Test Question

Rethinking Childhood Author(s): Leena Alanen Source: Acta Sociologica, Vol. 31, No. 1 (1988), pp. 53-67 Published by: Sage Publications, Ltd. Stable URL: http://www.jstor.org/stable/4200684 . Accessed: 11/05/2011 16:09

Your use of the JSTOR archive indicates your acceptance of JSTOR's Terms and Conditions of Use, available at . http://www.jstor.org/page/info/about/policies/terms.jsp. JSTOR's Terms and Conditions of Use provides, in part, that unless you have obtained prior permission, you may not download an entire issue of a journal or multiple copies of articles, and you may use content in the JSTOR archive only for your personal, non-commercial use.

Please contact the publisher regarding any further use of this work. Publisher contact information may be obtained at . http://www.jstor.org/action/showPublisher?publisherCode=sageltd. .

Each copy of any part of a JSTOR transmission must contain the same copyright notice that appears on the screen or printed page of such transmission.

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide range of content in a trusted digital archive. We use information technology and tools to increase productivity and facilitate new forms of scholarship. For more information about JSTOR, please contact [email protected].

Sage Publications, Ltd. is collaborating with JSTOR to digitize, preserve and extend access to Acta Sociologica.

http://www.jstor.org

Acta Sociologica 1988 (31), 1:53-67

Rethinking Childhood

Leena Alanen Institute for Educational Research, University of Jyvaskyla

The study of children is either totally absent in sociology or is treated within very limited contexts which are considered marginal for sociological theory and research. The paper aims at examining why there is no sociology of childhood and what kind of sociological rethinking is needed to bring children into sociology. First, the conventional approach in child-related themes - recognized as 'socialization' - is discussed. Secondly, sociological notions of 'the family' are taken up and criticized on the basis of the feminist project of deconstructing the family. Finally, childhood is considered as a social construct and some demands for research on children and childhood are developed that may help to avoid the long-held and unnecessarily limiting views on these topics and make way for the emergence of a sociology of childhood.

'The study of children and child development is both empirically and theoretically central to the discipline (of sociology)'. This assertion is made by the two editors of a recent book on sociological studies of child development (Adler & Adler 1986). They go on to justify why this is so, yet are forced to admit that there is a great paucity of sociological literature in this area. Another writer in the same book (Ambert 1986) documents this latter fact: there is indeed a near absence of studies on children in (North American) sociology. The fact has been recognized elsewhere too, and first attempts to clear the ground for a future sociology of childhood to emerge have been made (e.g. Wallner & Pohler-Funke 1978; Jenks 1982).

The aim of my paper is to examine sociology from the point of view of finding at least some tentative explanations for this absence of children in sociology. In her article in the book just mentioned, Ambert suggests two explanations. The first one derives from the originally macrosociological perspectives with which sociology as a discipline emerged. Children (like women) were seen as peripheral to the global systems under study or simply taken as future replacements for their adult members. Little attention was therefore devoted to children, who were not accepted as significant members although they did occupy their 'proper place' (again like women) in the lives of the more significant ones (men). This insignificance is still granted to children, whereas feminist scholarship has been able to challenge many taken-for-granted assumptions on womanhood of malestream sociology and has contributed to changes within the discipline itself (see e.g. DuBois et al. 1985). - The second explanation suggested by Ambert derives in its turn from the present reward system within the profession: research on childhood and children simply is

53

no avenue for participating in the central discussions and debates within sociology, which discourages serious research in the area.

The acceptance of the marginal relevance of childhood research for sociology clearly stands in opposition to the bold assertion of Adler & Adler - that this in fact is not the case and that the issue of childhood is central to the discipline. Accepting this leaves one with the problem of why children still remain at the edge of sociological knowledge. Ambert's statement that the perspectives in use in sociological research account for this, points to the obvious task of looking critically at its theoretical and conceptual approaches.

A convenient starting point for such an examination is given in the observation that children and childhood are in fact approached only within some limited topics. The family, and later the school, are such topics. Families and schools are, of course, perfectly legitimate objects in studying children and childhood, for the simple reason that the modern, Western childhood is a familialized and scholarized one. It is in these contexts then that today's children largely live out their everyday lives. They are also the contexts that often prove troublesome and problematic for both children and their significant others, a fact which calls for further serious studies on the lives of children in families and schools.

In research, the encapsulation of children within such limited notions has another side to it as well. Ennew (1986), in a thoughtful analysis of one childhood problem, begins her examination of the sexual exploitation of children by stating the set of ideas that define the current notion of modern childhood. These ideas are clearly at work within the issue she examines, both in the public alarm and in the studies that have been conducted around it.

The first set of ideas (Ennew 1986: 20) separates children from adults and defines the ideal family as a nuclear unit consisting of protected children and protecting adults. The second set, in turn, separates adults from children within the production process. The child is denied the (formal) status of a worker, for the reason that being a child demands protection from work and provision of training.

Not only do these two sets of ideas structure the way we think about the sexual exploitation of children, they largely set the baseline for both the ways we think about children in general and the ways we go about organizing childhood for the younger members of our societies. Underlying this thinking seems to be a triangular configuration made up of assumptions concerning the nature of an essentially non- social childhood, the family as an appropriate context for this kind of childhood, and socialization (the more academic term for what childhood processes are about). This configuration presents the (Western) child not as (yet) part of her society, but condemned into a curiously non-social existence - 'a family life sentence' (Billy Bragg: 'The Home Front', in: Talking with the Taxman about Poetry) - described as a period of lack of responsibility, with rights to protection and training, but not to autonomy, a configuration that is irrelevant both culturally and structurally to the majority of the world's children.

