Norah Carlin Archive   |   ETOL Main Page


Norah Carlin

Is the family part of the superstructure?

(Spring 1985)


From International Socialism 2 : 26, Spring 1985, pp. 113–117.
Transcribed by Christian Høgsbjerg.
Marked up by Einde O’ Callaghan for the Encyclopaedia of Trotskyism On-Line (ETOL).


While welcoming Chris Harman’s article, Women’s Liberation and Revolutionary Socialism (International Socialism 2 : 23, I felt that some of his arguments, though essentially correct, were based on confused reasoning and untenable assertions. I hope that by pointing these out and suggesting some better arguments I will be contributing support for our case that the struggles for socialism and women’s liberation are not separate but one.

The arguments I am concerned about hinge on Chris’s assertion that:

the logical thing is to see the family as part of the superstructure – something created by the needs of accumulation at a certain point in capitalist development, which capitalism now begins to undermine, but which it is prevented from abolishing because of its own crisis-prone nature. (p. 19)

The family is not, he says, ‘essential to capitalism in the same way as exploitation and accumulation.’ Yet he is quite clear that ‘there can be no end to women’s oppression under capitalism’ – indeed, this is one of his section headings, and he includes the impossibility of actually abolishing the family within capitalism under that heading.

Confused? I was at first, and I believe that the confusion is partly the result of Chris’s introducing the much misunderstood distinction between ‘base’ and ‘superstructure’ into his argument. But it is also the result of a common Marxist tendency in writings on the family, to leap straight from ‘the Origins’ to the twentieth century (or perhaps the nineteenth-century industrial revolution) without any consideration of the history of the family in class society in between. My point is that Chris’s case would be more coherent and persuasive without these impediments.

The article argues quite correctly against the Two Modes theory of Joan Smith and others – the idea that there is a ‘mode of reproduction’ in each society which is distinct and separate from the ‘mode of production’. Human arrangements for the reproduction of the species and for the day-to-day survival of its members (feeding, clothing and sheltering them) are always an essential part of the mode of production in a society, and have varied throughout the history of class society according to the level of the productive forces and the balance of relations between producing and exploiting classes. For example, in the slave societies of ancient Greece and Rome the slaves had no families: they were not allowed to marry or to claim ‘their own’ children, who belonged instead to their mother’s master. But in the feudal societies which followed, the serfs were tied to the soil as families, legally bound to reproduce the labour force of the manor as well as to labour for the lord themselves. Where feudal lords were relatively weak (mountain areas such as the Pyrenees), the peasants lived in clannish extended families, but where the lords were strong (in most of North-Western Europe) the nuclear peasant family was the norm. Many more examples could be given of the relationship between the family and the productive arrangements in different societies: for example, the multiplication of proletarian nuclear family households producing commodities for the market in the Domestic System of early capitalism; or the recreation of the family among the working class in the cotton towns of Lancashire in the nineteenth century where it had briefly seemed to be on the verge of disappearance.

Taking such historical developments into consideration, there are some difficulties about defining the family as ‘part of the superstructure’. In some societies – feudal and early capitalist – production actually takes place within the family unit. In another – industrial capitalism – there is an almost total divorce between the family and the workplace. In yet others, the producers may work in the households of other families (the slave workshops of Greece and Rome); or again they may be totally or partially deprived of family relations as workers on huge estates (parts of the Roman world and the American South). This makes it hard to say in general whether or not the family is part of the relations of production: it would seem that sometimes it is, but not always.

Even to see the problem in this way, however, would be to concede half the case to the Two Modes argument, for it is to suggest that production and reproduction are different kinds of things, one of which – production – is more ‘basic’ than the other.

I suspect that what underlies Chris Harman’s attitude is that he is unconsciously reproducing a fundamental tenet of most non-Marxist schools of sociology: that the ‘social’ is a separate realm from the ‘economic’. It is in this sense that many Marxists seem to have interpreted the distinction made by Marx and Engels between base and superstructure. Some of the best known quotations from Engels on this subject do indeed use the words ‘economic’ and ‘basis’ in conjunction, but it should be clear from the context and from other writings of Engels around the same time that he had no narrow conception of the ‘economic sphere’. For example, he did write:

The economic situation is the basis, but the various elements of the superstructure ... also exercise their influence on the course of historical struggle ... [To J. Bloch, 21–2 September 1890]

It is not that the economic situation is cause, solely active, while everything else is only passive effect. There is, rather, interaction on the basis of economic necessity, which ultimately always asserts itself. [To W. Borgius, 25 January 1894]

But in successive introductions to late editions of the Communist Manifesto Engels repeatedly defined the fundamental proposition of Marxism in the following terms:

That economic production and the structure of society of every historical epoch arising therefrom constitute the foundation for the political and intellectual history of that epoch. [Communist Manifesto, English edition, 1880, German edition 1883 (My emphasis)]

With regard to the subject under discussion in Harman’s article, the first letter of Engels quoted above explicitly says:

According to the materialist conception of history, the ultimately determining element in history is the production and reproduction of real life. [My emphasis]

By both these definitions, it seems to me, the family would be part of the base and not the superstructure.

