Part II
Gender divisions
of power - citizenship,
rights and control
5    Gender and European welfare states
Context, structure and agency
Henrik Bang, Per H.Jensen and Birgit Pfau-Effinger
5.1 Issues: the need to contextualise welfare state policies
The way in which the welfare state affects the position of women is a highly contested issue, as the perception of causes and effects varies with the theoretical viewpoint from where observations and interpretations take place. For example, theories of patriarchy (Walby 1990,1994) argue that the welfare state is an instrument used by men for the oppression of women, while the empowerment hypothesis (Hemes 1987; Sum 1994a), on the other hand, argues that welfare state institutions may pave the way for the improvement of women's lot as workers, mothers and citizens. Such controversies as to how the welfare state is conferred with meaning have intensified in recent years, not least as a result of the paradigmatic shift from 'modernity' to 'post-modernity', and from 'feminism' to 'post-feminism' (Brooks 1997).
'Modernist' feminism is marked by essentialism anchored in an egalitarian rhetoric. As such, modernist feminists seem to have found the truth of women's nature, since they have developed normative and universalistic assumptions about 'what is good for women'. Starting from such a unified gender identity, modernist feminists then argue how a welfare state ought to be constructed in order to meet women's needs and aspirations. For instance, the welfare state should guarantee women's right to be 'commodifieď (Orloff 1993) or, alternatively, the welfare state should provide the possibility for women to form and maintain an autonomous household (Lister 1995).
From an anti-essentialist perspective, institutions and individuals cannot be analysed independently of the historical and social conditions that constitute them in their specificity for a given society and a given moment in time (Bourdieu et al. 1991). Given this, welfare state institutions have no virtual existence of their own. Rather, the properties of individuals and institutions arise from the broader cultural system to which they belong. This means that social relations, culture and contextualisation must be given primacy in the analysis of the welfare state - and especially in a comparative perspective.
Anti-essentialism entails a move from abstract ideas of gender equality to a debate about gender equality in relative terms. For example, the promotion
116    Gender divisions of power
of part-time work might benefit women in one cultural context, while being disadvantageous to women in another cultural setting. As such, conclusions as to what is beneficial for women can differ across countries because of cultural differences, just as women can lean towards different cultural models of gender equality.
Even though essentialism and anti-essentialism are two distinct approaches, we have witnessed some attempts to integrate the two divergent perspectives. In a recent article, Birte Sum (1999) argues for normativity and contextuality simultaneously. On the one hand, from a normative standpoint Sum tends to universalize conceptions about what is good for women, arguing that specific types of welfare state institutions are in accordance with women's identities and aspirations. An example is that of good quality childcare institutions. On the other hand, Sum insists that contextualising the framework of citizenship is crucial for analysing the driving forces behind the development of women's social and political rights. We consider it logically impossible to combine a contextual approach with a general normative one. Furthermore, we will argue, much confusion in the debate occurs as a result of the essentialist understanding of anti-essentialism as 'relativistic' or 'nihilistic'. Anti-essentialism has its own contingent methodological and theoretical foundations. Therefore, the first aim of this chapter is to give answers to the following questions: What does it mean to include an anti-essentialist perspective on culture in analyses of the relationship between women and the welfare state in a comparative perspective? What does it mean to speak of contingency with regard to the framework of citizenship? In answering these questions the main empirical focus of this chapter is the different social constructions of motherhood in Britain, Denmark, Finland, Germany, Italy, the Netherlands and Sweden. Such a country comparison will highlight the need for contextualisation in the face of well-known variations in welfare state regimes (Esping-Andersen 1990).
Furthermore, anti-essentialism raises new challenges as to how to come to grips with the dynamics of social change. From the point of essentialism social change is generated either by a 'self-unfolding' social structure or by the deliberate decisions of a rational actor. Anti-essentialism, in contrast, invites us to reconsider the relevance of this modern understanding of order as opposed to disorder, and of structures as opposed to actors. The contextualisation of gender relations first of all seems to require a new notion of the political which allows for contingency and which does not take the hierarchical type of authority to be the only possible type. Therefore, the second aim of this chapter is more programmatic in discussing and identifying the transforming capacities of society.
5.2 Patterns: variations in the gender policies of welfare states and problems of interpretation
Since World War II we have witnessed an overall increase in female labour
Gender and European welfare states    117
force participation. The feminisation of the labour force in Europe is, however, marked by huge national and local variations. These variations become even more striking if we consider the labour force participation among mothers with children under 11 years of age. As we see from Table 5.1, the participation rate among mothers for our case study countries varies from 49 per cent in Italy to 84 per cent in Denmark. As the labour force participation of mothers is normally lower compared with all women in the age group 25-54, it seems as though motherhood in general discourages women from participating in the labour market. This discouragement effect, however, differs markedly across Europe. As Table 5.1 shows, motherhood leads to a dramatic drop in female labour force participation in Germany, the Netherlands and Britain, while the Danish experience seems to indicate that motherhood actually encourages labour force participation (see also Chapter 9). Similarly, there seems to be no clear-cut pattern in the relationship between full-time and part-time employment. As Table 5.1 shows, the Netherlands and Italy exhibit similar participation rates for mothers of young children, but diametrically opposed situations for shares of part-time and full-time work.