The triangularity of childhood, the family and socialization proves to be as if moulded into one piece that cannot be broken into parts for separate consideration. This clearly affects the way that childhood issues are discussed and implications drawn. The questioning of one component, e.g. the family as the setting for childhood socialization, takes place within accepted and unproblematized notions

54

of the other two, thus in fact blocking the possibilities of even imagining novel relations between the three components. This is to be seen, for example, when discussing the impact of the recent changes in family forms and 'non-traditional' life styles on children: the new, family arrangements are evaluated against the accepted knowledge on childhood and socialization, i.e., knowledge that is based on observations on a particular version of 'the family'. It is little wonder that this often results in explicit or implicit moralism and pressure towards nor- mative family policies - naturally in the interests of providing the best for our children.

The very same triangular configuration appears within social science as well, with implications for research that in the end may prove to be responsible for the absence of children in sociology. The gaps in our sociological knowledge of children and childhood are then there not without reason. The observations that childhood issues come up in limited contexts and that the need for a sociology of childhood has only recently been asked for provide grounds for asking how in fact the child is constituted within social theory and what the implications are.

Jenks (1982) confirms that the child is indeed constituted within social theory: it is assembled to support and perpetuate certain versions of man [sic], action, order, language and rationality. This specific constituting of the child within social theory then defines for sociology the particular presence or absence of children on research agendas. The suggestion that there are basic conceptual limitations for constructing a sociology of childhood points to the need for a 'strong' programme for this endeavour. This means, in analogy to the 'strong' programme already effectuated in feminist research (see Bell & Roberts 1984), that more is needed than merely staying content with the task of observing the absences of child-relevant knowledge and starting to fill them, thereby adding an extra dimension to sociology. The 'strong' programme proposes the need for critical questions to be raised about the very perspectives in sociology on children and childhood that are built into its present concepts and categories. These should be looked at from the viewpoint of their tendency to produce systematical distortions and omissions concerning the place of the child in social life. This questioning, in turn, would necessarily promote the transforming of (some of) the perspectives, potentially leading to changes in the discipline itself.

I have written this rather lengthy introduction to my paper in order to convince the reader as well as myself of the benefits that can be derived for both sociological research and practical child politics from questioning the very concepts that underly the conventional approach in sociology to child-related issues. For my own use, I have translated this approach - 'growing up in the family' - into the conceptual triangle of socialization/the family/childhood. It is this triangle, and its implications for sociology, that to me seem to demand rethinking.

The paper first examines how the child is approached in sociology and how this approach is justified. The approach, conceptualized as 'socialization', is further analysed to find the particular positioning that it grants to children in society. The partiality and distortions in this clearly demand correction.

'Socialization' points instantly to the second suspect contributing to the neglect of children in sociology: the family. Feminist criticism of conventional notions of the family and the project of deconstructing the family prove to form a methodologically

55

helpful bridge from the negative critique of 'socialization' to a positive notion of childhood as a social construct.

Finally the paper takes up some of the beginnings for constructing a sociology of childhood. This demands the reformulation of the triangular configuration that has long been the obsession of sociological studies on childhood. Growing up in the modem family may then appear not so much 'a life sentence' as a life project of the child's own making, too.

1. Looking for the child in sociology Jenks, in an introduction to a collection of articles and excerpts from social science literature entitled The Sociology of Childhood (1982), reaches the generalization that the child is indeed constituted in social theory. If real live children rarely appear in social science research, at least the 'theoretical' child has a function within it. It is precisely this function that proves to have a powerful influence on how children are treated in social science research and sociology in particular.

Jenks describes the constitution of the child in social theory as revealing a continuous paradox: The child is familiar to us and yet strange, she inhabits our world and yet seems to answer to another, she is essentially of ourselves and yet appears to display a different order of being (1982:9). This ambiguity in the relationship between the child and the adult, appearing both in common sense and in theory, remains imprisoned within the notion of difference. But difference between the two positions indicates the identity of each: the child cannot be imagined except in relation to a conception of the adult, just as it becomes impossible to produce a sense of the adult and his/her society without first positing the child (Jenks 1982:10).

The significance of this formulation is in the fact that it contains an unac- knowledged and therefore unsolved paradox. A conception of the child is reached only by leaving the child side of the relationship 'empty', i.e. without any substantial positive definition.1 The child is envisioned merely as different and particular - compared to the adult - and this difference is accounted for by simply focusing on the process of overcoming it. And indeed social science has concentrated - to Jenks in an overattentive fashion - on theorizing 'the integration of the child into a social world' (the title of just one of many books, this one edited by Richards (1974), and more conscious of the dominant focus than most others).

As far as social theory is concerned, then, the child remains negatively defined - defined only by what the child is not but is subsequently going to be, and not by what the child presently is. The child is depicted as pre-social, potentially social, in the process of becoming social - essentially undergoing socialization.

The concept of socialization supplies sociology precisely with this perspective which it shares with social science in general. 'Socialization' has indeed become the conventional categorical tool for approaching child-related issues, whether conceming the family, parenthood or the experience of being a child. This is evident simply by looking at sociology textbooks which - if indeed they have a chapter on children - usually do this under the rubric of 'socialization' (Ambert 1986:16; Thome 1987).

