(Presentation at the National Conference in Sociology, Røros, Norway 1993)

Sigurd N. Skirbekk:

Limits of predictive power in an individually based political ideology.
- Reflections upon some aspects of Swedish family policy

Abstract: Sweden has for decades been regarded as a prototype of rational modernity. But not all aspects of that country´s policy tell the story of rational control over development. In this paper a case of unexpected consequences of Swedish family policy is being presented. The latent dysfunctions of a supposedly rational policy is seen in relation to limitations of individually based political ideologies. The question is then raised whether sociological theories could offer more adequate explanations to processes in the reproductive area. A functional analysis, where reference is made to structures of institutions as well as to civilizations, is suggested as being the best type of theory available.


Sweden - the Prototype of Modernity?

For decades now, politicians and social scientists looked to Sweden as a kind of prototype of modernity. In this connection modernity has been understood in a cultural as well as in a political sense. Culturally, as principles for regulating society according to rational rules, based upon research and public support. Politically, as rules based upon specialized expertise and bureaucratic principles for universal and morally impartial treatment of individuals. The principles of modernity have usually been related to organizations of production. In principle, they could also be applied to the sphere of reproduction - and that has definitely been the case in the image of Swedish modernity.

The reproductive sphere in our neighbor country has also been the focus of other images. In certain circles Sweden has been regarded as a prototype of modern sin; or rather of a community without any consciousness of morality and sin, since all kinds of sexual attitudes and activities have supposedly been regarded either as belonging to the sphere of individual rights or as being a matter for social policy and the medical service. The contrasting view to this perspective of morality and sin has been a perspective focusing upon the difference between liberating potentialities through educated enlightenment as against the submissive tendencies inherent in superstition and constraining traditions. Spokesmen for pietism and liberal radicalism alike have legitimated themselves in a negative contrast to each other; thus creating closed intellectual circles where discussions of functional consequences have been displaced to the side line.

These images could in themselves be one reason for non-Swedes to look into the sphere of reproduction in that country. Such attempts could be particularly interesting for Norwegians, not least because everything Swedish has a tendency to become Norwegian some ten years later. Moreover, as Norwegians and neighbors we should perhaps have certain qualifications for being able to see Swedish conditions from both an inside and an outside point of view.

A more scholarly reason for paying attention to the reproductive sphere in Sweden is that the Swedes have produced relatively reliable statistics in this field for a long time, statistics that could serve as a basis for judging the adequacy of dominant ideologies used for guiding developments in these fields. Data might even tell us something about who has been the more ideologically deceived: those who have seen developments in this field as an example of increasing intellectual control, or those who have seen them as an example of failure and maladjustment.

In the field of reproduction the simplest kind of data would be the periodical measures of fertility rates. In Sweden such data are available in the form of periodical mass statistics from the middle of the 18th century.

Swedish fertility data have long been split according to whether the children were born inside or outside of marriage. This distinction has been regarded as publicly important; partly due a moral concern and partly as an outcome of social and financial considerations: Children born outside of wedlock very often meant single mothers who could not support their offspring, and who consequently would become a burden for the rest of society. Children with unwed mothers were often unwanted children.

In our century, with greater financial resources, the reasons for wanting to avoid having children born out of wedlock might partly have changed. Concern for the mother has been one type of predominant argument, concern for the children another. As the central professions and public spokespersons in these matters have changed over time, from the theologians to the medical profession and people of social policy, so the recommended remedies have also shifted, from concentration upon individual character and moral behavior, to public education in technical methods of contraception.

If the functional reason for the old morality and its justification was solely a concern for avoiding unwanted children and VD, the availability of new technical devices could be readily seen as a functional alternative to the old morality. But as social scientist we cannot be sure of the adequacy of such interpretations, even if these have had central spokesmen among the men of medicine. There are sociological reasons for assuming that profound norms and institutions have more than one manifest function, and that the latent functions of a moral arrangement are quite often wider than the traditional justifications given. For a sociologist it should for instance be relevant to query whether certain inhibitions to individual instincts and needs have had functions for the formation of monogamous family units, especially in periods when this formation is based mainly upon certain cultivated emotional ties rather than on support from social customs and economic necessity.

