If there is one universally recognized function of the family it is ‘consumption’. It would be tedious to list all the books and articles which mention this, because there is no sociologist, and more generally no author dealing with the family, who does not at least allude to it. It is presented as one of the principal functions of ‘the modern family’.
If it is granted that the family is the institution (or one of the institutions) which fulfils this function, we might have expected that the next step would have been to study the ways in which the family satisfies what are undoubtedly seen as some basic biological needs of its individual members. But despite the social and theoretical importance of both the family and the ‘function’ of consumption, there is a strikingly poor literature on the topic. Not a single known study of the family takes consumption as its theme of research, or even sets out the ground for such research.
If consumption and the family were not the object of specific investigations, we might at least have expected to see it discussed in general theoretical introductions. But, after its obligatory and quasi-ritual mention, it is little developed. Indeed, the assertion of the existence of a consumption function is often put in the form of a negatively phrased sentence. That is to say, the function of consumption is presented as the only remaining function of the family within the economic order: what remains to it of a glorious past, of the global economic role it used to play. Its mention is an integral part of the – often advanced, never substantiated – thesis that the family in general (and not certain forms of the family) has recently been excluded from any role in production whatsoever (and not only from production for the market). It is as if consumption was put forward to give credit to the thesis of the loss of the family’s role in production, and at the same time to affirm that – despite this vicissitude – the family continues to be necessary within the economic order. Hence, even at a theoretical level, the function of consumption is not treated in and of itself by those who study the family. E. M. Duvall’s sentence (1957, p. 58) ‘Families have shifted from production to consumption’ is exemplary of this kind of thinking. It is considered only in a general historical perspective: from the point of view of the evolution of the family and its gains and losses of ‘functionality’.
When we look at past work on family consumption, it rapidly becomes clear that the term ‘consumption’ is used to designate market demand. The titles of articles and journals lead us to think that what is being studied is individual consumption, but the consumption they describe is not that of any actual person, but is rather the purchase of goods and services on the market by households (generally in the person of the housewife). Such studies let it be thought that the family, which is a collective agent on the market, is equally a collective agent in consumption. 1 An INSEE study (the French equivalent of the Government Social Survey’s Family Expenditure Survey) says explicitly:
The field covered by this enquiry is that of expenditure on goods and services: purchase of products, consumption taking place outside the home, and payment for loans and services.
It is clear here that all consumption by members of the household (and this includes children at boarding school for instance), wherever it occurs, is taken into account in evaluating the standard of living of households.
Thus the use of the term consumption implies that individual consumption is being studied, while the way in which consumption is observed in practice – the relating of all consumption to the household – requires that distribution within the family should be studied. But not only do studies of individual consumption or sharing within the family not exist, the themes are not so much as broached, even theoretically, and it is precisely the choice of the household as the unit of observation which prevents such studies being possible using existing data. Taking the family as a unit does not allow family consumption to be studied – only the consumption of aggregates of families. What is studied is no longer the families themselves, but the way in which they differ from each other, or form groups. Moreover, the only difference between families which these studies are explicitly interested in studying, is that of ‘the comparative standard of living of different socio-economic groups’ (Jousselin 1972, p. 141).
This comparison itself actually also suffers from the definition of the household, e.g. servants (waged and apprenticed) lodging in a household are held to be part of it from the point of view of consumption. The result is that studies of the standard of living of farm workers’ households, for example, do not include those who lodge with – and who are consequently trapped by – the household of their boss. And these are those whose standard of living is lowest. Excluding them from farm workers’ households has the effect of raising the average standard of living of the latter, while their being ‘captive’ has the effect of lowering the average standard of living of the households to which they are attached, i.e. those of the class of their masters. These two effects together lead to a not inconsiderable diminution of the economic distance between the two classes.
But distortions brought about in comparisons of social categories are a minor defect compared to the major sin of considering the very place – the household – where certain class relations are exercised (e.g. those of servant and master) as the place where they are annulled.