'Socialization' possesses a compelling insight, which for example Thorne (1987)

56

notes but doesn't remain content with: born without language or knowledge of social organization, children do become inducted into the social worlds that surround them. But for Thorne as well as others (e.g. Speier 1976) the socialization framework also embeds problems for child research: it emphatically brings out 'the adult ideological viewpoint' as well as functionalist and determinist assumptions. Its limitations help to distort the sociological understanding concerming the dynamics of social structure, human agency and historical change (Reyer 1980; Thorne 1987), making it hard even to imagine, and even harder to conceptualize children as veritable social actors.

2. Socialization: the career of a concept Both the term and the concept of socialization originally emerged as a corollary of the concept of society, in an age when arguments for a separate scientific discipline of sociology were being advanced. In this context, 'socialization' referred to social forces drawing individuals together into a community. It was understood to make social life possible and was to be considered a reality in its own right, or in the words of Durkheim (who named the notion), a reality sui generis, to be treated as a social fact in clear distinction of its effects on individuals.

The emergence of 'socialization' was then the emergence of a genuinely socio- logical category, which furthermore was included within the vocabulary of a realist sociology (see, e.g., Johnson, Dandeker & Ashworth 1984). This sociology - contrary to empiricist and subjectivist sociologies which were later to dominate the field - insists on social phenomena being things-in-themselves, outside and independent of all those elements that it comprises, such as individuals, their particular circumstances and consciousness. The social reality of socialization was to be seen in its constraining effect on individuals, this appearing as the inter- nalization by them of social facts originally external and independent of them. Socialization, then, was taken as an exclusively social process, which nevertheless had secondary effects through internalization (a psychological process).

The original and developing concept of socialization contained within itself a strong moral sense: the society in which the forces of socialization are in effect is a civilized one, i.e. one in which the originally unsocial human nature is gradually superseded. The civilized society, of course, then proves to be the society of adults. - The original concept of socialization also contained an assumption of the fundamental polarity of human nature between an uncivilized, unsocialized side and a civilized and socialized side. Here the child was 'naturally' located on the uncivilized side, as a being yet to be socialized in the power field of social forces.

It is therefore the idea of a reality residing on the level of the social and exercising a determinate influence on individuals that maintains the original link of the issue of socialization to sociological analysis - an idea that later was named sociological determinism. However, sociological determinism within a realist sociology is dif- ferent from sociological determinism transplanted to other sociologies. The shift in the meaning of socialization during its career exemplifies this. The term clearly emerged as an entirely different concept, as in the course of time it became included in the more individualistic and increasingly empiricist research in the United States (for an interpretation of the fate of 'socialization' in the US, see Wentworth

57

1980). The term was given a new interpretation that made use of the sociological determinist view on the empirical world, but above all it reinterpreted socialization as referring to the psychological process of internalization. The social reality that Durkheim spoke of came to be reduced to externally given conditions or milieu for the actual drama of socialization-as-internalization. Besides the current research problems in early American sociology two other sources of influence for this reduction were Freud and his psychoanalytical thinking, and the anthropology of the Culture and Personality school, including the childhood studies by Margaret Mead (for accounts on these influences see Clausen 1968 and Wentworth 1980). The reduction of socialization to internalization left to research the task of accounting for the process whereby society, the social environment and the cultural system in which children are born and grow up are taken as external givens and contents to be internalized - and thereby reproduced - by them.

This shift in the fundamental meaning of the sociological category of socialization is, of course, paradoxical for sociology itself, as it removed the whole concept out of sociology into the domain of psychology. Nevertheless, it was this very transformation that from the 1950s onwards was the base for the emergence of the tradition of specific studies on socialization - a development that was given theor- etical foundation by Talcott Parsons who inscribed the notion of socialization-as- internalization into his grand theory.

3. The 'socialization' perspective to childhood The feature perhaps most easily detected in the present-day notion of socialization has been named by Speier (1976, see also Thorne 1987) 'the adult ideological viewpoint', stressing the exclusive power of adults in defining children. This view- point, within the 'socialization' framework, can be seen to apply to various agents, depending on the particular context of socialization. However, adulthood need not be their common denominator. More significant is the fact that they share a structural position of power in relation to those undergoing socialization (here: children). There are grounds then for renaming this viewpoint an elitist perspective. The process of socialization is essentially looked at from the viewpoint of society's institutions and organizational apparatuses and their representatives, these assumed to have systemic constraints or vested interests for reproducing themselves and therefore exercise influence on children in order to produce in them such outcomes that are conducive to that reproduction. It is easy here to see that the other side of the elitism coin is functionalism.

As a consequence of the viewpoint's inherent elitism and functionalism the intentions and interests of children as participants in their own socialization are effectively excluded, presumably on the assumption that they more or less converge with those of the elites. This, of course, helps to model children as passive objects and victims of influences external to them, unable and unwilling to resist. The outcomes of the socialization process - following the elitist viewpoint - can therefore be accounted for merely by referring to constraints in children's environments. A case of sociological determinism is given here.