This issue could, in other words, be used in analyzing whether the assumed rationality of dominant explanatory theories really fulfills the more pro-found part of the program for modernity: To replace tradition with theory.

The case of Swedish family policy

Demographic data from Sweden can hardly be said to demonstrate very sensational changes from one decade to another during an almost two hundred year period from the middle of the 18th to the beginning of the 20th century, even if the aggregated changes are considerable. The population increased in size. People gradually come to live longer. Most people married and had children, even if the number of children per family has decreased in recent generations. The proportion of children born out of wedlock showed on the whole some tendencies to increase.

In this century, however, we can register several rather dramatic demographic changes over a relatively short time. Take for example data on children born inside and outside of wedlock. During the 1920´s statistical data showed that children born to unmarried couples had reached about 15 % of all birth. Some spoke of a moral decay. Others, professionals and people connected to the political Left, said that the challenge of these children would have to be met with new educational measures. Young Swedes had to become more enlightened about the possibilities of contraception. In the beginning these programs could be advocated as part of an increased moral responsibility, later as part of a supposed morally neutral technical instruction. Under the management of RFSU, sex education has been legitimated as part of an ideology for human liberation. But throughout the first decades of this century, Sweden´s official programs for sex education had as a goal to reduce the number of children born to unwed parents.

Apparently these programs worked. Between 1930 and 1950 the proportion of children born outside marriage decreased, from about 15% to some 10%. This success led to new efforts to strengthen the educational programs. But then unexpected things began to happen during the 1950´s. The proportion of children born out of wedlock did not decrease. In the 1960´s it regained its former levels, and during the 1970´s and later in the 1980´s it more or less exploded (See Fig.1). In 1986 more children were born out of wedlock than to married couples. The Swedish rates of marriage fell in the period 1970-1978 to the lowest level in Europe. A Norwegian demographer wrote in 1981 that if the new birth cohorts should continue the actual trends of marriage in Sweden, only one half of that cohort would ever marry.

Fig 1. Sweden - Children born out of wedlock per 10 000 newborn

(Statistiska Centralbyrån)

We could ask: What had happened that should not have happened? - Young people were obviously no longer so eager to get married. Many not only postponed marriage, they did not marry at all. Parallel to the new cultural signals of the sex education programs we could witness a change in sexual behavior in relation to marriage. The whole institution of marriage was increasingly perceived as a burdensome trap, with many legal and social obligations and with few benefits that could not be obtained independent of marriage.

Swedish authorities reacted. By the end of the 1960´s it had become clear that marriage as an institution had lost support among young Swedes. However, the nation as a whole did not stand to benefit if marriage were made obsolete, nor if matrimonial relations were made so private and loose that one-parent families became predominant and many children grew up without a close relation to their biological fathers. An official committee with a broad mandate was appointed.

The first report from this committee - called Familj och äktenskap 1 (Family and Marriage 1) - was published in 1972. It presented arguments for why marriage is a social concern and not simply a matter of private choice by reference to the social sciences, an expression of a deliberately rationalistic tradition in Swedish Social Democratic thought: "Childbirth can in a modern society, with its medical resources and far reaching differentiation and specialization, theoretically be isolated from the family and become reduced to a solely biological event. Sociology knows, to our recognition, of no society where that has been the case", the report said, and went on to refer to "psychological, pedagogical and sociological literature"

showing that harmonious development of a child´s personality during its first years of life is crucial. "The more norms the child meets, and the less cohesive these norms are, the more important the family will be in helping to blend the norms together". For this task it was regrettable that the father´s role as breadwinner prevented him from taking appropriate part in the child´s development, while the mother could have a too dominant influence. Reference was thereby made to the radical program for women and men sharing jobs both at work and at home, at the same time as the implicit message was made clear: Children need stable contact with their fathers; politics on marriage and the family had to make this a goal.

