as the implicit model of consumption is to read the whole history of eating through one recent way of obtaining food.
Many human activities show the same thing. Eating vegetables from a household garden, gathering berries, drinking water from a spring and eating a meal cooked by someone else at home need involve no transaction on the part of the person eating. The presence of a material transfer does not establish transactional coordination. Conversely, a transaction can occur without anything being consumed in the narrower sense. Someone buys land, acquires a painting in order to preserve it, or buys a tool that will be used for decades. A purchase can take place without the purchased thing being exhausted, destroyed or incorporated.
Three separable operations therefore need to be distinguished. Acquisition is obtaining access to something, by gathering, inheriting, receiving, taking or buying. Transaction is a particular form of transfer between distinguishable parties, organized through reciprocal or contractually specified arrangements. Incorporation or use is engaging with something during, after or entirely independently of its acquisition. In a supermarket the three may coincide within an hour; in most of human history and much of present life they do not. A theory that takes their coincidence for granted mistakes a historically particular arrangement for a universal relationship.
The distinction matters because a transactional conception makes the consumer appear as a bounded actor whose relationship to the world is organized through acquiring and using things. Much living is neither organized through bounded transactions nor adequately described as interaction between individuals and alienable goods. Consumption is thus often inserted into transactive dualism even when the coordination under study is not transactional at all. The first delimitation takes the form of a question: does an actual transaction occur, and if so, what part does it play in the coordination under investigation?
IV. Second delimitation: the occurrence of consumption does not establish its centrality
An activity can contain consumption without being about consumption. A birthday party involves eating cake. A family meal involves eating food. A religious festival may involve distributing and ingesting particular substances, as when worshippers in Bengal receive prasad, food offered to a deity and returned as blessing. A parent feeds a child. Consumption occurs in each case in a recognizable sense, and in none of them does that fact establish that consumption is the central phenomenon.
Take the birthday. The cake may be bought, displayed, cut, divided and eaten. The vital stakes, however, may concern recognition, attachment, continuity, shared memory and the celebration of someone's continued presence among the people who love them. Eating the cake is one component of a much wider living coordination, and much of what makes it matter happens before the first slice: the candles, the singing, the wish, the faces turned towards one person. A consumption-centred interpretation risks mistaking a visible, materially identifiable activity for the reason the occasion matters.
Feeding an infant is a sharper case. Milk is ingested, and nourishment matters enormously. Feeding also involves bodily attunement, warmth, holding, gaze, protection and a responsiveness that develops between two living beings over months. A theory of consumption cannot assume that the ingested substance supplies the primary explanation of the interaction. The opposite reduction is equally mistaken. In conditions of hunger, malnutrition or illness, adequate nourishment may be the most urgent stake of all, and reducing feeding to bonding would be a callous error. Which stake has priority must be established in the particular situation, since the analytical category cannot stipulate it.
Nothing here implies that eating is merely instrumental to survival. Pleasure, appetite, curiosity, sensory exploration and enjoyment are themselves significant, and a theory that reduced food to calories would miss most of what eating is for human beings and many other animals. These are better investigated as sensory pleasure, appetite and embodied engagement than absorbed into the general category of consumption. The question concerns what matters within the actual living coordination, and survival is only one of the answers it can receive.
The second delimitation parallels the one that applies to exchange. The existence of a transaction does not establish the explanatory centrality of exchange, and the occurrence of ingestion, purchase, use or expenditure does not establish the explanatory centrality of consumption. A theory has to identify what is actually at stake before deciding whether consumption explains anything about it.
V. Third delimitation: when everything becomes consumption, consumption explains almost nothing
If consumption includes buying, eating, wearing, watching, listening, reading, contemplating, enjoying, expressing identity and maintaining relationships, what does it exclude? Almost any engagement with anything can be redescribed as consumption, and the category acquires a seemingly unlimited applicability. An unlimited capacity to classify, however, is no evidence of a correspondingly wide capacity to explain.