The distribution between part-time and full-time for working mothers may be dependent on demand or supply-side factors. From a demand-side perspective Anne Marie Berg (1989: 229) has argued that 'part-time work in itself is the ideal manifestation of the female population as a "reserve army of labour" '. But what are the meanings and consequences of such statements? Does it mean that women serve as a reserve army to a larger extent in the Netherlands and Britain than in Finland and Italy, where only a very small proportion of mothers are working part-time? Conversely, from a supply-side perspective it has been argued that part-time jobs may bridge the juxtaposition between family and work and as such contribute to the integration of women into society as a whole (Nätti 1993). According to this line of argument, jobs have been constructed as part-time jobs in order to meet women's needs, and hence women in the Netherlands and Britain would be better integrated into the labour market than those in Finland and Italy. Thus, part-time work may be infused with negative or positive meanings depending on whether we look at the phenomenon from a demand- or a supply-side perspective (see also Chapter 2).
When we are doing comparative research another question is whether part-time work, as a social phenomenon, is actually comparable cross-nationally. In Britain, for instance, women's part-time jobs are largely located in the private service sector; often this is unskilled work, where usually the degree of unionisation is low, wages are low, employment is unstable, employment protection is weak, hours are low, and so on. In the Scandinavian countries, in contrast, part-time work is largely located in the public service sector, the work is skilled work as social programmes are the primary fields of public employment, the degree of unionisation is high, wages are relatively high, employment is stable, hours are not far below full-time, and so on. Furthermore, in the Scandinavian countries part-timers can fully participate in the social security and benefits system as individual workers. [The fact
CO
CT)
C
O u
3
si O.
KS
3
I
-o
h
í

C    H
«; Š
ä
■o, SP
3 Š
«1!
fto
II
5S
S1?
^    a    Äŕ *T)
>-l    «U,   S   <N

r—       co oo co <o —■ —<
•* oo
—i  CM
LO O
N* CO
(O "* CO
co o co m
Tf> cm r«, cti
co co p» in
CT) -h O) ^ iO tO
—; CM —; —;
cm to iri c-i co co co i~»
	c 'S		
E	T3  <s	S	%
c	*1 C/2   [14	ÍH	a
Q			W
nm* in to r-~
-O P3
-S
U s
a
o Q
u £
■S3
áž«
Gender and European welfare states    119
that part-time workers have access to unemployment benefits may contribute to the explanation of the relatively high unemployment rates among women in the Nordic countries (see Table 5.1), as there is an incentive for part-timers to be registered and stay in the labour market as unemployed.] Thus, as the substance of part-time work differs markedly across Europe, we can state that part-time work is not part-time work, is not part-time work.
In the social science literature the integration of women into paid employment is treated mainly as a supply-side phenomenon. However, within mainstream literature it is possible to make a distinction between two major approaches. First, the economic literature argues that the relative wages of women are crucial for the decision of mothers to enter the labour market, inasmuch as the higher the relative wages of women the higher the utility of entering paid employment (see, for example, Gustafsson andjacobsson 1985; Mincer 1985; Blau and Ferber 1986; Becker 1993). As such, the decision of mothers to participate in the labour market is seen to be based on calculative reasoning. The relative wages of women may also be structured by the welfare state in the form of equal pay and equal treatment legislation, on the one hand, and the welfare state as employer, on the other. Second, childcare and parental leave institutions enter into most discussions of women and the welfare state, in that there is a substantial literature on how these institutions allow women's autonomy and independence by facilitating female labour force participation (see, for example, Kamerman and Kahn 1981; Leira 1992; Pringle 1998). That is, childcare and parental leave policies are considered to be central indicators for the measurement of the commitment of the welfare state to integrate women into the workforce. From these two major perspectives the key elements in a definition of citizenship of mothers has been outlined (Table 5.2).
Tables like this are commonly used in comparing different types of welfare regimes (Jensen 1996; Pfau-Effinger 1998a). Most often such tables are based on an underlying assumption about causal dynamics and linear model thinking. Thus, on the one hand, we may argue from Table 5.2 that cross-national differences in the pace, degree and form of women's labour force participation are an outcome of institutional differences. From such a perspective, the behaviour of women may be interpreted as an automatic or mechanistic response to the opportunity structures of the welfare states; that is, the behaviour of women is more or less programmed or determined by the state. On the other hand, we may argue that women's labour force participation is a deliberate response to the utility of a given behaviour. Thus, women's informed and premeditated decision to participate in the labour market is based on instrumental rationality.
We find such interpretations understandable, but also highly problematic. A first reading of Table 5.2 may indeed support linear model thinking as differences between welfare states seem to 'fit' with variations in the labour force participation of mothers. A second and more careful reading, however, reveals that there are no simple interrelationships between these key elements
-o tí
B!
-o
tí .3
X! 3
a,
»T
ho C
v
S CT) CT)
« -O
o .S
li
i Ť
o  S
•S   c
rt   u
.S-B
u   u
•S,   bo
C    a
«s    a
u    c«
O
XI ■ ta
J3
H
CM
3
•&
It?
-ŕ1   ^ Q   S-
I
<Č5
1   I
«,*, -$>?