A further feature in the socialization perspective to childhood, again following its inherent elitism, is the firm and exclusive commitment to the outcomes of the

58

process at the expense of the process. But of course not all possible outcomes are given attention. Only those that are expected, intended or desired by the agents that the perspective depicts as instigators of socialization processes are observed. This outcome-centredness futher maintains two limitations within the frameworks. First, despite frequent mentions of concern for the process, it is in fact remarkably neglected in actual research. This is evident by looking at the methods most often applied: techniques of correlating environmental conditions and outcomes ('inputs'/ 'outputs') dominate, leaving the process inside a black box.

Second, the exclusive attention paid in research to outcomes expected, intended or desired by elites necessarily brings about a systematical neglect of any other outcomes. This not only avoids the possibility that children as participants might contribute something themselves - the case is dismissed by elitism at the beginning - it also works against any conflicts or contradictions being noticed in the process or outcomes that may even partly refute those expected by elites. Functionalist presumpYtions are in effect here again, giving rise to an overly harmonious view of socialization.

The most serious simplification and indeed omission in the socialization per- spective concerns the process itself. Following from what has been said it is clear that what is commonly taken to be the whole process of socialization, is at most one part of it, namely the part that is best conceived of in psychological terms and theories. Socialization-as-internalization continues to monopolize our under- standing of socialization. The ample reservoir of psychological theories available to account for it renders it even more difficult for students of socialization to take notice of this basic deficiency. The original sociological notion of socialization as a social process is lost, and only recently this fact has been taken up and arguments for a sociology of socialization raised. A correction at this principal point in the socialization perspective will help to correct its other distortions and simplifications, too, granting even children the status of social actors, albeit in unequal social relations.

4. Making 'socialization' social The assumptions of determinism, linearity, functionality and harmony in the process of socialization, its reduction to internalization and its outcome-centredness have all been observed and criticized within the research tradition, but mostly individually and in piecemeal fashion (see, e.g. Long & Hadden 1985). Thus the assumption of functionality in socialization has proved to be, if not entirely false, then at most only a partial truth. More precisely conducted empirical studies have shown that socialization does cover other phenomena besides those that are regarded as conducive to the reproduction of society's institutions and structures. These include contradictory effects and results that cancel each other out.

It seems that the limitation easiest to overcome in the socialization perspective is its unreflected commitment to the viewpoint of elites. Research methods do exist that give voice to groups that have normally been silent in research (particularly ethnography). By using these methods at least more 'democratic' and complete descriptions on the experience of socialization may come out (see e.g. Willis 1978). Such methods may also help to produce evidence on the remarkable competence

59

that children - even small children - have in constructing their everyday social relations (see e.g. Goode 1986; Waksler 1986). This in turn may contribute to children being seen in a new light, not just as objects primarily being acted on (even if this is largely true for children most of the time), but also as social actors in their own right.

Ethnographic, 'phenomenological' or cultural analyses on the life of the child will furthermore help to bring about a more complex notion of socialization in admitting that the assumption of linearity - that is, the assumption of socialization resulting from its given preconditions - must fail. Instead, a certain degree of autonomy will have to be granted to the processes involved, pointing to the need to reconsider these as social processes.

This, it is admitted, implies in turn that children be viewed as social actors and agents of their own lives. It is methodologically wrong then to victimize children in research, no matter how much they appear as victims in their various real life situations. This view then leads to a reconsideration of children and childhood also in the context of on-going theoretical debates about the relationship between social action and social structure (see e.g. Burns 1986 and Hindess 1986). Childhood will therefore be moved from its conventional micro-world of family, school and significant others, and placed within the terrain of central theoretical concerns in sociology as well. This implies the consideration of childhood in structural terms (see, here, Qvortrup 1985).

So, children seen as social actors has far-reaching implications not the least for the socialization perspective itself. In fact this view risks ruining the whole conventional idea of socialization, as it forces a reconsideration of its most prob- lematical point - the understanding of socialization as a process.

This point has recently been addressed in the scientific literature (e.g. Wentworth 1980; Long & Hadden 1985). Geulen (1987) approaches the very problem when suggesting that socialization research and developmental psychology, while sharing a concern for analysing and explaining human development, display opposing strengths and weaknesses. The socialization approach, compared to the alternative approach to human development, is increasingly being recognized as having a deficient understanding of the process aspect, whereas developmental psychologies in particular conceptualize this aspect in various ways. For some researchers in the socialization tradition a reliance on these more elaborated psychological process concepts (e.g. of Piaget or Erikson) has appeared as the solution to their problem. Geulen however demonstrates the inadequacy of this. There are logical limitations for such an integration of the two perspectives, and the resulting eclecticism cannot be accepted. The problem of how to conceptualize socialization as a process therefore remains.

The difficulty obviously lies in the recognition that socialization is in fact a social process, as already implied earlier in this paper (and originally established by Durkheim). The problem is a theoretical one and resists piecemeal solutions. A return to Durkheim's original notion of socialization will hardly provide a sociologically satisfying answer, although his work may be inspiring in this project (see here e.g. Bhaskar 1979:117-121). The criticism levelled against the way of treating children in the socialization perspective, however, suggests avenues for rethinking that have been progressing in other areas of social science. As the

60

conventional notion of socialization has tended to objectify those undergoing socialization, approaches that stress agency seem more promising. And as 'soci- alization' furthermore has tended to centre on outcomes of an individual (or psychological) nature, approaches stressing the socially constructed nature of social life and conditions point to social relations, structures of social interaction, insti- tutionalized social practices and shared (but also contested) meanings as outcomes of socialization as well. These being integrated into a new understanding of soci- alization implies that socialization might be reconceptualized as construction instead of internalization (whichever its particular psychological interpretation).