How to aschieve this goal was not described in terms of structural premises for the institution of marriage, but rather as a change in the conditions of cost and benefit in the choice for the individual. The committee said: "The difference between living together outside marriage and doing so in marriage can consequently be reduced through a combination of limiting the legal implications of marriage and introducing regulations for actual cohabitation with the same content as the regulations for marriage." Through this combination members of the committee thought it would be realistic to expect a halt in the falling marriage rates, and that it would be possible to establish more stable and responsible relations where fathers could take greater care of their children.

With some modifications, the proposals for a rather radical new law of marriage along these lines were accepted by the Swedish Parliament as of 1. 1.1974. Statistical figures from the 1970`s can tell something about the effect of this law. Several data that are of interest in this connection. The first trend to note is that the steep drop in marriages from the mid-1960´s suddenly stopped in 1974. The changing conditions under the new law had obviously had their effect; there are no other plausible explanations for this change - for instance, sudden new cultural signals at that time or a considerable variations in the cohorts of relevant ages. Second, the new trend of increased popularity for marriages was very short-lived; after a peak the curve later stabilized, but at a considerably lower level than in the 1950´s and the early 1960´s. Third, parallel to an increase in marriages we also find an increase in divorces. If the new law should be seen as responsible for more marriages in Sweden after 1974, it also has to be held responsible for more divorces. The law had made it easier to get married and easier to get divorced; this dual change was part of the philosophy of putting fewer duties into marriage and more duties into living together in order to equalize the alternatives without distorting the individual freedom of choice.

Was then the law reform a success or not? - If the only goal had been to put an immediate stop to a falling tendency of marriage rates, then it could be said to have achieved that. It had also demonstrated that the law can direct human behavior. But if the main goal had been to keep more people in stable matrimonial relations, then it could hardly be judged a success. It is also highly dubious if it could be judged successful in relation to the goal of creating social frameworks for greater contact between children and their biological fathers.

Some defenders of this law would say that it had at least achieved something; they would explain the limitation of these achievements as the price to pay for a liberal concern for individual freedom of choice. The concern for men´s, and not least women´s, right to choose their own way of life had put limitations on the efficiency of legal regulations, so it has been said. An opposing view would be that such statements mainly serve as ideological justification for a development that has in fact got out of control.

The latter view could claim some support from other Swedish studies of attitudes to marriage among cohabiting couples. In the mid-1970´s Bo Levin and Jan Trost surveyed the issue of individual choice in this field. They found that any conscious choice by the individuals involved, based upon ideological value-preferences for cohabitation instead of marriage, was a rare phenomenon. The more usual answer among the cohabiting couple, when asked why they had "chosen" living together, would rather be: "it just happened that way". A study based upon interviews with 279 lawyers who had had much to do with this clientele concluded that the most commonly expressed reason for choosing to live together was that the clients preferred not to "bind themselves". In another survey 53 % of a sample of cohabiting individuals said that both they and their partner intended to get married, while 33% said that they themselves but not their partner wished to marry. Freedom for one part is not necessarily freedom for the other. In the last-mentioned article, the generally reluctant attitudes toward marriage were not explained as a result of opposition to strict matrimonial patterns, but rather as a consequence of weak and vague social norms on these matters in Swedish society.

If the trends toward more cohabitation at the expense of marriage was to be interpreted primarily in terms of individual choice, experimentation and self-realization, we could have expected increasing quality among the select sample of married couples, and at least a tendency toward greater stability among those who did marry. Several defenders of the Swedish case have argued along the lines of quality vs. quantity. But even this general conclusion conflicts with available statistical data.

Contrary to many expectations, the divorce rate has increased as selection through cohabitation has become more dominant. In Sweden the ratio of annual recorded divorces to 100 marriages had by the mid-1970´s even surpassed that of the United States. Sweden has long been a leading Western nation with regard to widespread acceptance and practicing of living together outside marriage. In the 1970´s Sweden also demonstrated a lead in divorce rates, as is clear from Fig 2.