J3    §>
ft, "a


C
ti
ft, -§
2
■«   £
s?    s;
a   8
e   ■«
«   K
•í;   *>
-c;    **>
JJI
jj
g.š §:s
-5 t;
m*J-š
c«    w    «    tň
*ŠZ & ►£ »■&
/* r r r1
t/J     M     W     W
■n       „j
£!    ^ i
■* (M "l Tŕ
—i r- ct> —<
o o o
V2       V2       n
O CO o r»
CM cm en co r-~ m
en i*« to
o CM
CO CO —i •*  CO CM
cs r-*-
m o r-
co CTi r^
m CT) CO
CD  MO
o ö d
■* cm r~ CTi co co r-~ un
B!
«"2 S
c  u  O   a
tí   u -a   u
u   >   S   J)
QcŽEu
CT> —< CT)
■* io m
-a c c«

-S S
J3
O | <u
z
c
S
e
o U c
3
w
3 W
U —
81
Gender and European welfare states    121
of citizenship. For instance, from Table 5.2 it is very difficult to explain the low level of participation in the labour market among mothers in Italy. First, as we see from Table 5.2, the ratio of female to male earnings is very high in Italy and equals, for example, the Danish situation. Thus, from an economic perspective, participation in the labour market has a high degree of utility. However, mothers in Italy do not enter the formal labour market to the same degree as in Scandinavia. Second, as Table 5.2 reveals, the combined effect of (1) day care institutions for children between 3 and 6 and (2) maternity leave arrangements is relatively high and equals, for instance, the situation in Denmark. That is, welfare institutions in Denmark and Italy allow mothers to participate in the labour market to a similar extent, but mothers in Italy still exhibit far lower participation rates than in Denmark.
Such inconsistencies are often labelled 'misfits'. 'Misfits' are, however, only rarely mentioned in mainstream literature on gender and the welfare state and, if they are, they are mentioned parenthetically. For instance, Millar and Warman (1996: 21) write that 'the fit is perhaps not so close here'without addressing what are the consequences of such misfits, and how can we account for the misfits?
Misfits often occur because similar institutions, or institutions with similar names, have different meanings and different steering effects in different social and cultural contexts. For instance, high female wages may be interpreted as an egalitarian measure stimulating the entry of women into paid employment. High female wages, however, may also be interpreted as a measure making it uneconomic for employers to take on women workers. Historically, this has been the case in Italy, where female wages reached 83 per cent of male wages as early as 1945-46 (Lorini 1975). Or, conversely, institutions which from the 'outside' look very different may actually have the same nature or structure. For instance, historically Sweden has had high-quality maternity leave schemes in terms of coverage, duration and entitlements while, alternatively, Denmark has had a high degree of coverage by high-quality day care institutions (see Table 5.2). Nonetheless, maternity leave and day care institutions may be functional alternatives to the same problem, that is to organise conditions allowing entry of mothers into the labour market. As such, different institutions with different names may actually have the same meaning and harbour the same goals.
Similarly, it is not sufficient to observe whether welfare schemes are present or absent in one country or another, because properties of welfare institutions differ markedly cross-nationally. This is the case, for instance, for day care institutions (Moss 1990). First, day care may be constructed in order to improve the socialisation, life quality and welfare of children vulnerable to neglect or abuse. Second, day care may be constructed to bridge the needs of mothers in combining childrearing with paid employment. Third, day care may be considered as an institution of relief in order to improve the welfare and functioning of the family. All these motives may be present in the day care institutions in a given society. However, one motive often predominates.
122    Gender divisions of power
In Britain, for instance, public day care has up to now been effectively limited to 'problem children'. Services provided in the Nordic countries are much influenced by the intention of enabling women to combine motherhood with paid employment, which, among other things, is reflected in the opening hours of the day care institutions. In Italy social and educational motives predominate. This is also the case in West Germany. Here, public day care is seen as part of the educational system and as a means to socialise children in groups at a time when families are getting smaller and having siblings is less common (Kaufmann 1995). Only recently has the political aim to give mothers the possibility of combining parenthood and employment (at least by working part-time) been added in Germany, although this is still limited to pre-school children aged from 3 to 6 years. A right for children to participate in public or publicly financed childcare was introduced in 1996 and includes the possibility for full-time care, even though many childcare institutions are still limited to part-time care. This is not simply a question of public provision of childcare but also one of the demands of parents who often do not want their children to spend the whole day in a childcare institution. Although former East Germany is now integrated into the same welfare state, local government there normally provides both full-time childcare for pre-school children and after school care as well, and this better matches the demands of women who are traditionally more oriented towards continuous full-time employment (Pfau-Effinger 2000). Thus, in Germany, even within the same welfare state context, different cultural and institutional traditions, and different demands from citizens, have contributed to a differing public provision.
A focus on the presence or absence of one scheme or another also tends to neglect the quality of such schemes in terms of level, coverage and staff. In Germany, for instance, we find rather generous maternity arrangements in which women receive benefits for 14 weeks at 100 per cent of previous earnings. Such generous benefits do not, however, necessarily overcome the problems of juggling family and employment. In Germany, the generous maternity arrangements are counteracted by the low level of coverage of services for children. That is, the generous maternity arrangements in Germany do not function as a 'bridge' back to employment. Instead they function within an overall framework of public policies directed at promoting traditional family norms, and encouraging women to take on a 'homemaker' role for at least three years. Therefore, the maternity arrangements in Germany are a flexible exit mechanism from the labour market, quite different to the entry intentions of maternity leave arrangements in the Nordic countries.