5. Deconstructing the family A 'constructionist' understanding of social phenomena has been advanced in family research - an area of interest for deconstructing the configuration of childhood, socialization and the family. Feminist inspired family research in particular has made us aware of the insight that social conditions should be approached from the perspective of their being constructions. The social condition primarily dealt with by feminist research is of course the women's (unequal) condition, and the family is the focus on the assumption that it is the family that forms an important, perhaps even decisive, part of the women's condition.

Family sociology has been slow, if not altogether impotent, in finding answers to the question of gender inequality. The extremely atheoretical nature of most family sociology, deriving from its patent empiricism, is one expfanation. In it a unitary notion of 'the family' has been taken for granted and made the basis for family research and theory. Bernardes (1985) for instance depicts the situation in family sociology by stating that traditional family sociology has adopted an 'ideal type' model of a presumed 'nuclear family' or 'typical family'. One reason for the prevalence of the unitary view of 'the family' is the frequent reduction of the family to a natural or biological unit (see Coward 1983, Barrett & McIntosh 1982) commonly depicted within functionalist frames. Another explanation lies in the fact that the idea of 'the family' as a unitary institution is part of the arsenal of the theory of industrial society (see Harris, 1983, Close 1985). This seems to have become in its simplified versions a kind of common-sense sociology and has as such effectively maintained the idea of 'the family' as an essence successful in taking various appearances in various social environments.

However, since the 1960s 'the family' has not been without criticism. First the radical critique of society's institutions and the concomitant search for alternatives applied, of course, to this particular institution as well. Within family research itself an empirical critique came for example from family history, comparative family research and descriptive studies on everyday realities of family living (see e.g. Anderson 1979, Rapoport et al. 1982). The variety of 'family forms' or 'family types' that were found to exist could only with great hesitation be subsumed under some general concept of 'the family'. Here Barrett & McIntosh (1982:83) note that as each writer demonstrates an experiential certainty of the object under study, the reader is left with the impression that they are simply not talking about the same thing. Family researchers then agree that there are groups of people commonly

61

identified as 'families'. But the theoretical problem of how these empirical families relate to 'the family' has remained open. Consequently, an implicit adherence to an essential understanding of 'the family' is retained, although expressions like 'the British family' are explicitly abandoned (Laslett, in Rapoport et al. 1982).

Contrary to this, the critique of 'the family' coming from the emerging feminist studies could not remain theoretically content with the unitary view. One of the channels by which a questioning of 'the family' came forth was the observation of the diverse meanings, experiences and consequences that living in families had even for members of the 'same' family - men, women, adults, children, sons, daughters. The differentiation of these were clearly distorted or denied by various con- ceptualizations that recur in literature on the family (Thorne 1982). The observation of diversity within families made it impossible to entertain the solid fact of an essential family. This image had to be more of an imaginary thing, a constructed (not a natural or inevitable) unity of a multitude of meanings behind the term 'family'. These as well as their justifications needed to be discovered and related to the actual life of real families. Family sociology then was not so much of the family but of families, including a concern for 'the family' as an ideological and cultural construct.

The search for meanings for 'the family', their interrelations and the etymologies that followed, brought a realization that the term 'family' covers a wide range of material arrangements, sets of relationships and ideologies. These include at least household and residence arrangements, kinship systems, sexual relationships, ideol- ogies of femininity and motherhood and of the private domain (see e.g. Flandrin 1979, Harris 1981, Coward 1983). Thus the proper object for a sociology of families was to be found 'behind' the taken-for-granted unity of 'the family', and if unity there be, it had to be a problem to explain, not a fact to assume.

New questions were opportune for family theory now. The dissolution of the imaginary unity of 'the family' questioned the theoretical object of family sociology itself. If family phenomena and 'the family' are part of the empirical surface, and if behind or underlying these there is a multitude of material and ideological structures and processes, then to theorize family is to theorize these elementary structures and processes, their inter-relationships and the ways in which family- identified empirical phenomena are constructed within them.

These questions were first tackled by Juliet Mitchell in her pioneering article from 1966 (see Mitchell 1971) depicting the family unit as a distinctly social construct. To her it is the ideology of the family that manages to represent the family as a (quasinatural) unit as well as gives it an atemporal quality. This has clearly escaped the attention of past family theory (for reasons that have been analysed by Coward (1983), for example). Instead, Mitchell goes on to dissolve the false unity of 'the family' by what can be conveniently termed destructuration, that is, she decomposes the family into the three elementary structures of reproduction, sexuality and socialization.

Mitchell's way of dissolving the family has its weak points (see e.g. Vogel 1983:13- 17) but its theoretical merit endures: she puts forward a radically different task for family research to solve when denying the category 'family' any explicit theoretical presence and locating the theoretical object of family research on a reality that substantially underlies and effectively constitutes observable empirical phenomena.

62

Mitchell's as well as many others' (see e.g. Rubin 1975 and Rapp 1982) project means a clear break from the patent empiricism of mainstream family sociology.