Other data also indicate that we have been witnessing a change in the cultural meaning of marriage and cohabitation, and not merely some changes in choice between fixed alternatives as the conditions of costs and benefits changed. For instance, the chances that a cohabiting couple will later marry have fallen drastically as living together has become common in society. We have data that demonstrate this tendency in Sweden as well as in Denmark and

Fig 2

(Source: Michael Anderson)

Norway. The majority of cohabiting relations that started for women below 25 years of age did end in marriage for the cohorts born in the 1940´s, but this holds true for only a minority of younger cohorts.

Not only has cohabitation largely replaced the institution of marriage as a frame for fertility, but the proportion of women giving birth outside marriage and outside cohabitation has increased as well. - With Durkheim, we could perhaps even say that married couples became "less married" as cohabitation and divorces became increasingly common.

The inadequacy of political ideologies

We could have delved more deeply into data on the reproductive sphere in Sweden. However, the intention of this paper is not to present a complete demographic overview, but rather to show that the direction of development in a country assumed to be guided by rationality and expertise, has not followed the intentions of intervening experts. We should then ask what the leaders of society had overlooked in their supposedly rational premises and conclusions.

It is easy to say that politicians are usually so concerned with their legitimating ideologies, and the application of these that they do not always seek out the best theories available. If they are convinced they already represent the objective interest of the people, they may not even bother to ask what the majority of the people think about specific matters. In the case mentioned, the conservative minority in the committee who wrote Familj och äktenskap 1 - the report that paved the way for the marriage reform of 1974 - blamed the radical majority for their ideas as to what the Swedish people wanted at that time. Other contemporary studies of Swedish opinion also contradict the notion that the elite in democracies generally represent the value-preferences of the majority of the people.

But the inadequacy of political pattern of thinking cannot be reduced to particular politicians in a single country or to a question of the political Right or Left. Elsewhere, as in the UK and USA, we have witnessed how supposedly conservative politicians have failed in their programs for strengthening family ties by paying more attention to individual responsibility and less to social security. Divorce rates, for instance, increased during both President Reagan´s and Prime Minister Thatcher´s rule, and so did the proportion of children born to unwed mothers. The Swedish case may in some way appear extreme, but it is not unique.

Nor can the unpredicted direction of development be explained simply by some general reference to the people´s choice in opposition to the political elite. Several studies of value attitudes among the majority, by the elite commonly described as "prejudice", indicate a widespread skepticism to dominant liberal trends of development in this area. Explaining the falling number of births below replacement level by referring to women´s liberation and women´s choice, does not hold water. More or less compatible data from different European countries on women´s preferred number of children versus calculations of actual number of children show a considerable discrepancy. (See Fig. 4). In contrast to several expectations, the direction of the discrepancy does not indicate that women are giving birth to more children than their prime preference - something which might previously have been the case, if we limited out attention to conditions in the beginning of this century, when fertility fell from 4-5 children to 2-3 children per woman. This explanation cannot in the same way be made valid for the fertility drop in recent years, when the European nations, with few exceptions, have fallen to a level of 1.3 to 1.9 children per woman, clearly below the magic figure of 2.1 children per woman of fertility age, defined as the level of replacement for a population.

The discrepancy between women´s prime preferences and their actual number of births has to be explained in relation to family structures and to the new cultural climate around reproduction. Theories based merely upon individual choice and value-preferences are inadequate to explain the kind of processes we have been witnessing in the reproductive sphere. The consequences of these changes should also be judged in a supra individual perspective. For instance, keen arguments can be adduced for why the recent drop of fertility below the replacement level is related to the change of family structures; a certain rise in fertility among cohabiting couples in the period do not balance the fall in fertility as the married part of the population has decreased.

Conclusions along structural lines like these will undermine assumptions often made by politicians on the Right as well as to the Left. Both social Liberals and Social Democrats have tried to run society mainly by influencing conditions of costs and benefits for assumed free and rational individuals, on the market or within the social political bureaucracy. This way of thinking can coincide with certain notions of culture as mainly being symbols for social identity or for individual entertainment.