We can see that parental leave arrangements are designed to achieve different aims in different parts of Europe. It makes a difference, for instance, whether the parental leave arrangements are designed to complement a female 'homemaker' model, as in Germany, or are constructed with the aims of supporting equality among husband and wife in the caring and upbringing of children, as in Scandinavia. If the latter is the case, parental leave will be
Gender and European welfare states    123
paid and will normally be supplemented with a paternity leave scheme (see Table 5.2). On the other side of the coin, Britain has only very recently introduced a parental leave scheme and it remains unpaid. That Britain is a latecomer to parental leave, and that the leave is unpaid, mirrors a specific cultural and historical legacy according to which society is committed to values such as individualism and privacy, which in turn favour a traditional gender division of labour. Parental leave in Germany can be seen as something in between. It was designed in the middle of the 1980s with the intention of keeping mothers of children under 3 years old out of the labour market, and this fits into the cultural ideals of private childhood and a homemaker role for mothers with small children. It is paid, but means tested, so that the care-giver is dependent on a (usually male) breadwinner. It also mirrors in part the orientation towards waged work in the younger generation. Thus parental leave in Germany builds a bridge back to employment by including a right for parents to return to the former employer for up to three years. This is ambivalent, however, since the law did not include a right for parents to work part-time at the previous employer - which is the type of employment which mothers of young children prefer for it fits with the new cultural orientations. This is one reason why childbirth in Germany still ends, in many cases, with a long period of unemployment for women (Pfau-Effinger and Geissler 1992; Hoist 2000).
So far, it seems as though welfare state institutions have no virtual existence of their own, as the properties of similar institutions vary cross-nationally. We assume that the properties of institutions arise from the broader cultural system in which the institutions are located. Therefore, to analyse, classify and explain differences in welfare states with respect to gender policies, it is necessary to consider the existence of different cultural and institutional traditions. This is because cultural ideas penetrate the functioning of welfare state institutions.
5.3 Explanations: welfare state policies in the context of gender arrangements
In this section we wish to present an explanatory framework as to how culture and social actors may be included in the analysis of welfare states. This will then allow us to understand the properties of welfare state institutions and their dynamics of change. In the first section we present a conceptual framework as to how welfare state policies and culture interrelate, and in the second section we discuss more generally how to avoid essentialism in gender analyses.
The interrelations of welfare state policies with cultural and institutional frameworks and with social agency
What does it mean to include 'culture' in analysis or to 'contextualise' the framework of social citizenship? We suggest the theoretical framework of
124    Gender divisions of power
'gender arrangement' as described by Pfau-Effinger (1996, 1998a, 1999a) as a way of doing this. This theory conceptualises the interplay of welfare state policies with culture, institutions, social structures and social actors, and considers the dynamic of change which may develop within these mutual interrelations. This is a further development of the 'gender system' and 'gender contract' approach of Yvonne Hirdman (1988,1990; see Pfau-Effinger 1994, 1998a).1
According to the idea of a gender arrangement, a gender cultural system can be identified which includes values and ideals about gender relations, such as the gender division of labour, power relations and dependencies, motherhood and fatherhood. The gender cultural system is the result of conflicts, negotiation processes and compromises of social actors with differing power at an earlier stage of historical development. It has a strong impact upon the gender discourses and practices of each new generation of actors. The gender cultural system also forms a main reference point for social action and policy formulation at the level of the welfare state, where it is implemented as norms and expectations about what is normal, and in other institutions like the labour market and the family, as well as by collective actors and individuals themselves. However, under certain circumstances individuals or groups of social actors develop new ideas, or adopt ideas from different contexts, and may try to negotiate a new dominant gender arrangement. Change at the level of gender culture can thereby contribute to change in welfare state policies. The outcome is influenced by the power relations between these actors, for instance between the feminist movement as a new social movement on the one hand and conservative political parties on the other. The cultural foundations of welfare state policies are thus mutually interrelated with the dominant cultural ideals about family and gender relations in the population, and these also form a cultural reference point for other institutions such as the labour market and the family (Pfau-Effinger 2000).
Crucially, however, the way welfare state policies refer to those gender cultural models which are dominant in the population can vary. The interrelations may be more harmonious or, alternatively, they can to a certain degree be characterised by asynchronies and discrepancies. For example, the relationship between welfare state policies and gender culture was relatively harmonious in the Netherlands in the 1950s and 1960s, when welfare state policies exclusively promoted the housewife model of the male breadwinner family to which the broad majority of women and men were oriented (Ishwaran 1959; Heiligers 1992). In the 1970s, however, contradictions and asynchronies developed, for the cultural orientations of women were changing whereas welfare state policies still promoted the housewife model (Heiligers 1992; Bussemaker and van Kersbergen 1994; Knijn 1994a,b).
We do not support an ideational approach here, however, where the explanation of welfare state policies is restricted to the influence of cultural values and ideals alone. Welfare state policies are also the result of differing interests of social groups - and some social groups have more power than
Gender and European welfare states    125
others. Welfare state policies can therefore contribute substantially to the development of social inequality, including gender inequality. In this view, gender culture, institutions, gender structures and social actors are mutually interrelated within the gender arrangement, and the concept therefore forms an appropriate framework for cross-national analyses of differing welfare state policies and their dynamics of change (see Chapter 10).