Engels was always an opponent of what he called the vulgarisation of historical materialism, and some of Chris Harman’s formulations in the article in question were undoubtedly vulgar. For instance, he attempted to argue that the dynamic tension between forces of production and relations of production which shapes the course of human history is absent from the sphere of reproduction because ‘human beings are not constantly finding new ways of reproducing themselves (cloning in one epoch, laying eggs in another, live births in a third) ...’(p. 16) He does not go on to say ‘test tube babies in a fourth!’, presumably because he realises that history might prove him wrong!

But the argument is not just facetious, it is silly. Human reproduction in the narrow sense of procreation is not immune to technological change. Has Chris never used a condom, or slept with a woman on the Pill? Or does he think that the difference between Karl Marx’s experience of family life and his own is purely ideological?

Nor do technological changes in contraception lack that dynamic tension in relation to production that we see elsewhere. The twentieth-century spread of the condom, followed by the more reliable Pill, among the working class has been both a cause and a result of the increasing participation of women in the labour force. I should have thought that such changes precisely fit Chris Harman’s definition of the dynamic: ‘Every increase in the ability of human beings to control nature produces new interrelations between human beings themselves, and therefore begins to transform the pre-existing relations of production.’ (p. 16)

But long before the spread of mechanical means of contraception, it can easily be seen that having babies was a social and not a purely natural fact. Who had babies, who married whom and at what age, how many children were born and how many survived – all these reflect changeable social relations, varying from one mode of production to another and even from one specific economic situation to another. Demographic facts do not simply respond to economic change; they may even help to create new economic facts, such as the industrial labour force in the late eighteenth century. They may vary, in one society and one economic situation, between classes: for example, in the eighteenth century the French peasantry restricted births (presumably by coitus interruptus or abstention) in order to keep the family landholding intact, while the urban proletariat had more illegitimate births, more unions libres (common law marriages) and more children generally. Unless we realise that changes in population are overwhelmingly the result of human activity and human choice, we end up with the ‘hidden hand’ of mysterious and uncontrollable demography replacing the ‘hidden hand’ of the economic process.

Procreation is, however, only a part of the reproduction of the labour force: the rearing of children and maintenance of adult workers accounts for far more time and effort. Chris Harman’s article argues that Kath Ennis’s demonstration that ‘capitalism had an economic interest in socialising certain aspects of housework, so enabling women to be exploited through the labour market’ proved, by ‘irrefutable logic’, that the family is not essential to capitalism ‘in the same way as exploitation and accumulation’. (pp. 18–19)

The logic certainly can be refuted, because capitalism’s interest in socialising certain aspects of housework in order to make women available and willing to work for wages does not prove that capitalism has an interest in socialising all aspects of housework. On the contrary, the partial socialisation of housework has certain advantages for capitalism that the complete abolition of the family would lack. The dual position of women as workers and housewives facilitates their exploitation on worse wages and conditions than men. The fact that women move in and out of the labour force according to family needs keeps the market for women’s labour (and to some extent the market for men’s labour also) in the shadow of a reserve army of women seeking work or about to seek it. Finally, the privatised family system of reproducing the labour force is more flexible than any socialised system: in hard times, families are tightening their belts and changing their priorities, not rioting in the dining rooms or tearing apart the dormitories. This brings us back to the point that capitalism is prevented from abolishing the family ‘because of its own crisis-prone nature’.

It is not only ageing capitalism that has crises: I seem to remember that both youthful capitalism and middle-aged capitalism had short-term crises as well as the long-term tendency for the rate of profit to fall. Chris Harman himself has just written an excellent book explaining that crisis is an inescapable fact of capitalism. Yet he seems to argue in his article that capitalism’s ‘crisis-prone nature’ is somehow not ‘essential to capitalism in the same way as exploitation and accumulation’. This seems to me to be splitting hairs.

The distinction between base and superstructure, hallowed though it is by the Marxist tradition, also sometimes seems to me to be splitting hairs and to be misleading more often than it is useful. For the superstructure is surely no less essential, or easier to get rid of, than the socio-economic base of class society. This should be obvious when we think of the state, which both Marx and Engels undoubtedly regarded as part of the superstructure: it is not autonomous, nor capable of being destroyed without destroying capitalist society, nor can capitalist society be destroyed without destroying it. It is, perhaps, more variable, in that one mode of production – e.g. capitalism – can support more than one form of state – e.g. bourgeois democracy, fascism, state capitalism. In this sense, the family resembles the state, for there is no doubt that one mode of production can accommodate variations in the form of the family (as in the case of Western European feudalism mentioned above). But to place it therefore ‘in the superstructure’ would be to ignore its place in the productive process in some class societies, and its place in the reproductive process in all.

In conclusion, I hope that we can continue to defend the position that women’s liberation and the struggle for socialism are one fight and should be one movement, without resorting to the dubious arguments that occasionally crept into Chris Harman’s article.


Norah Carlin Archive   |   ETOL Main Page

Last updated: 29 March 2016