Fig. 3 Ideal and realized number of children in some Western countries

(Source: Tab. 9.8 in Svein Blom: Familieverdier, Rap. 22, ISS, Tr.heim 86)

Could sociology do better?

When political ideologies prove inadequate, sociologists should see it as their job to offer better explanatory theories and models for understanding contemporary processes. A sociologist of knowledge should also be able to provide explanations for why dominant ideologies may prevail despite of their relative inadequacy. In the field mentioned, the challenge for sociologists should thus have been much wider than just to find measures to enable accurate and neutral registrations of what is going on, presupposing that ongoing social change is part of a universal law of development or at least a result of democratic preferences.

The raison d´etre for sociology as a specific subject is that supra-individual structures can be used as an explanatory basis for social processes. Cultural and social structures, norms and roles, predispose individual actors for collective forms of behavior. This behavior might be functional for the survival of the society of which the actor is a part, or it may in the long run prove dysfunctional. In either case, behavior should be related to dominant forms of supra-individual structures and not only to the question of individual actors´ social conformity or deviance. When widespread behavior should be judged dysfunctional, this is often because the structures of societies are inadequate. There is no necessary connection between a dominant social structures and an adaptive capacity .

It is easy to locate some statistical evidence for a social development, and call that a description. Neither will it usually be very difficult to find some factors correlated to this development, and eventually to cite them as an explanation. When we have data related to the survival of a population, like the present data on a reproduction rate below replacement level, it is not even difficult to give a functional judgement of the development, which in this case must be judged as dysfunctional. More difficult is to find an adequate level of generalizations for relating structures to functions. This requires a general sociological theory. Without an analytically constructed system of substantive theory, we are - as Bernard Barber put it - often interesting but never scientific.

In order to be adequate for an intervening orientation, theories of society have to take into account the necessity of fulfilling certain functions, various contributions to the survival of society. Theories of modern societies have to consider the necessity of differentiated institutional frameworks around different tasks. The concept of institution can, at least initially, be useful for relating cluster of norms to functional requisites and to modern specialization. The task of reproduction and the task of integration cannot be adequately solved by norms deriving from the institution of production or the institution of political government, even if several spokesmen for liberalism and socialism have tried to reduce public concern to these two spheres. Sociologists should have a great deal to say about the kind of norms that have to govern the reproductive sphere if it is to fulfill its tasks. Sociological reasons can be adduced for why these norms will hardly be maintained without a separate institution of marriage.

Some might think such reflections look like an attempt to revitalize a presumedly obsolete Parsonean sociology. That would not be an accurate assumption. Talcott Parsons did not develop a theory for separating functional and dysfunctional behavior, even if several of his theories on the functional necessity of structural differentiation in a modern society could be valid enough. The missing conceptualization of dysfunctional alternatives even made him see forces of individuation and system equilibrium as regulating the specialized functions of the new family institution, when it would have been more accurate to notice a combined infiltration of commercial and political norms of individual self-realization in a privatized reproductive sphere. Not even Robert Merton managed to develop operational criteria for determining dysfunctionality, even though he set out several considerations which should be taken into account in functional analysis.

Theories of functionality should be kept strictly to criteria for long-term survival, so as not to be confused with all kinds of particular goals and individual preferences. A structure is "functional" if it contributes to the survival of entities of which that structure is a part. It is "dysfunctional" if it in the long run mainly contributes to diminishing the survival capacities of life or civilization at a higher level. With such a definition, it is not very problematic to conclude that a family structure which leads to a reproduction rate around 80 % of the replacement level, is in the long run dysfunctional for the populations and societies concerned, even if the majority of individuals in those societies may have learned to evaluate their lifestyle according to other measures.