This theoretical framework can also be used to analyse the degree to which welfare state policies can contribute to the empowerment of women and the promotion of gender equality. We would argue that welfare states promote the aim of gender equality best if they refer to dominant ideals and orientations towards gender equality among women. The Netherlands represents a good example of the importance of the analysis of the interrelations between welfare state policies and the cultural ideals of women. Part-time work by women (and also by men) has been substantially promoted by the state, in a relatively protected form and of similar quality to full-time jobs (Plantenga 1996). One could draw the simple conclusion that this promotion of part-time work for women was a more traditional gender policy, based on the exclusion of women from full-time employment. However, this would be a misleading interpretation. An analysis of change in cultural orientations in the attitudes and behaviour of women (and men) towards waged work helps to understand and interpret this development more adequately. In the Netherlands, a new and more egalitarian gender cultural model, based on part-time work by mothers and fathers, has developed at the level of cultural orientations by the majority of adults, mainly by women but also by a considerable proportion of men (see Halier and Höllinger 1994; Pfau-Effinger, 1998a). The promotion of part-time work by the state can be seen here as an element of an equalising gender policy, and is also supported by the feminist movement as a contribution to greater gender equality. In social practice this model had been realised only to a rather limited degree by the millennium. Even though the rate of men working part-time was the highest in Europe by 1998 (67 per cent; OECD 1998: 192), it is still mainly women who work part-time, and this means that they are dependent on the income of their male breadwinner as long as they are partnered. To empower women adequately with respect to the dominant cultural model of gender equality, a stronger promotion of part-time work for men would be needed. Moreover, some kind of substitute for the loss of full-time income would also be necessary, in order to free part-time workers from dependency on a full-time breadwinner. There is, however, strong pressure for change towards more equal family and employment patterns. Moreover, post-divorce social protection for carers is relatively high. A universal retirement scheme with a minimum retirement pension which is above the subsistence level also minimises the risk of poverty for workers who previously combined employment and care by working part-time (Bussemaker and van Kersbergen 1994; Knijn 1994b; Plantenga 1996). This is different for instance to Germany, where no minimum pensions exist (Veil 1996; Ginn and Arber 1998).
In other words, the promotion of part-time work in this particular context
126    Gender divisions of power
of time and space was an important step towards greater gender equality (Plantenga 1996; Pfau-Effinger 1998b). In contrast, in Finland a cultural model of full-time employment for all adults and comprehensive public childcare is dominant (Haavio-Mannila 1985; Anttonen 1997). Here, the traditional gender cultural model of both partners contributing to agricultural work was transformed, during the transition to an industrial and service society, into a dual breadwinner/state carer model. The tradition of full participation of all women in employment was maintained during this process, but adapted to the new situation of work outside the home by a strong expansion of the public social service sector and of comprehensive public childcare provision. The cultural tradition of the housewife family, and the idea of private, individualised childhood in the family, was never dominant in Finnish history. Therefore part-time work by mothers does not have any cultural basis, and does not match the employment orientations of women. Instead, women in Finland are usually oriented to continuous full-time employment (Pfau-Effinger 1994,1998a, 1999). The social practice of women fits to this orientation and the share of women working part-time has since the beginning of the 1970s always been about 10 per cent (OECD 1996: 192; OECD 1998: 192). In this societal context, any promotion of part-time work by the state would not contribute to the empowerment of women.
Cross-national differences in the gender policies of welfare states can therefore be substantially explained by differences in the gender cultural foundations to which welfare state policies refer, and by the interests of those social groups which are influential in the reconstruction of old compromises, or in the struggle for new compromises. It should be noted that the gender policies of welfare states overlap in a specific way with those policies of welfare states which refer more closely to the differing class interests of capital and labour, the varying outcome of which has been conceptualised by Esping-Andersen (1990) as 'welfare regimes'. The respective combination of gender cultures with class-based welfare regimes contributes substantially to cross-national differences of social policy between societies with similar gender cultures (Pfau-Effinger 2000). Simon Duncan and Ros Edwards (1999) have introduced a framework using such a combination, which they call the 'genderfare' model for the classification of social policy. This is based on the assumption that policy variations reflect variations in both the capital-labour contract and the gender contract (or gender culture), and that these are mutually interrelated. According to this model, welfare states vary according to the specific way both kinds of contracts are shaped and interrelate.
The classification of gender cultural models
To analyse gender policies of welfare states in their cultural context, we need a classification of gender cultural models. Pfau-Effinger (1998a, 1999) suggests a classification based on the following criteria:
Gender and European welfare states    127
1      Cultural ideals about the gender division of labour, the main spheres of work for women and men, the social valuation of these spheres, and the way dependencies between women and men are constructed.
2     The cultural construction of the relationship between generations, that is the construction of childhood, motherhood and fatherhood.
Using this classification model it is possible to distinguish between at least six gender cultural models in Western Europe, which include more traditional and new cultural models. The different models can be characterised as follows (Pfau-Effinger 1998a, 1999):
(1)  THE FAMILY ECONOMIC GENDER MODEL
This model is based on the cooperation of women and men in their own family business (farm or craft business), in which both genders contribute substantially to the survival of the family economy. Children are treated as members of the family economic unit, that is as workers, as soon as they are physically able to contribute. There may exist a strong sexual division of labour within the family economy, which varies according to context of time and space. This model was previously widespread in the agrarian regions of Europe, particularly where the agrarian structure was based on small family farms (Honnegger and Heintz 1981).