The absence of studies of distribution has a positive meaning. It means that the only pertinent perspective is how the family is a unit within a larger whole, because this is the only perspective considered. Above all, it lets it be thought that the family, a unity vis-à-vis the outside, is also one within itself. One of the images which the term ‘unit of consumption’ evokes is that of common – i.e. homogeneous – consumption. It connotes at one and the same time common consumption, and undifferentiated consumption.
Differences of consumption within families
However, such connotations of common and undifferentiated consumption are contradicted by the facts of everyday experience. Here the disparities of consumption between family members are not only visible, but recognized as constitutive of family structure. Differences in consumption are seen as correlated with the existence of different family statuses. Differential consumption plays a major role both in the perception of these statuses by outsiders and in the appreciation of their particular statuses by those involved.
Existing studies of consumption are, however, based on the opposite assumption. And they do not rest content with ignoring individual consumption: they pretend to know about it without having studied it. Thus:
The average annual consumption per head … is obtained simply [sic] by dividing the values entered in the table … by the number of persons.
(INSEE 1973)
It should be remembered, however, that among the individuals whom we are thus invited to consider as benefiting in equal shares from all the goods consumed in the household to which they are attributed, are not only children in boarding schools, and soldiers on military service, but also servants, waged employees and apprentices. Thus, while pretending to ignore the whole topic, existing studies of consumption in fact assume (impose) a theory of distribution – an egalitarian theory.
It is likely that the processes described above are no chance effects and that their convergence is no coincidence. The use of the term ‘unit of consumption’ – which in denoting a simple unit of reckoning connotes a unitas (union and communion) – tends to make the study of distribution seem pointless; and statistical practice, for its part, by always taking the household as the only unit of observation, makes any empirical research impossible. All these processes converge to prevent any study of real distribution, for on the one hand such a study would risk undermining the whole basis of existing research by showing it to be founded on an implicit postulate – that of egalitarian distribution; while on the other hand it could not but confirm what is apprehended impressionistically by everyday experience – the existence of differential consumption.2
Study of the consumption function of the family should consist of studying the role of the family as the distribution centre for its members, and research should take as its object the effect of family status on individual consumption. But, as has been noted above, not one of the studies which refers to the family as a unit of consumption so much as outlines the limits of what does or does not enter into this unit, so the very framework of the research has still to be defined. Does individual consumption within the family involve consumption effected collectively, with all the members of the family present, regardless of place? Or is it consumption that occurs at home, whichever members may be present? Or is it consumption by members of the family, whatever the place and whichever individuals are present? Among the criteria which could be envisaged, besides the place (at home or outside) and the presence or absence of the family as a collective, must be the nature of the consumption. Would specific consumption (e.g. connected with a job) be opposed to common consumption, or consumption of the same sort of thing (e.g. consumption of food) – these latter alone being considered familial?
If the subject of research is the role of the family as a distribution point, it seems obvious that all individual consumption should be considered as familial since it is based on the status of the individual in the family whatever the place, the modalities, or the form it may take. But in the absence of even elementary reflections and investigation in this area we must proceed empirically and cautiously. In fact it is much more a matter of using examples to set out the directing hypothesis for a new approach to family consumption than of stating the methodological outlines for a systematic study.
I shall try to set out the outlines of such an approach in the remainder of this article, using examples chosen from the area of non-specific consumption, effected mainly at home, even if not in the presence of the whole family, since such consumption, and particularly the consumption of food, is the most evidently familial. It is the family seated around the table which most approximates the image of a really communist community, of a really equitable distribution, and which seems most sheltered from the effects of hierarchy.