If the family structure of certain dominant groups in a society is dysfunctional with regard to biological and cultural reproduction, other groups could over time be expected to take over the social leadership. If a society cannot mobilize alternative groups with a more functional way of life, the whole society can be expected to lose influence. If most societies within a civilization adjust to dysfunctional reproduction - as for the time being seems to be the case in Europe - other people from other civilizations can most certainly be expected to take over the land and the lead in a foreseeable future. It is the job of sociologist to say such things, even if this contradict basic assumptions in popular ideologies.

People with a natural scientific orientation have had a tendency to reduce social explanations to material conditions. People with the culture of the humanities as their source of orientation might on the other hand have had a tendency to see collective activities as deductions from a common Weltanshauung or from a basic belief system. Sociologists should not have to make a sweeping choice between materialistic and idealistic interpretations. The particular perspective in a general social science is rather to see society as an entity of interrelated subsystems, each structured in relation to differentiated tasks. If a modern and differentiated society is ruled solely according to religious norms, it may destruct its adaptive potentialities. If mainly ruled by familistic norms, it may become unjust, ineffective and perhaps turn in a mafiose direction. If political organs should rule everything, it may become very bureaucratic and perhaps totalitarian. If economy should dictate the common norms, basic trust and integration will most certainly suffer. And so we can expect in societies where one or more subsystems are deinstitutionalized and ideologically individualized or where the institutional frames are foreign and dysfunctional for the tasks concerned.

This could be the basis for a sociological model applicable for analyzing long term consequences of a family policy without privileged support in an institution of marriage. We could ask to what extent commercially initiated notions of "love" can be functional for family formation. We could further ask to what extent economic support alone, congruent with the norms of social policy, can motivate couples for an adequate reproductive behavior.

In order to construct adequate theories for functional and dysfunctional adjustment, sociologists have to include two levels of theoretical considerations not included in the previous theories of structural functionalism: the level of supra-state civilization and the level of supra-civilizational ecology. Ecology tells us about natural limits of consumption and production, by some ideologists seen as the main means for solving all kind of social problems. Comparative studies of civilizations can tell us about the functional limits of dominant myths and concepts about reality within supra state cultural circles.

A leading scholar of civilizational studies within the field of sociology was Pitirim A. Sorokin. Even if Talcott Parsons in many ways was a pupil of Sorokin, Parsons never made civilization a question of functional or dysfunctional adjustment. For Parsons, as for most other sociologists, the subject of sociology remained a field for supra-individual, not for sub-civilizational studies.

This theoretical limitation of sociology as a science has also had its impact in the field of family sociology. Neither our dependencies upon greater civilizational structures, nor the requisites for maintaining the cohesive character of such civilizations have been part of this branch of sociology. On the other hand, within family sociology some researchers have related contemporary problems to characteristics not of a particular society, but of a specific civilization. One of those who have made family studies in relation to various civilizations and to different phases of development within the civilizations studied, is Carle C. Zimmermann. Mainly on the basis of historical data from Greece and Rome, he even meant he could put forward several characteristics typical of a late family pattern no longer functional for the maintenance of those long-dead civilizations.

In his presentation of Zimmermann´s theories of "the atomistic family" British sociologist Michael Anderson referred to these theories as to a path that was not followed. This might be an accurate description. But it indicated more about contemporary sociology than just an overlooking of certain authors who should have deserved more attention. It also says something about sociologists who have been more interested in judging social change in relation to certain measures of freedom or equality than in analyzing contemporary processes in relation to measures of functionality and dysfunctionality. Individual freedom and social equality are the two main values in liberal and social democratic policy. In the short run, sociologists might have gained popularity for making these measures the final test for critical sociology. The sociological project, however, was originally designed to grasp wider than that. If sociologists does not develop models and theories for how values of freedom and order, equality and differentiation are interdependent, and for how realization of these values depends on the institution of production as well as for reproduction, the institution of political government as well as institutions for cultural integration, the subject of sociology may loose its basis for original contributions to the understanding of modern societies. In this perspective the challenges raised by Swedish family policy could be a test case for our abilities.