Developments within sociological theory, with the significance of both Marxist and feminist research, have been conducive to further theoretical work in the line originally opened by Mitchell. Theoretical critiques of structuralism, genealogical studies inspired by Michel Foucault, debates on theories of ideology and culture have all made an impact in many recent works towards renewing family theory. In broad terms these works may be depicted as attempts at deconstruction of the family (compare with Mitchell's destructuration). Here. deconstruction means establishing both the false unity of the family (the concept) and the fully social, constructed nature of family structures, values and processes (see Barrett & McIntosh 1982, Barrett 1985, Stacey 1984). The specific historical, political, cultural and class variation of empirical family phenomena is to be demonstrated and traced back to underlying processes, conditions and relations between social forces and actors that effectively have functioned to construct and maintain present structures like kinship systems, residence and household arrangements, gender relationships, particular ideologies, etc.

The deconstructionist move within social science stresses the need for rigorous historical studies on the often not entirely peaceful developments and struggles between various and mostly unequally positioned social forces that make the construction sites of the structures underlying observable family phenomena. His- tories of family-relevant constructs are still rare, but historically minded studies have been conducted on the social construction of motherhood, womanhood, sexuality, gendered subjectivity, skilled (and unskilled, i.e. women's and children's) work, the modern suburban family, etc. (Dally 1982; Margolis 1984; Lewis 1980; Weeks 1981; Foucault 1979; Hollway 1984; Urwin 1985; Hartman 1979; Coyle 1982; Cockburn 1983; Game & Pringle 1983; Riley 1983; Ng 1986). Subsequent research for example on the role of the (welfare) state in constructing domestic life (Cass 1983), normal, i.e., economically viable family units (Brophy & Smart 1982; Langan 1985; Ursel 1986) or the relations between children and their parents (Freeman 1984; Dingwall et al. 1984) in turn builds upon the same insights.

In this company the modem prolonged, familial childhood appears as a social construct as well. That modem childhood has been constructed and undergoes further reconstruction by the (welfare) state for one is recognized (e.g. Leira 1986). The history of childhood literature, however, gives a longer view on the construction of modern childhood.

6. Childhood as a construct Philippe Aries's work (1962) on the discovery of childhood seems to tell the story of how the idea of childhood first emerged, coupled with a new image of family intimacy, and how it was subsequently put into practice within schemes of child care and education. The history of childhood appears - in this interpretation of Aries's intention - as a piece of the history of ideas, conducive to the modern sense of childhood that we now know.

A more thorough analysis of Aries's historical method and his mostly implicit

63

social theory, however, helps to produce another view for thinking about childhood. In this reading (see e.g. Muller 1979) childhood emerges, not as an idea of the child in the first place, but as a particular social status within specially constituted institutional frames. These frames were a practical accomplishment of the bourgeois class, seeking by political means to secure its own reproduction in structurally changing circumstances. 'The child' emerged in this context as a social and practical construct to be realized for the younger members of that particular class. It is for this project that various schemes of child care and education were developed, leading to the formation of extraordinary social worlds for their inhabitants - notably 'the intimate family' and the school - and consequently to the formation of a particular 'habitus' as well. This product - modem childhood - was later made available for other social classes (Donzelot 1980), although not without a struggle (which Donzelot chooses largely to ignore in his analysis of the French experience). This practical accomplishment subsequently made the ground for new formations as well, from the emergence of 'motherhood' and other child-related professions and experts to the institutions of knowledge production about children, based on observing the newly emerged social worlds of childhood and perceiving their inhabitants accurately as 'children', different from adults.

It is easy to detect here the closeness of Aries's thinking to particular structuralist schools in French sociology (Foucault, Bourdieu). However, the possible con- tribution to a sociology of childhood, of this kind of reading of the history of childhood, does not depend on one's judgment on these schools. It does not even depend on the credibility of Aries's results.2 The important point is, to my mind, the insight that childhood need not be reduced to an idea or a mentality that needed first to conquer people's minds in order to be instituted for the younger members of society. On the contrary, childhood was - and in its modem versions is - the ever-constituted result of decisions and actions of particular historical social actors, in their economical, political and cultural struggles that potentially concern the whole spectrum of their interests. To account for childhood then calls for analyses of these broad social processes that in their interaction come to constitute - rather than deliberately aim to constitute - social practices that define childhood. - The implications of this approach to childhood for a sociology of childhood are obvious.

7. Conclusion

Sociologists need not - and should not - content themselves with the notion that children belong primarily to the province of psychology, education and pediatrics. The relevance of psychological, medical or educational knowledge is not under dispute even when asserting that there are crucial limits to that knowledge set by the particular versions of childhood that we have been observing and now may observe. If it can be accepted, as I suggest, based on recent developments in sociological (and historical) research, that childhood is under continuous recon- struction within a large field of societal constraints and confronting social forces, the implications should be clear. One of the most obvious of them is that research on children and childhood needs to be conducted in a genuinely interdisciplinary fashion.

My main argument has been for a sociology of childhood, as this area is very

64

much absent in social science. Reasons for this have been traced and perspectives for advance have been seen in many areas, calling for critical examination of sociological theories and categories and for empirical research informed by this critique.

The latter would need to avoid the conventional victimization of children in social science researchi, granting them instead the status of participants and constructors in the very processes that make their - and our - world. This derives from approaching social life as a dynamic field of confrontations and struggles between social forces. The approach then allows for viewing even children as a structural 'class' in relation to other classes and capable of collective action and therefore capable of engaging in social struggles. There may be objections for treating children as a social class, some of them stressing the social barriers for this kind of class formation. Nevertheless, even if children might not appear as a social class in its own right, childhood remains a political issue that is enmeshed in on-going everyday struggles. Research that helps to uncover these struggles in reality grants children agency and thereby aligns with children. A sociology of childhood, and for children, if not (yet) by children themselves, hopefully emerges from these alignments.