I shall also deal with families on very low incomes, since there is a sentiment that inequality is less cruel when it is a case of individuals getting more or less of what is already a surplus, rather than when it is a question of individuals getting more or less than the minimum needed for a healthy life. It tends to be thought that families on the breadline must and do share what little they have.3
Both experts and the uninitiated like to situate the western version of ‘subsistence’ within rural, and above all within peasant, families. Here production for self-consumption is relatively important compared to that destined for the market, and this suggests a self-sufficiency, especially in food, which though far from existing in reality, is close to the golden age of the popular imagination (which is curiously situated in the nineteenth century).
It is in this type of family that some of the lowest incomes within industrial society are to be found and here that the standard of living is at its lowest. It is also here that it is most recognized that all family members do hard physical work. It is thus the last place where one would expect to find differential consumption of food. Therefore, if it can be shown that such differential consumption does exist, there seem good grounds for expecting it to exist in all families: that it is part and parcel of the structure of the family. That is to say, the stress given here to the rural family, and to the consumption of food, is not due to a particular interest in these areas as such, but rather to a belief that once the fact of differential consumption is established and established here (for the reasons evoked above), it will require further research into its principles and functioning, i.e. its existence as an institution. It will involve, in sum, the freeing of a problematic which will allow us to return to new concrete studies since future research will no longer be aimless: its problematic will have been constructed.
The distribution of food in peasant families
There is, needless to say, no scientifically collected information which can be used in considering whether there is or is not differential consumption in poor farm families. On the contrary, so-called scientific data have been collected in just such a way as to mask it. But, as was said earlier, the point of this essay is not to present new facts, but to look at facts, which are universally known to the social actors, from a new angle.4 So I shall therefore draw on descriptive studies and personal knowledge.
In the traditional rural family (of the eighteenth and early nineteenth centuries in Britain, and still today in marginal family smallholdings of the type that predominate in south-west France and much of southern Europe), consumption of food varies greatly according to the individual’s status in the family. This variation concerns the quantity of food and sets apart primarily children and adults, and women and men. But among the adults the old eat less than those who are mature, and the junior members eat less than the head of the family. It is he who takes the biggest pieces. He also takes the best: variation concerns quality as much as quantity.
Children are fed exclusively on milk, flour and sugar until two or three years old. The old, particularly the infirm elderly, return to a similar regime based on cereals and milk, bread-soups (panade) and broths.
Meat is rarely on the menu, and even more rarely on the menu for everyone. It often appears on the table to be consumed only by the head of the family, especially if it is butcher’s meat. Less expensive meat – chickens reared on the farm, preserves made at home – are not subject to such exclusive privilege. However, women and children will never have the choice piece, which is reserved for the father (or, on social occasions, for distinguished guests). Thus according to Cazaurang (1968), the prime pieces of ham, a prime food in itself, fall to the future son-in-law. Infants and the elderly never touch it. Alcohol is another food whose consumption is strongly differentiated. It is for adult men, to the exclusion of women and children.
Respect for food prohibitions is obtained by both coercion and the internalization of these prohibitions. The physical infirmity of young children and the old makes coercion so easy that it becomes not useless but invisible. It is mainly necessary, and becomes visible, in relation to children during the period when they are ‘thieves’: i.e. when they have not yet internalized the prohibitions.
Hence many types of food which are kept in the kitchen are put in high-up places, on hanging shelves (planches à pain) or on the tops of cupboards, where only people of adult height can reach them. This coercion by height is so classic that many folk tales have as their hero a child who has decided to outmanoeuvre it. The tale generally tells of the confident solution of the problem by the hero using a stool, and of the unhappy outcome in a punishment, either mediated (inflicted by an adult hand) or immediate (coming from the sky in the shape of indigestion). A brand of jam has even chosen for its trademark the picture of a little girl dipping her fingers into a jar: she is perched on a chair.
But if certain foodstuffs are physically protected only from children, others are protected from the whole family:
Provisions which it is thought should not be allowed into the kitchen are put in the bedroom, especially in the master’s bedroom. For pieces of pork meat, such as sausages, the stay in the upper storey enables them to finish drying out. Further, it shields them from the temptation of the young, who are always hungry. The same line of thought leads to the week’s supply of bread being put on a shelf from which it is only given out as needed
(Cazaurang 1968, p. 97, t II).