(2)  THE MALE BREADWINNER/FEMALE HOME CARER MODEL
This model conforms to the idea of the basic differentiation of society into public and private spheres. Women and men are seen to be complementarily competent for one of these spheres: men are regarded as breadwinners, earning the income for the family with waged work in the public sphere, whereas women are primarily regarded as being responsible for the work in the private household, including childcare. This is also based on a social construction of childhood, according to which children need special care to be supported comprehensively as individuals. This model was dominant at the cultural level in many Western European countries after the transition to industrial society, for example in Germany since the turn of the nineteenth century, and has lasted much longer in other countries such as Ireland. It also forms the cultural basis for model (3):
(3)  THE MALE BREADWINNER/FEMALE PART-TIME CARER MODEL
This is a modernised version of the male breadwinner model. The idea is that women and men are to an equal degree integrated into waged work as long as there are no dependent children in the household. During the phase of active motherhood, however, the combination of part-time waged work and caring is seen as appropriate for mothers. This model is dominant at the cultural level in West Germany and Great Britain, for instance.
128    Gender divisions of power
(4)  THE DUAL BREADWINNER/STATE CARER MODEL
This model conforms to the idea of the 'completed labour market society' (Beck 1994), with full-time integration of both sexes into the employment system. Women and men in marriage are seen as individuals, who are both breadwinners earning income for their own livelihood and that of their children. Like the male breadwinner/female carer model, childhood is constructed as a phase of life with its own worth, in which the individual child needs much care and support. But in contrast, caring for children is not primarily seen as the task of the family, but to a considerable extent the task of the welfare state, which is also seen as more competent for fulfilling this task than private households. The dominant gender cultural model in Scandinavian countries, particularly in Finland, approaches this model. Whereas the Scandinavian version of this model is in principle based on the idea of gender equality, a Russian version of this cultural model is more connected to the idea of male superiority, as can also be said for the dominant model in former East Germany (Nickel 1995). In France this model seems to be combined with pervasive ideas of women's femininity and motherhood.
(5)  THE DUAL BREADWINNER/DUAL CARER MODEL
This model reflects the notion of a symmetrical and equitable integration of both genders into society. In contrast to the preceding model, childrearing is to a large extent a responsibility of the family. The basic idea is that the family economy consists of an equal distribution of domestic labour - in particular childminding - and waged labour between a female and a male head of household. This is possible only because the labour market is organised in such a manner that structurally allows parents to fulfil a 'dual responsibility'. Such a model requires that domestic labour be financed on the basis of a family wage or on the basis of a state transfer system. This model has gained dominance at the cultural level in the Netherlands.
(6)  THE DUAL EARNER/MARKETISED FEMALE CARER MODEL
This model is based on the idea of full integration of women and men into full-time waged work. The family is seen as responsible for organising and paying for marketised childcare, by using commercial childcare facilities or by employing childminders in the household. This is a new model which has gained increased importance in countries such as Britain and the USA, but is not yet dominant at the cultural level (Crompton 1998; Yeandle 1999).
There also exist mixes between these models, as in Sweden, where a dual breadwinner/dual carer model is combined with elements of private care for children, and mothers in the phase of active motherhood tend to work part-time (although mostly with long hours not much below the level of full-time employment). In some countries different gender cultural models are
Gender and European welfare states    129
dominant at the same time, as in Germany where in former West Germany the male breadwinner/female part-time carer model is dominant, while in former East Germany the dual breadwinner/dual carer model is the central model. A co-existence of two dominant models can also occur when the gender arrangement is in transformation from one dominant family model to another, as in the Netherlands, where at the cultural level the dual breadwinner/dual carer model is dominant, whereas in practice most couples still live in correspondence with the male breadwinner/female part-time carer model (Pfau-Effinger 2000).
It should be remembered that even though there is a mutual interrelation between the economic structure and gender cultural models there is no deterministic impact of the former on the latter. Thus in Holland of the seventeenth century, which was an agrarian society (with elements of a commercial society), instead of the family economic model (which was the cultural basis for the family in many agrarian societies) the housewife model of the family was apparently dominant at the cultural level of the whole society (Schama 1988; Pfau-Effinger 2000).
Cross-national differences in welfare state policies, as indicated in Table 5.1, can to a considerable part be explained by the gender cultural model(s) to which welfare state policies in each country refer. This is not a simple interrelation, however. It should be noted that social policies are constituted by complex interrelations between different kinds of institutions and policies, as outlined above, including the way in which they interrelate with welfare state outcomes of the capital-labour relation (see the 'genderfare' approach of Duncan and Edwards 1999). In times of social, institutional and cultural change these interrelations may be characterised by asynchronies, and traditional and more egalitarian policies may be combined in discrepant or even contradictory ways.