The test is the null case. What would count as evidence that an activity is not consumption? Someone listens to music: music consumption. Someone refuses to listen: selective consumption, resistant consumption, or non-consumption as a consumer preference. Someone buys fashionable clothes: consumer identity. Someone refuses them: ethical or anti-consumer identity. When acquisition and refusal, enjoyment and rejection, participation and abstention are all classified within the same framework, its explanatory scope can no longer be tested. Refusal can certainly be studied in relation to consumption, and a boycott of a particular brand is organized around a commercial transaction that is deliberately withheld. Abstention should not, however, be read automatically as another positive instance of the category under examination, or the category becomes unfalsifiable. The null case tests the boundaries of the category, and it does not settle everything. An activity may unambiguously involve consumption while consumption has little relevance to what matters about it, and the decisive test is relevance to the question being asked, of which membership in the category is only a part.
Much of the expansion proceeds by metaphor. Food is consumed through ingestion. Fuel is consumed through combustion. A film is consumed through viewing, music through listening, an experience through participation. These usages involve quite different physical, causal and relational processes. A watched film is not used up, and a song heard a hundred times is, if anything, more fully present in the listener's life than before. The shared word establishes an analogy, and an analogy creates no explanatory unity. Nelson Goodman observed that metaphor can generate the similarity it seems to record, and that any two things resemble each other in indefinitely many respects (Goodman 1972). To call listening to music consumption helps to produce the comparison it appears merely to describe, and the discovery that music and food can be compared tells us little until we know why that respect of comparison matters.
A category can nonetheless generate an entire research field. Once music, clothing, pharmaceuticals, tourism and food are all called consumption, they become available for comparison as kinds of consumer practice, with journals, handbooks and conferences to match. Similarities produced by classification must not be confused with similarities in the phenomena. The resulting literature can become extraordinarily extensive without becoming more precise about the processes it studies.
The illusion of theoretical integration follows. A framework appears comprehensive because it can redescribe a vast range of activities in its preferred vocabulary, which is a different achievement from explaining how those activities work. The anthropology of consumption can seem to unite numerous research areas precisely because it has stopped discriminating adequately between them. The proper test is whether describing an activity as consumption identifies specific processes, causal relationships, constraints or vital stakes that would otherwise be hard to understand, and the mere possibility of so describing it settles nothing. Classificatory applicability must never be mistaken for explanatory relevance. The same comparison can be relevant to one inquiry and idle in another: a household expenditure survey has good reasons to set food and music side by side as claims on the same income, and a study of why someone returns to one song through the years of a life has almost none.
VI. Fourth delimitation: how the category of consumption reorganizes living
The modern consumer is far more than an observer's description. Consumer categories operate through markets, public institutions, legal arrangements, marketing, digital platforms and systems of measurement, and these classifications can alter the activities they describe.
Students are a clear case. Education involves teaching, learning, attention, effort, intellectual companionship, material resources and continuing changes in understanding and ability. Universities can nonetheless come to organize their dealings with students through the language of service provision, consumer satisfaction and measurable outcomes, and in the United Kingdom consumer protection law has been applied to that relationship. The classification selects particular aspects of education for institutional attention, and it can then shape what students expect, what universities prioritize and which kinds of educational value become visible. This is an instance of recursive uptake, and it does not show that every student or university already behaved transactionally before the classification arrived.
Patients described as consumers of healthcare show the same process. An ongoing effort to sustain or restore living coordination becomes organized around episodes of service, provider choice, satisfaction ratings and measurable outputs. Some of these measures are useful, and choice and feedback can protect patients against neglect. The patient's vital stakes, however, can extend far beyond what the consumer classification captures: a person with a chronic condition needs continuity, trust and a life that can go on around the illness, and none of these appears in a satisfaction score.
The language of content consumption is an especially clear contemporary example. Watching films, listening to music, talking with friends and following the news are classified through attention, engagement, viewing time and subscriptions. These metrics shape how platforms arrange content and how creators design it, so that songs are written to hold attention in their first seconds and videos are cut to the length the metrics reward. The consumer classification becomes increasingly consequential for the activities it measures.
A classification can thus help reorganize living until living increasingly resembles the classification. That does not show that the classification identified the fundamental phenomenon before the reorganization occurred. The fourth delimitation directs attention to the recursive effects of treating living beings as consumers, and away from the assumption that consumption was always the natural foundation of what they do.
VII. Mediational reductionism: why anthropology privileges meaning over living
The four delimitations apply to consumption theory in general. A further problem is specifically anthropological. A familiar justification for an anthropology of consumption holds that human consumption differs from animal consumption because it is meaningful. People do not merely eat, dress or acquire objects; their practices express cultural values, identities, social distinctions and collective classifications. The insight is often useful. The difficulty arises when multisymbolization is raised into the defining or determining feature of human engagement with material things.