Some of the measures which back up prohibitions with physical obstacles apply to the whole household – except for the women, or rather except for the mistress of the house. These measures would in fact be inconvenient if applied to her because it is she who prepares all the food. She therefore has access to all the foodstuffs, even to those which she does not eat. But this access is clearly tied to her operations as preparer. Alcohol escapes her operations because its preparation is a masculine prerogative. The physical taboo to which it is subject may extend to the mistress of the house: often the ‘master’s’ bottle is touched only by his hand.
Repression in all its aspects – punishments and threats, verbal injunctions, physical obstacles and taboos on contact – play only a security role, except with regard to children, or perhaps even (as in the case of the bottle of alcohol) only a symbolic role in founding and maintaining differential consumption. For this is essentially a customary act (i.e. the constraints are internalized and reproduced as spontaneous behaviour by those involved). A whole corpus of proverbs, sayings and beliefs are both tokens of the content of the roles and the justification for these roles.
Sometimes these precepts seem like observations from experience – ‘women eat less than men’. Sometimes they are in the shape of advice on hygiene – ‘such food is “bad” or “good” ’ – with the prescriptive aspect on differential consumption only appearing in the second part of the phrase, where it is revealed that this ‘goodness’ or ‘badness’ strikes the organs in a selective manner according to the status of their possessors. Thus ‘jam spoils (only) children’s teeth’, ‘wine gives (only) men strength’, etc. The waiter in the restaurant where the young David Copperfield was stranded when he was travelling alone explained to him in the same vein that the beer which had been served to the youth would be fatal for him, and he saved David from death by gulping it down for him. At other times the norm is prescribed under the guise of aesthetic considerations – ‘There’s nothing more ugly than a drunken woman’ – or moral – ‘a woman who drinks is worthless’ (femme de vin, femme de rien) – which completely masks the repressive aspect, since it leaves those concerned free to be ‘ugly’ or ‘worthless’ and passes in silence over the anticipated benefit of such repression, i.e. the monopoly of a prized commodity. One could set against this the naïvity of an old farmer in the Lot region of France who exclaimed: ‘Goodness me, a woman smoking … it’s the first time I’ve ever seen that … but why not, after all, smoking’s a pleasure.’
The total absence of proteins from the diet of infants and the elderly leads to food deficiencies which have serious repercussions on the development of the former and the ageing of the latter, and the life expectancy of everyone. Their relative absence in the diet of women leads to consequences for their general state of health, whose effects are doubled by the physiological burden of pregnancies, as was previously evidenced by the very high rates of maternal and infant mortality in rural areas. Nevertheless, it is held that babies and children do not need meat, and that women have ‘less need’ of it. Men, however, ‘need’ such noble food. Vegetables which do not ‘hold to the body’ and do not ‘sustain a man’, apparently nourish women and children.
Indigenous theory suggests a relationship between the stature of the individual and the quantity of food necessary for his or her constitution.5 That this is a rationalization and not a principle of distribution is evident from the number of exceptions it suffers: a husband, a master, a father, or an eldest son, however puny he may be, does not give up his privileged share to a wife, a worker, a child, or a younger sibling, however heavily built or tall.
The theory of differential needs allows a third level of argument – that of differential expenditure of energy. This form of argument does not rest on the measure of energy really expended by the individual, but establishes an impersonal relationship between an activity and the expenditure of energy. This relationship is based on classifying activities into ‘heavy’ and ‘light’ work, but the classification is not based on the actual expenditure of energy required by the activity considered, but rather by the nature of the activities. It is not the technical operation itself which is the real criterion of the classification (carrying water is considered to be ‘light work’, carrying manure is ‘heavy’), nor is it the labour of the task (cutting corn with a scythe is ‘heavy work’, gathering it into bundles and binding it is seen as ‘light work’). Rather, throughout France, carrying water and gathering are, or were, exclusively work for women, while other sorts of carrying and harvesting were men’s work. The criteria of classification of work into ‘heavy’ and ‘light’ rests, in fact, on the status of those who usually do it.