The interrelations between welfare state policies and social practices of individuals are also very complex. The social action of individuals is not a simple outcome and is not determined by state policies, although this is often assumed when data on behaviour (such as the labour force participation rates of women) are used as an indicator for welfare state policies. Such an assumption does not take into account the fact that the social behaviour of individuals is a process which takes place in a very complex field of influences, where cultural ideals and values also play an important role. Thus Duncan and Edwards (1997,1999) have criticised the assumption of'rational economic man' on which analyses of the impact of welfare state policies on behaviour are often based. According to their argument, individuals do not simply act according to principles of 'economic rationality' but also with respect to principles of'moral rationality'. In turn these principles are related to cultural ideals and values. Similarly Pfau-Effinger (1998a, 2000) found that cross-national differences in the development of female labour force participation rates and part-time working in Finland, the Netherlands and West Germany could not be explained by simply referring to welfare state policies, but rather by the complex interplay of culture, institutions, structures and social actors.
130    Gender divisions of power
As long as the respective gender arrangement is coherent, the labour force participation rate of women may to a large degree conform with the aims of welfare state policies. This is because in such cases welfare state policies and behaviour refer to the same set of cultural values and ideals on gender relations, as for example in West Germany and the Netherlands in the 1950s and 1960s. Things can be different, however, in times of social and cultural change in the respective gender arrangement. Women may act according to new cultural orientations in spite of welfare state policies which are traditionally oriented, and sanction their new behaviour negatively, so that this even seems to be 'irrational'.2 For example, in the Netherlands and West Germany in the 1970s many women became culturally more oriented towards waged work and increasingly participated in the labour force, whereas welfare state policies were still conservative and promoted the housewife model of the family. Change may therefore take place at the level of culture and in the behaviour of individuals, while welfare state policies still promote more traditional gender cultural models - particularly in those cases in which the power relations still favour traditional (usually male) elites (Pfau-Effinger 2000) .3
To conclude, cross-national variations in the gender policies of welfare states can be more adequately understood if these policies are analysed with respect to the ways in which they are embedded in the social context. This includes the complex interrelations of culture, institutions, social structures and social agency.
5.4 Afterword: from essentialism to anti-essentialism in gender analysis
The chapter so far has shown that gender theory must reconsider the relevance of the state-civil society opposition for understanding and explaining changes and continuities in gender relations (Bang and Serensen 1999). It is far from evident that either the state, or civil society, is the natural starting point for assessing how new gender identities and activities affect, and are affected by, political institutions. The legitimate domination of the state may indeed be employed to set people free in certain contexts. Yet this freedom is acquired at the expense of political self- and co-governance by social actors (Bang et al. 1997). Furthermore, integrative norms and other forms of social capital maybe the 'glue' that holds a society together. However, they are simultaneously the site of orthodoxies in the social which prevent the pluralisation of gender differences, and the balancing of uneven gender relations.
As Foucault was one of the first to stress, 'power isn't localised in the State apparatus and ... nothing in society will be changed if the mechanisms of power that function outside, below and alongside the State apparatuses, on a much more minute and everyday level, are not also changed' (Foucault 1980: 60). Norms and administrative power do walk hand in hand in the social.
Gender and European welfare states    131
However, this does not necessarily imply that role and identity building must always be enwrapped in a circle of power and counter-power, as Foucault suggests. Power could function on the basis of the reciprocal acceptance of difference rather than as a relationship of superordination and subordination. Hence, power might be employed to do away with relations of class, race and sex, and other kinds of uneven relations, which subvert difference (Bang and Dyrberg2000).
We see a possibility of using a notion of political solidarity, understood as the reciprocal acceptance of difference, to merge Bourdieu's theory of culture and the new institutionalism in political analysis as an alternative logic of exclusion and inclusion (March and Olsen 1989, 1995; Bourdieu 1990; Swartz 1997). This may help to demonstrate that although political authority has traditionally been interpreted as resembling the legitimate domination of the state, it could be extended into the social as a condition for establishing relations of self- and co-governance within the various fields of power-knowledge constituting a society. If political authority is understood as conditioning access to solving common problems, it becomes evident that the ideal of gender equality is dependent on the transformative capacity of authority, that is, on power (Giddens 1979, 1992). What is of importance here is the fact that analyses of the inclusion or exclusion of the ideals and discursive practices of gender equality do not obey the logics of the rational calculation of utilities and the internalisation of a common good. They rely more on specifications of the power conditions that enable or hinder the articulation of personal and collective differences. Hence, new concepts and ideals are needed that do not trace gender equality to some comprehensive doctrine of 'order', 'rationality', 'normative integration', or 'class struggle' but simply to the mutual acceptance and recognition of difference (Butler and Scott 1992).
The social contextualisation of gender relations and the introduction of a cultural framework for different societies, as understood in sections 5.2 and 5.3, thus require a new notion of political authority as a guarantor of difference, contingency and change. Political authority is effective only when people accept it and consider themselves bound by it (Easton 1953). However, its acceptance need not be univocal; nor does its social bonding require a comprehensive normative agreement. It is a type of power-knowledge that allows for continuous disagreement and struggle, as a continuous presence in the social, which may appear in multiple, irreducible forms (Bang 1998).