Two senses of meaning need to be kept apart. If meaning refers to a living being's practical orientation towards what matters, then animals have it too: they discriminate between food and poison, danger and safety, companions and strangers. If meaning refers to symbolically articulated significance, then human beings elaborate it far beyond anything found in other animals. Neither proposition shows that symbolic interpretation should supply the general explanation of human activities involving materials. Human beings do not cease to be organisms because they possess language.
Consider someone who cooks the same cheap, familiar meal several times a week. The explanation may lie in hunger, taste, habit, availability, cost, nourishment, convenience and the need to get on with other things. It may owe nothing to the meal expressing a cultural identity, a class position or a symbolic opposition. Symbolic classifications can certainly influence food, and the counterclaim that meaning matters only in branded consumer capitalism would be equally wrong: food prohibitions, fasting, offerings and caste rules show how deeply symbols can organize eating. The more rigorous proposition is that the importance of symbolic meaning must be investigated rather than assumed.
Douglas and Isherwood made the assumption into a programme. Their advice to readers of The World of Goods was to forget that commodities are good for eating, clothing and shelter, to forget their usefulness, and to try instead the idea that commodities are good for thinking, treating goods as an information system and a nonverbal medium through which people make and maintain social relationships (Douglas and Isherwood 1979). The instruction is explicit mediational reductionism, and it is admirably candid about what it requires: embodiment, dwelling and multimateriality must be set aside so that multisymbolization can do the explaining. Pierre Bourdieu's Distinction (1979) showed with great empirical force that preferences in food, clothing, music and furnishing are patterned by class and serve to distinguish classes from one another. "Taste classifies, and it classifies the classifier." These are real observations about patterned practice. Why, though, should the symbolic differentiation of taste have explanatory priority over bodily requirements, material availability, established routines and the practical work of living? Class distinctions are real, and the problem lies in the promotion of one observable pattern into a general interpretation of the activity.
Daniel Miller's ethnography of shopping in north London shows what happens when the interpretive frame is loosened by its own evidence. The women Miller accompanied were, he argued, mostly shopping for others, and their purchases were acts of love and devotion directed at partners and children, a kind of everyday sacrifice (Miller 1998). The finding is striking, and its implication runs further than Miller took it. If what is at stake in shopping is care for particular people, then shopping is an instrument of being-with, and consumption may not be the most relevant comparison for understanding it. The alternative question is what living coordination is occurring and what its participants are trying to maintain, alter or make possible. Symbolic meaning may turn out to be indispensable, marginal or absent from the relevant explanation, and that cannot be known in advance. Meaning is one mediation of living, and living does not take place for its sake.
VIII. Graeber: the critique that came closest to escaping consumption
David Graeber's essay "Consumption" (2011) is the indispensable predecessor to this argument, and the problem of the category was his discovery before it was this article's. Graeber asked why activities as dissimilar as buying refrigerator magnets, putting on eyeliner and cooking dinner should be treated as instances of the same phenomenon, and he traced the answer to the history of a metaphor. The English verb derives from the Latin consumere, meaning to seize or take over completely, and Graeber showed how eating, the destruction of something by incorporating it, became the dominant idiom for desire and gratification in European thought (Graeber 2011). He also challenged the assumption that human fulfilment is fundamentally organized around acquiring and destroying things.
His historical account connected the concept to possessive individualism. As markets expanded and property became the model of social relations, people increasingly came to see themselves as isolated beings, bounded possessors whose desires were directed at things that could be acquired, enjoyed, incorporated or used up. Graeber identified a basic tension between that image and the relational constitution of human life, and in doing so he anticipated much of the criticism of transactive dualism.
His positive proposal was deliberately provocative. Instead of consumption, he suggested, anthropology should consider "the production of people (of which the production of things is simply a subordinate moment)": the activities through which human beings form, sustain and transform one another in relationships (Graeber 2011). He did not insist that everything must be either production or consumption, and he gave the example of a softball game, which is clearly neither. That recognition matters, because it concedes that the dichotomy itself has limits.