Notes The article is based on my paper given at an interdisciplinary conference on the life and development of children: 'Growing into a Modern World' in Trondheim (Norway), June 10- 13, 1987. 1 Contrary to this, the image of the strange, indeed 'enigmatic' child is a recurrent theme in

both verbal and non-verbal arts (see Kuhn 1982). 2 Wilson (1980), for one, is highly critical of Aries's work from a historian's point of view,

seeing many weaknesses in the ways Aries generates evidence to support his argument.

References Adler, Patricia A. & Adler, Peter 1986. Introduction. In P. A. Adler & P. Adler (eds.),

Sociological Studies of Child Development, vol. 1: Greenwich, Connecticut: JAI Press. 3- 10.

Ambert, Anne-Marie 1986. Sociology of Sociology: The Place of Children in North American Sociology. In P. A. Adler & P. Adler (eds.), Sociological Studies of Child Development, vol. 1: 11-31. Greenwich, Connecticut: JAI Press.

Anderson, Michael 1979. The Relevance of Family History. In C. Harris (ed.), The Sociology of the Family. Sociological Review Monograph 28. University of Keele, Staffordshire.

Aries, Philippe 1962. Centuries of Childhood. A Social History of Family Life. London: Jonathan Cape.

Barrett, Michele 1985. Introduction. In Friedrich Engels. The Origin of the Family, Private Property and the State. Harmondsworth: Penguin.

Barrett, Michele & McIntosh, Mary 1982. The Anti-Social Family. London: Verso. Bell, C. & Roberts, H. (eds.) 1984. Social Researching: Politics, Problems, Practice. London

& Boston: Routledge & Kegan Paul. Bernardes, Jon 1985. Do We Really Know What 'The Family' Is? In Paul Close & Rosemary

Collins (eds.), Family and Economy in Modern Society. London: The Macmillan Press. Bhaskar, Roy 1979. On the Possibility of Social Scientific Knowledge and the Limits of

Naturalism. In John Mepham & Ruben David-Hillel (eds.), Issues in Marxist Philosophy, Vol. III: Epistemology, Science, Ideology. Brighton: The Harvester Press.

Brophy, J. & Smart, C. 1981. From Disregard to Disrepute: The Position of Women in Family Law. Feminist Review, No. 9.

65

Burns, Tom R. 1986. Actors, Transactions and Social Structure. In Ulf Himmelstrand (ed.), Sociology: From Crisis to Science? vol. 2. London: Sage Publications.

Cass, Bettina 1983. Population Policies and State Policies: State Construction of Domestic Life. In Cora V. Baldock & Bettina Cass (eds.), Women, Social Welfare and the State. Sydney: George Allen & Unwin.

Clausen, J. A. 1968. Introduction. In J. A. Clausen (ed.), Socialization and Society. Boston: Little, Brown & Co.

Close, Paul 1985. Family Form and Economic Production. In Paul Close & Rosemary Collins (eds.), Family and Economy in Modem Society. London: The Macmillan Press.

Cockburn, Cynthia 1983. Brothers. Male Dominance and Technological Change.. London: Pluto Press.

Coward, Rosalin 1983. Patriarchal Precedents. Sexuality and Social Relations. London: Rout- ledge & Kegan Paul.

Coyle, Ann 1982. Sex and Skill in the Organization of the Clothing Industry. In J. West (ed.), Work, Women and the Labour Market. London: Routledge & Kegan Paul.

Daily, Ann 1982. Inventing Motherhood. The Consequences of an Ideal. London: Burnett Books.

Dingwall, Robert, Eekelaar, John & Murray, Topsy 1983. The Protection of Children. State Intervention and Family Life. London: Basil Blackwell.

Donzelot, Jacques 1980. The Policing of Families. London: Hutchinson. DuBois, E. C., Kelly, G. P., Kennedy, E. L., Korsmeyer, C. W. & Robinson, L. S. 1985.

Feminist Scholarship. Kindling in the Groves of Academe. Urbana and Chicago: University of Illinois Press.

Ennew, Judith 1986. The Sexual Exploitation of Children. London: Polity Press. Flandrin, Jean-Louis 1979. Families in Former Times. Cambridge: Cambridge University

Press. Foucault, Michel 1979. The History of Sexuality, vol. 1, An Introduction. London: Allen Lane. Freeman, Michael D. A. (ed.) 1984. State, Law and the Family. Critical Perspectives. London:

Tavistock Publications. Game, Ann & Pringle, Rosemary 1983. The Making of the Australian Family. In Ailsa

Burns, Gill Bottomley & Penny Jools (eds.), The Family in the Modern World: Australian Perspectives. Sydney: George Allen & Unwin.

Geulen, Dieter 1987. Zur Integration von entwicklungspsychologischer Theorie und empirischer Sozialisationsforschung. Zeitschrift far Sozialisationsforschung und Erzie- hungssoziologie 7, (1), 2-25.

Goode, David A. 1986. Kids, Culture and Innocents. Human Studies 9, (1), 83-106. Harris, C. C. 1983. The Family and Industrial Society. London: George Allen & Unwin. Hartmann, Heidi 1979. Capitalism, Patriarchy and Job Segregation by Sex. In Zillah

Eisenstein (ed.). Capitalist Patriarchy and the Case for Socialist Feminism. New York: Monthly Review Press.