Certain work, reserved for men and hence supposedly ‘heavy’ in some regions, is reserved for women in others and there changes its qualifications. This applies, to give just a couple of examples from among many sex-related tasks, to earthing up potatoes and driving draught animals. When women do supposedly ‘heavy’ work in one particular region – either in an exceptional way, at certain times of year, or in an ordinary way, as in Brittany or in the Alps where they do all agricultural work – the evaluation of the energy they expend and need is not thereby modified. This is not surprising since this expenditure and their real needs are never measured nor compared. The simple counting of hours of physical activity per day (more than a third higher on average for women than for men) would lead one to think that, contrary to indigenous belief, women’s expenditure and hence need for energy would be greater than men’s. But the theory of ‘needs’, while invoking explicitly or referring itself implicitly to objective physiological imperatives, in fact ignores them totally.
Does it, then, take into account subjective needs and desires? Still less. It is clear that in determining the ‘needs’ of a given individual, the evaluation of those ‘concerned’ does not enter into it. The feeling of hunger experienced by children and adolescents does not lead to a conclusion that they need food. On the contrary, in reply to requests there is a set response: ‘you don’t need it’, which suggests that need is different from, external to, and even antinomic to desire. The theory of needs thus calls on objectivity as against subjectivity, albeit (as we have seen) refusing any objective measurement.
This double contradiction is well expressed in the previously cited passage by Cazaurang:
It shields them (the pieces of meat) from the temptation of the young, who are always hungry … the week’s supply of bread … is only given out as needed.
The needs to which he alludes are thus not those of the youngsters. Their present hunger will not be satisfied, and had their previous needs been covered, they would not be ‘hungry’.
This quotation shows that a state of hunger is considered to be normal among the young, or rather that satiation does not form part of the needs which are recognized for them. ‘To eat one’s fill’ is one of the pleasures of life and this objective always runs the risk of not being achieved. Nonetheless, a chronic feeling of hunger is not considered an attribute of adults as an age class, although it is attributed as a distinctive characteristic to adolescents in rural society. ‘Hunger’ as concerning, not specific cases, but a whole category of individuals, ‘the young’, is considered not a characteristic of their social condition but an irremediable physiological fact. In other societies, such as that of North America or even urban France, a state of perpetual non-satiation appears as just as subjectively undesirable and objectively injurious among the young as among adults.6
When peasant farmers say – and most do concede it – that ‘we live better than before’, it is often primarily to evoke those changes which have occurred in everyday experience. In this regard, today is compared advantageously to yesterday. This ‘before’ is repeatedly evoked with bitterness, as a period of deprivation of food, and in all cases this relates to childhood.
I remember, when I was a kid, I went out in the morning with the sheep. I went out with a ‘drubbing’ and that’s all I had till evening (from an interview in the Lot Region).
The maintenance of differentials in consumption
If coercion is primarily used to make up for the lack of internalization of prohibitions among the young, and to create them, this is never so perfect that some slackening is not to be found. Between pure coercion and pure internalization, gossip plays a role, calling in the last resort on the presence of others and on shame, or its inverse, honour.
As Cazaurang again says:
A small gesture of an earlier mistress of the house is worth pointing out. She used to profit by the absence of other family members to yield to her gluttony. She would make herself some separate small dishes or simply coffee. If an intruder arrived unexpectedly, the sinful object was swiftly slid into the unlit oven near the hearth
(Cazaurang 1968, p. 124, my emphasis).
If for the young, food prohibitions – even when internalized – remain as constraints, especially since they are linked to a necessarily transitory status, for women they are integrated into a wider repressive system which allows a greater flexibility in its details. This system is the ideology of the role of wife and mother.