His alternative nonetheless placed great weight on the mutual creation of persons. The making of persons describes important configurations of being-with, and it does not exhaust living coordination. A person sleeping alone, an organism holding its body temperature steady, an animal searching for food and someone recovering from an illness may all be sustaining something indispensable without primarily engaging in the mutual production of persons. Describing care, pleasure and companionship as production also risks keeping the very political-economic vocabulary under examination, with production simply taking the place that consumption had occupied. The strongest criticism concerns what came next: Graeber saw the conceptual problem clearly, and his alternative did not systematically distinguish all the mediations and kinds of coordination that the consumption category had collapsed together.
Graeber asked why anthropology had grouped so many activities under consumption. The further question is how those activities can be investigated without replacing consumption with another indefinitely expandable classification. Rather than being fundamentally production, social reproduction, person-making or even metabolic remediation, they are different living coordinations whose organization and vital stakes have to be established one by one.
IX. Conspicuous consumption: why ostentation is not a type of consumption
Thorstein Veblen's The Theory of the Leisure Class (1899) gave the field one of its most durable concepts. "Conspicuous consumption of valuable goods", Veblen wrote, "is a means of reputability to the gentleman of leisure" (Veblen 1899), and for more than a century the phrase has stood for consumption used to display wealth and standing. What Veblen was explaining, however, was larger than the consumption of things. His concern was pecuniary emulation, reputability, the public demonstration of wealth, and the social significance of conspicuous leisure, expenditure and waste. His book is a theory of how people establish and communicate distinctions of standing, in which consuming valuable goods is one of several means.
Imagine someone wearing an extraordinarily expensive watch. Far from being consumed in any ordinary sense, the watch may be carefully maintained, protected from damage and kept to be passed on to a child. Its importance may lie in being recognized as expensive, exclusive or prestigious, and the relevant operations are display, comparison, recognition and perhaps competition. Why should this be called consumption? A more precise inquiry would investigate ostentation, wealth signalling, status competition, recognition-seeking or competitive differentiation, depending on what is actually happening. These terms are not interchangeable. Ostentation need not involve competition, since someone may display wealth to people with whom they are not competing at all, and status competition need not involve ostentation, since rivals can compete through restraint, connections or knowledge. Their very differences make them analytically preferable to a broad category that absorbs them all. None of this requires denying that the watch was bought. For a study of household debt, the expenditure may be the most relevant fact about it; for a study of prestige, display and recognition; for a study of inheritance, the watch's passage from one generation to the next. The same object supports several relevant comparisons, and none of them determines the others.
Ostentation also occurs without consumption. A family displays inherited jewellery at a wedding. A man uses a ceremonial title. A household shows visitors through an impressive house or an ancestral collection. The display may be highly consequential, and it involves no new purchase, no depletion of resources and no destruction of goods. People also display standing by giving to charity, hosting a feast, demonstrating exceptional knowledge or appearing in the company of influential friends. There may be transactions in some of these activities, but the common phenomenon, where there is one, concerns status and its recognition. The presence of expensive goods does not make consumption the defining operation.
Consumption equally occurs without ostentation. Someone buys costly food, clothing or equipment for the pleasure of it, out of practical necessity, for comfort or out of attachment, with no intention of impressing anyone. A climber's expensive rope is bought to hold a fall. Even where a commodity is visible to others, it does not follow that display explains the conduct. Whether competitive recognition is at stake has to be established.
Veblen's insight deserves to be kept. He identified an important, historically situated relationship between expenditure, ownership, honour, reputability and emulation, and he recognized that much of it ran through habitual conformity to a "standard of decency" in expenditure rather than through conscious attempts to outdo others (Veblen 1899). The objection concerns what the label later did: it made status processes appear to be fundamentally varieties of consumption. Where expenditure is causally central, it should be examined as expenditure. Where display or recognition is central, those processes deserve explanatory standing of their own. A misplaced classification can also hide a relevant similarity. An inherited watch worn to be noticed and a charitable donation made to be noticed fall under different economic headings, ownership and giving, though both may involve closely related processes of status display.
The same holds for Bourdieu. Taste can be patterned by class position and can contribute to distinction, and aesthetic enjoyment, practical preference and status competition are nonetheless different phenomena. That someone enjoys classical music does not show that status differentiation is the principal reason they listen to it, and that a preference is socially patterned does not establish its decisive stakes for the person who holds it. A theory of distinction can identify real regularities while overstating their explanatory relevance.