Hindess, Barry 1986. Actors and Social Relations. In Mark L. Wardell, Stephen P. Turner (eds.), Sociological Theory in Transition. London: Allen & Unwin.

Hollway, Wendy 1985. Gender Difference and the Production of Subjectivity. In Julian Henriques et al. Changing the Subject. London: Methuen.

Jenks, Chris 1982. T7he Sociology of Childhood. Essential Readings. London: Batsford Aca- demiic and Educational Ltd.

Johnson, Terry, Dandeker, Christoper & Clive Ashworth 1984. The Structure of Social Theory. London: Macmillan.

Kuhn, Reinhard 1982. Corruption in Paradise. The Child in Western Literature. Hanover and London: University Press of New England.

Langan, Mary 1985. The Unitary Approach: A Feminist Critique. In Eve Brook & Ann Davis (eds.), Women, the Family and Social Work. London: Tavistock.

Leira, Arnlaug 1986. Barna, familien of velferdsstaten. Nytt om kvinneforskning 10 (2), 3-7.

Long, Theodore E. & Hadden, Jeffrey K. 1985. A Reconception of Socialization. Sociological Theory 1985, 3 (1), 39-49.

Lewis, Jane 1980. The Politics of Motherhood. London: Croom Helm.

66

Margolis, Maxine L. 1984. Mothers and Such. Views of American Women and Why They Changed. Berkeley and Los Angeles: University of Califomia Press.

Mitchell, Juliet 1971. Woman's Estate. Harmondsworth, Middlesex: Penguin. Miller, Sebastian F. 1979. Wie ist die 'Geschichte der Kindheit' zu lesen? Literatur Rundschau

(Sonderheft der Neuen Praxis), 2. Jg., Heft 2, 19-32. Ng, Roxana 1986. The Social Construction of Immigrant Women in Canada. In Roberta

Hamilton & Michele Barrett (eds.), The Politics of Diversity. London: Verso. Qvortrup, Jens 1985. Placing Children in the Division of Labour. In Paul Close & Rosemary

Collins (eds.), Family and Economy in Modern Society. London: Macmillan. Rapoport, R. N., Fogarty, M. P. & Rapoport, R. (eds.) 1982. Families in Britain. London:

Routledge & Kegan Paul. Rapp, Rayna 1982. Family and Class in Contemporary America: Notes Towards an Under-

standing of Ideology. In B. Thorne & M. Yalom (eds.), Rethinking the Family. Some Feminist Questions. New York: Longman.

Reyer, Jiurgen 1980. Sozialgeschichte der Erziehung als historische Sozialisationsforschung? Zeitschrift fiur Pddagogik, Jg. 26, Heft 1, 51-72.

Riley, Denise 1983. War in the Nursery. Theories of the Child and the Mother. London: Virago.

Richards, Martin M. P. (ed.) 1974. The Integration of the Child into a Social World. London: Cambridge University Press.

Rubin, Gayle 1975. The Traffic in Women: Notes on the 'Political Economy' of Sex. In R. R. Reiter (ed.), Toward an Anthropology of Women. New York: Monthly Review Press.

Speier, Matthew 1976. The Adult Ideological Viewpoint in Studies of Childhood. In Arlene Skolnick (ed.): Rethinking Childhood. Perspectives on Development and Society. Boston - Toronto: Little, Brown & Co.

Stacey, Judith 1984. Should the Family Perish? Socialist Review, No. 74. Thorne, Barrie 1982. Feminist Rethinking of the Family: An Overview. In Barrie Thorne

with Marilyn Yalom (eds.), Rethinking the Family. Some Feminist Questions. New York: Longman.

Thorne, Barrie 1987. Re-visioning Women and Social Change: Where Are the Children? Gender & Society I (1). 85-109.

Thorne, Barrie with Marilyn Yalom (eds.) 1982 Rethinking the Family. Some Feminist Questions. New York: Longman.

Ursel, Jane 1986. The State and the Maintenance of Patriarchy: A Case Study of Family, Labour and Welfare Legislation in Canada. In James Dickinson & Bob Russell (eds.), Family, Economy and State. The Social Reproduction Process Under Capitalism. London: Croom Helm.

Urwin, Cathy 1985. Constructing Motherhood: The Persuasion of Normal Development. In Carol Steedman et al. (eds.), Language, Gender and Childhood. London: Routledge & Kegan Paul.

Vogel, Lise 1983. Marxism and the Oppression of Women. Toward a Unitary Theory. London: Pluto Press.

Waksler, F. C. 1986. Studying Children: Phenomenological Insights. Human Studies 8 (1), 71-82.

Wallner, Ernst M. & Pohler-Funke, Margret 1978. Soziologie der Kindheit. Heidelberg: Quelle & Meyer.

Weeks, Jeffrey 1981. Sex, Politics and Society. Harlow, Essex: Longman. Wentworth, William M. 1980. Context and Understanding. An Inquiry into Socialization

Theory. New York: Elsevier. Willis, Paul 1978. Learning to Labour. Farnborough: Saxon Hause. Wilson, Adrian 1980. The Infancy of the History of Childhood: An Appraisal of Philippe

Aries. History and Theory. Studies in the Philosophy of History, vol. XIX, no. 2, p. 132- 153.

67