Women are in practice managers of the home and like all overseers they find themselves confronting situations for which no instructions exist. At such times a general principle takes over from the precise prohibitions which have become inappropriate. This general principle is simple: the wife and mother should always preserve the privileges of the husband and father, and ‘sacrifice’ herself.
Different modalities are used to this end in different societies. In Tunisia, for example, differential consumption is effected in a radically different way. Men have two or three meals a day while women have only one or two, and these meals never coincide. The women eat foodstuffs prepared once a year and obtained from second quality produce. The meals they make for men on the other hand use fresh and best quality ingredients. The rigorous separation of time, place and the basic substance of the meals makes any competition for the food between men and women impossible (Ferchiou 1968).
In France today, except for a few specific prohibitions – such as alcohol and tobacco – men and women eat ‘from the same table’ (au même pain et pot). Differential consumption derives essentially not from prohibition on this or that food, but from attributing women the smallest and most mediocre share of each food. It is difficult to say if it is the circumstances – sharing the same meal – which make necessary the creation and application of a general principle, or whether it is the existence of this principle which makes possible the preparation of but one meal. Perhaps it would be more appropriate to say that only such a principle could give an account of the variability of content of differential consumption.
In a particular social situation, in a given family and at a given standard of living, the content is not so flexible: the same dishes appear regularly on the table each week and it is not necessary to work out a new evaluation and a new distribution each time. The shares are fixed once and for all: in each family and in each chicken there is ‘father’s bit’.
Here again restrictions are experienced differently according to the degree of internalization and the transitory character or definitive status to which they are attached. For children, especially male children, they are persecutions on which they are revenged from the first occasion when they have access to the ‘father’s bit’ which they have coveted for years. Women, however, think that they have chosen the piece to which they are entitled.
For instance, a young peasant farmer invited two city women to share his tea and opened a tin of paté. His aunt, an old woman who kept house for him because his mother was ill, was there. On her bread she put only the fat from around the paté, which had been scorned by the three other diners. The system not only requires women to restrain themselves, but also allows them a certain latitude by making them responsible for taking the decision as to the form of their restriction. Thus the meat of the paté had doubtless never been expressly forbidden to this old woman; but the obligation to leave the best part to others had been internalized as a moral imperative. She could have complied with it in a different way; she acted on her own initiative in giving herself the worst part; and above all the precise way of doing it was up to her. This attribution is experienced as a free choice and is often explained by the ‘ordinary’ motivation for choice: personal preference. When asked, the old woman replied that she liked fat.
But there is absolutely no need for sacrifice to be liked: it becomes second nature. The mistress of the house takes the smallest steak without thinking, and will not take one at all if by chance there are not enough for everyone. She will say ‘I don’t want any’; and nobody is surprised, she least of all, that it should always be the same person who ‘doesn’t want any’. There is also no need at all for her to refer to the ideology of sacrifice as an integral part of feminine nature, nor that she be aware of her generosity or abnegation. Recourse to a universal principle supposes an out of the ordinary situation where the purely mechanical conduct of everyday life no longer suffices to guide action.
When one moves from the country to the town, and from low income sectors to higher sectors, consumption of food increases and differential consumption becomes less marked in this area. Since the level of food consumption is higher, it might be expected that basic needs are better covered and that differences of consumption would more and more concern less visible qualities and modalities. Indeed, food being sufficiently abundant, it might be expected that differences in food consumption would tend to disappear completely and be replaced by, or only exist in, other areas.
However, the flexible character of differential consumption, the fact (discussed above) that it is not the specific content but the principles of attribution which are defined, allows other expressions of subordination when for one reason or another the household’s scale of relative values is modified. One example can illustrate this move back to using food to express status differences, and this also illustrates the flexibility of the system.