Conspicuous consumption, then, may be ostentation or status competition that happens to involve commodities, and it need not be consumption that has become conspicuous. The category invites researchers to ask what kind of consumption an activity represents before establishing what activity is taking place. More generally, many apparent subtypes of consumption may be better understood as entirely different phenomena. Renaming them changes more than the label: it changes what requires explanation, which evidence becomes relevant and what would count as a counterexample, and it lets rivalry and recognition be investigated where no commodity is bought or consumed at all.
X. Three ethnographic demonstrations: eating, refusing and taking medicines
Bengali offers a striking first case. The verb khāoyā, to eat, reaches far beyond food. One takes a bribe by "eating" it (ghush khāoyā), is beaten by "eating" blows (mār khāoyā), is deceived by "eating" a trick (dhokā khāoyā), and embezzles by "digesting" money (ṭākā hojom karā); to deprive someone of their livelihood is to "kill their rice" (bhāt mārā), and medicines too are eaten (oshudh khāoyā) (Ecks 2013). It is tempting to read these idioms as evidence of a unified cultural theory of consumption, a world organized around the belly and its appetites, or, following George Foster's image of limited good, a world in which one person's eating is always another's loss (Foster 1965). The idioms, however, distinguish quite different concerns: aggression, appropriation, vulnerability, livelihood, deception and the continuing requirements of living. To be beaten is to undergo something, to take a bribe is to appropriate something, and to lose one's rice is to lose the means of going on. The language of eating articulates concerns whose objects are often not ingestion at all, and linguistic similarity is here especially easy to mistake for analytical unity. For a study of the semantics of eating, the comparison is itself the object; for a study of corruption or of violence, the shared verb is a curiosity.
A second case comes from India's government schools, where midday meals are provided to children. There have been repeated reports of children, often at their parents' urging, refusing meals cooked by a Dalit cook. The episode can be classified as non-consumption or food refusal, a comparison that is relevant to a nutritional survey of the midday-meal scheme and barely touches what is happening between the children, their parents and the cook. The consequential processes are caste hierarchy, exclusion, embodied aversion learned at home, the organization of commensality, symbolic classifications of purity and pollution, and the institutional arrangements through which food is cooked and served. The refusal may also leave children hungry, so that nourishment is at stake alongside rank. Neither symbolic status nor nutrition alone captures its meaning and consequences. The analytically important object is the maintenance and enforcement of caste distinctions through the refusal of food, and calling it a consumer choice would hide exactly what makes it matter.
A third case concerns psychopharmaceuticals. What does an anthropologist explain by saying that someone consumes an antidepressant? The phrase may refer to ingestion, to a pharmaceutical purchase, to participation in treatment, or to an attempt to alter suffering and restore capacities for living. These are different operations. Consumption is far from irrelevant to medication, since ingestion, dosage and metabolism matter decisively for pharmacology, and the relevance of the comparison shifts with the question. A person may take the medicine in order to sleep, to work, to keep a marriage going, to reduce distress or to regain some continuity in everyday life. Advertising may promise more than that: in India, antidepressant marketing has offered patients the prospect of becoming fuller citizens of a modern society through the right drug (Ecks 2005). The advertisement's promises, the drug's effects on the body, the prescriber's involvement and the patient's own expectations may all matter, and none of them automatically determines the whole coordination.
The three cases point the same way. Bengali idioms of eating, caste-based food refusal and the taking of psychopharmaceuticals are different phenomena with different mediational configurations, causal relationships and vital stakes, and nothing is gained by grouping them together merely because all of them can be described in the vocabulary of consumption.
XI. Beyond consumption: a method for investigating what is actually happening
The positive contribution of this argument is a method for preventing a category from obscuring the phenomena it names, and another encompassing theory of consumption would defeat its purpose. The method has ten parts.
The first is to suspend the category. When an activity conventionally classified as consumption is encountered, the classification is set aside and the activity is described without the words consumption, consumer or consumer behaviour. What are the participants doing? Which things, organisms, environments and symbolic arrangements are involved? What changes in the course of the activity? Instead of music consumption, one describes listening, singing, remembering, enjoying, attending, performing or sharing music. Instead of conspicuous consumption, one describes displaying expensive objects, comparing standing, seeking recognition or maintaining prestige. The resulting descriptions may turn out to have little in common, and that is a finding. Suspending the category also makes it visible: the claim that an activity is consumption becomes the proposition that it is being investigated as consumption, which denies nothing and makes the choice examinable.