In the last ten years France, and Paris in particular, experienced a shortage of potatoes which lasted for a fortnight. Since the demand for this basic commodity is relatively inelastic, prices rose and queues formed in front of the greengrocers. When questioned in one of these queues by a radio interviewer, a woman replied: ‘I’ll keep the potatoes for my husband. The children and I will eat pasta or rice.’ In spite of the relative expense of potatoes it would not have been beyond the budget of the family to get enough for everyone, given the subjective importance they attach to them. On the other hand, if the value of the gratification did not compensate for the budgetary sacrifice, as the renunciation by the wife and children suggests, the husband should also logically have eaten pasta and rice. The solution adopted seems to be explained neither by the physiological impossibility for the husband to absorb products (replacements in this particular situation) which were in any case consumed in almost as regular a way as potatoes, nor by the economic situation of the family, but rather by the symbolic necessity of marking privileged and statutory access to goods which are rare (or become rare) – this access being both the sign and at the same time the reason for the hierarchy of consumption.
If differentiation were studied in all sectors of consumption it is likely that the following principle and its corollaries would be confirmed:
1The rarest goods in each sector, and the most prestigious sectors of consumption, are subject to privileged access.
2The relative difference between the standard of living of different family members stays more or less constant in all social situations (and increases in absolute value as the privileged access concerns more and more costly goods and/or the differentiation is exercised on an enlarged global volume).
Indeed, with growth in the part of the budget which is available for spending on things other than food, forms of consumption develop which were previously of little importance or non-existent. The raising of the general standard of living may thus allow the development of differentiation in certain existing areas. In addition, it allows the emergence of new areas of consumption which are fresh fields for the exercise of differentiation. For example, the acquisition of a car by a household in which previously everybody travelled by public transport, not only considerably increases the global difference in consumption – variance in the standard of living – between the user of the car and other family members, but above all it introduces differentiation into an area – transport – which up till then was undifferentiated.
The study of differential consumption cannot be reduced to the study of quantitative differences in access to particular goods, however, it is also qualitative. Does a child being taken for a Sunday outing consume the family car in the same way as the father who drives it? Above all, does it consume the same outing? The problems which are currently being put forcefully in rural areas when two generations live together, reveal – if one listens to those concerned – that the conflicts experienced divide not ‘the generations’, but rather concern the ‘freely chosen’ consumption which the ‘invited’ children want, and the ‘compelled’ consumption which is ‘given’ them (imposed) by their parent-hosts.
These examples seem to indicate that ways of consuming are perhaps more important than quantities consumed. But up to now the study of consumption has always been preoccupied with – has always meant exclusively – volumes, and the very existence of modes of consumption has not even been hinted at.
Yet consumption after all concerns not only goods but also services, and if the classical economic studies sometimes include under the rubric of ‘self-consumption’ goods which are made at home, they always ignore the services produced in the household. However, despite what the titles of research on household budgets may lead us to think, household consumption does not only involve what is bought on the market: we do not eat raw steaks or unpeeled potatoes at our family tables. We consume not only primary materials but also their preparation: the housework of the ‘mistress of the house’ (work of which the preparation of food constitutes only a part). The provider of these services does not consume them in equal manner with the non-providers, for diverse reasons (of which some are obvious: e.g. you cannot serve at table and be served at the same time). This point is considered further in Chapter 5, ‘Housework or domestic work’.
Taking these services into consideration overturns not only the existing accounted evaluation of family consumption, it overturns at the same stroke the evaluation of family production, because these services are also ‘self-produced’. Above all, it re-poses, at the level of production, the problems of the meaning of the very term ‘unit’ when applied to the family – i.e. the problem of the internal functioning of the family as an economic institution.
* This paper first appeared in Cahiers Internationaux de Sociologie (1974), pp. 23–41. An English translation was included in C. C. Harris et al. (eds.), The Sociology of the Family: New Directions for Britain, Sociological Review Monograph, 28, (1979).