The second is to identify the operations precisely. The activities compressed into consumption include acquiring, buying, eating, digesting, using, maintaining, repairing, preserving, displaying, enjoying, listening, refusing, discarding, appropriating and destroying. These are not interchangeable, and there is no reason to assume they belong to one larger phenomenon simply because objects are involved. Some occur together; others are independent of one another.
The third is to determine whether transactions occur. Is there an exchange between identifiable parties? Is the transfer reciprocal or contractual? Does the transaction explain the subsequent activity, or merely provide one of its conditions? Could the activity occur without any transaction? These questions separate the acquisition of commodities from the many ways of engaging with them afterwards.
The fourth is to reconstruct the five mediations, investigating how the activity is coordinated through embodiment, being-with, dwelling, multimateriality and multisymbolization. The purpose is to avoid prejudging which mediations matter and how, and five obligatory paragraphs on every case would serve no one. In eating, bodily hunger may be decisive, while the availability of food, the demands of others, an institutional timetable or a religious prohibition may also shape what happens. The relative weight of these has to be established empirically.
The fifth is to identify the recursive counterparts. The consumer category encourages a simplified image of an individual interacting with an object. Someone eating alone, someone being fed by another person, and two people sharing a meal are engaged in quite different patterns of coordination, involving nonrecursive counterparts such as food and plates, selfrecursive coordination with one's own appetite and body, and interrecursive coordination with others who respond to one's responses. Living beings can interact without any transaction, and predation, feeding, care and competition must not be collapsed merely because something passes between the parties.
The sixth is to trace the recursivity levels. At L1, practices proceed without explicit attention. At L2, something becomes uncomfortable, difficult or unsustainable. At L3, participants articulate the difficulty. At L4, they classify it through established categories such as unhealthy eating, luxury consumption, consumer choice or addiction. At L5, researchers and others may examine those categories, including through a general theory of consumption. The critical question is whether the L4 and L5 descriptions capture what is occurring or substitute a conceptual framework for investigation. The levels are recursively connected: a classification of unhealthy eating can change how someone eats, and an institutional diagnosis of consumer demand can reorganize production and supply. Such effects must be investigated rather than assumed.
The seventh is to identify the actual stakes. What is the person or living being trying to accomplish, sustain, avoid, restore or transform? What would be lost if the coordination were disrupted? Is nourishment at stake, or bodily comfort, recognition, being-with, safety, pleasure, continuity, access to materials, a hoped-for future? The answer must not be predetermined by a familiar theory of needs, identity, class or social reproduction. Nor can the researcher assume that everything vitally valuable has already been articulated by participants, since much of what matters works at L1 and becomes noticeable only when coordination fails or begins to snag at L2.
The eighth is to select the most precise explanatory vocabulary, and only after examining the processes. A study of luxury watches may require an account of ostentation, prestige or status competition. A study of infant feeding may require an account of nourishment and interrecursive care. A study of medicine use may require an account of therapeutic attempts and metabolic remediation. A study of sustainable fashion may concern resource depletion, manufacturing conditions, waste, repair and ethical classification. These are potentially different objects of inquiry, and nothing guarantees that they are different species of one genus. Choosing among them means testing alternative comparisons against the same case and asking what each excludes: which consequential similarity is hidden because activities have been assigned to different categories, and which misleading similarity is created by placing them in the same one.
The ninth is to retain consumption where it helps. The word should not be abolished. Physiology studies the ingestion and metabolism of substances, energy research measures fuel consumption, economics uses operational definitions of household consumption expenditure, and anthropologists can study historically specific consumer institutions and practices. In each of these the comparison earns its place by its relevance to a specific question. The objection is to making consumption a general explanation of activities that have little in common beyond their inclusion in the category.
A brief illustration shows how the steps work together. Return to the guest with the expensive handbag. Suspending the category, one describes what she is doing: she carries a bag that she bought two years ago, keeps it in a dust bag between outings, and holds it so that its logo faces the room. No transaction occurs at the party, and the purchase that made the scene possible lies two years back. Nothing is being used up. Among the mediations, multimateriality and multisymbolization carry the weight, the bag as an object and the brand as a sign, while being-with organizes the whole scene: the guest is attending to particular people, perhaps a former colleague who has done better than she has, perhaps a host whose approval she wants. Embodiment enters through the way she holds herself, and dwelling barely enters at all. What is at stake may be recognition, or reassurance after a difficult year, or rivalry with one person in the room. The precise vocabulary may then be ostentation, recognition-seeking or status competition, and the evidence that distinguishes them lies in whom she watches, what she says and how she feels when no one notices. The word consumption contributes nothing to any of this.
The tenth is to avoid constructing another impossible object. Living coordination must not become a new term into which every empirical difference disappears. The five mediations do not establish that everything has the same cause, function or value; they provide a way of investigating processes without letting a previously selected category dictate what those processes must be. The value of the framework depends on whether it yields more precise empirical distinctions, and a vocabulary of wider applicability would be no improvement. Replacing the word consumption everywhere with the words living coordination would miss the point, which is to discover which particular living coordinations the word consumption has concealed.
XII. Conclusion: “Consumption” Is Not One Problem
Return to the opening scenes. The chimpanzee eats a banana. The woman gathers mushrooms. The guest displays a handbag. The man listens to a song. The patient takes an antidepressant. An anthropologist can classify all of them as consumption, and the classification does not establish that they are instances of one explanatory phenomenon. For some questions the comparison is illuminating, and for most of what matters in these five lives it is idle.
The anthropology of consumption combines several mistakes that reinforce one another. It projects transactional arrangements beyond the situations in which transactions occur. It mistakes the occurrence of consumption for its explanatory centrality. It expands the category until its limits can no longer be found. It privileges symbolic meaning without establishing that meaning is the decisive mediation. It turns distinct phenomena, ostentation, status competition, nourishment, care and enjoyment, into supposed varieties of consumption. And it often neglects how consumer categories themselves reorganize living coordination. The transaction supplies the model, symbolic meaning supplies the justification for extending it, and the indefinite expansion of the category supplies the illusion of explanatory comprehensiveness. Together with an equally expandable concept of exchange, the result is an image of human life as fundamentally a matter of obtaining, exchanging, using and disposing of objects.
A discipline committed to understanding the diversity of human life has thus developed a category that makes fundamentally different activities look like variations of one thing. The opposite direction begins with actual living coordination, establishes its mediations, counterparts, recursive processes and vital stakes, determines which activities occur and what explanatory role they play, and keeps the term consumption only where it makes a demonstrable analytical contribution. The result is a more discriminating anthropology, in which many activities once classified as consumption recover their identities as distinct phenomena worth investigating in their own right, and an anthropology of consumption that encompasses still more activities would be its opposite.
Anthropologists have too often mistaken a remarkably elastic classification for a coherent object of investigation, and the problem with the anthropology of consumption lies there, far more than in any varieties of consumption it might have overlooked. Eating, caring, displaying, acquiring, listening, enjoying and sustaining life do not become the same activity because we call them consumption. Anthropology will understand them better once it stops explaining why they are all forms of consumption and begins asking which comparisons matter for what it wants to know about them.
References
Bourdieu, Pierre. 1979. La distinction: Critique sociale du jugement. Paris: Éditions de Minuit.
Carrier, James G. 2021. "Consumption and Meaning." In Economic Anthropology, 103–116. Newcastle upon Tyne: Agenda Publishing.
Douglas, Mary, and Baron Isherwood. 1979. The World of Goods: Towards an Anthropology of Consumption. New York: Basic Books.
Ecks, Stefan. 2005. "Pharmaceutical Citizenship: Antidepressant Marketing and the Promise of Demarginalization in India." Anthropology & Medicine 12 (3): 239–254.
Ecks, Stefan. 2013. Eating Drugs: Psychopharmaceutical Pluralism in India. New York: New York University Press.
Foster, George M. 1965. "Peasant Society and the Image of Limited Good." American Anthropologist 67 (2): 293–315.
Goodman, Nelson. 1972. "Seven Strictures on Similarity." In Problems and Projects, 437–446. Indianapolis: Bobbs-Merrill.
Graeber, David. 2011. "Consumption." Current Anthropology 52 (4): 489–511.
Miller, Daniel. 1987. Material Culture and Mass Consumption. Oxford: Blackwell.
Miller, Daniel. 1998. A Theory of Shopping. Cambridge: Polity Press.
Trentmann, Frank. 2016. Empire of Things: How We Became a World of Consumers, from the Fifteenth Century to the Twenty-First. London: Allen Lane.
Veblen, Thorstein. 1899. The Theory of the Leisure Class: An Economic Study of Institutions. New York: Macmillan.