The multiplicity of relatively autonomous forms made available by authority indicates why the governance and integration of gender differences should not be narrowed down to a matter of combining 'strong steering' and 'thick solidarity'. However, this is often what happens in gender analysis. There still is a tendency to envelop gender problems in the more or less hidden agenda of the opposition between the state as the playground of the dynamic, the rational and the powerful, and civil society as the social embodiment of the stable, the understanding and the good (Keane 1991;
132    Gender divisions of power
Cohen and Arato 1994; Hirst 1994; Etzioni 1995; Held 1996). Perhaps what is really wrong is the hidden presupposition of much gender analysis that just as women must grow hair on their breasts, if they are to gain influence over the big games that rational men play within the sphere of the state, so men must be soft and harmony-seeking when partaking in the solving of the small issues of everyday life in civil society. This is, of course, just another version of essentialism. Rather, we find it more appropriate to speak of gender within a framework of continuous battles and disputes over the distribution of valued things (Bourdieu and Passeron 1997). In such a framework, gender inequality is not merely a matter of sealing the life-world off from power, or alternatively of meeting 'the system' with counter-power and counter-rationality. It is much more a matter of exclusion - of particular meanings, of certain identities, of special discourses, of specific resources. Exclusions and so on are not the sign of'distorted communication' or 'social anomalies'. They are on the contrary a constitutive part of the social, making inclusion possible. Exclusion manifests lack, or the privileging, of certain modes of self- and co-governance in the social, which may both enable and constrain and lead to freedom as well as to domination.
Hence, we approach gender problems as a matter of political inclusion and exclusion rather than as a problem of decommodification, depaternal-isation or empowerment. We adopt an anti-essentialist stance towards all comprehensive gender doctrines, whether of a biological, functionalist, evolutionist, statist, or 'society centrist' nature. However, we do not deny the relevance of analytical theory. We merely hold that there are only contingent foundations and that any theoretical foundation has a political aspect which makes it crucial to ask what it is that a theory authorises by its foundation and, as a consequence, what it excludes. The continuous presence of the political in the social necessarily implies that gender relations could always have been shaped otherwise - even if their conditioning by culture and structure often make them peculiarly resistant towards change (De Certeau 1997; Swartz 1997; Gunnell 1998). We believe that this contingent foundationalism requires at least three methodological breaks with modern gender analysis:
1      The first break is with the tendency to oppose order and disorder, space and time, or the static and the dynamic.
2      The second break is with the actor-structure dichotomy and its many echoing oppositions such as those between rational man and irrational society, preference calculations and integrative norms, self-interest and the common good.
3      The third break is with the dichotomisation of power and knowledge, and thereby of authority and democracy.
We will briefly deal with these in turn.
Gender and European welfare states    133
Social capital - beyond the dichotomisation of order and disorder
The state-civil society opposition lays out many snares for gender analysis. One of the most basic is the tendency to empty the social of all political tensions and conflict. By perceiving unity, homogeneity and consensus as the constituting elements of civil society, the focus is removed from its constitutive political aspects: the handling of differences, diversity and dispute in everyday life (Connolly 1991, 1995; Mouffe 1992, 1993).
Robert D. Putnam, whose notion of social capital has been widely acclaimed, puts this well. 'Political participation,' he writes, 'refers to our relations with political institutions. Social capital refers to our relations with one another' (Putnam 1995: 665). Gender relations and social interactions are here simply defined as non-political. They become political only when they are politicised as claims upon the state which 'affect the prospects for effective, responsive government' (Putnam 1993: 16). The political has in this conception nothing to do with the social 'as such' which is why we must distinguish 'social trust - trust in other people - and political trust - trust in political authorities' (Putnam 1995: 665). There is nothing political outside the sphere of governmental power and the formal institutions of the state.
For Putnam, gender interactions appear as non-political and as important solely for the building of social capital. To the extent that one can speak of political capital, it is something that derives from the support for appropriate political authority by active citizens and social movements. A virtuous social order, the argument is, requires that the state is effective and responsive to the needs for equilibrating relations in the social. Hierarchy, political coercion and other forms of asymmetries in political domination are acceptable as long as they guarantee progress and enjoy legitimacy, and as long as the building of social capital is made independent from them. Only by keeping the political out of the social will 'virtuous circles result in social equilibrium with high levels of cooperation, trust, reciprocity, civic engagement, and collective well-being. These traits define the civic community' (Putnam 1993: 177). If the political is allowed to intrude on the social, social capital will erode. For 'the «ncivic community is also self-reinforcing. Defection, distrust, shirking, exploitation, isolation, disorder, and stagnation intensify one another in a suffocating miasma of vicious circles' (ibid.).
In Putnam's frame of reference, democratic political authority functions as an instrument of rational choice and as a medium for social order preventing such vicious circles from occurring. Like Talcott Parsons, Putnam tends to take it for granted that the functional problem for political authority is the 'minimizing of disruptive behavior and the motivation to it' (Parsons 1951: 30). This presumption leads both authors to put primary stress on the legitimation of authority as a binding or appropriate norm, deriving from the autonomy of the social. However, authority is not linked to the problem of social or normative order but to the problem of how to distribute values for a certain field or domain. It is a type of communicative relationship
134    Gender divisions of power
between a sender (A) and a receiver (B). It occurs (1) when B receives an explicit message from A; (2) when B then accepts A's message as the basis of decision and action; and (3) when B's grounds for doing so are the practical recognition that messages received in this way must or ought to be obeyed without evaluating the merits of the proposed conduct in the light of one's own comprehensive doctrines or normative judgments (Easton 1955: 28-29). Hence, the opposite of authority is not disorder but communication breakdowns, meaninglessness, and patternlessness.
It is important for gender analysis to recognise that if the terms 'order' and 'stability' are to have any sensible meaning they must represent a condition in which the rate of disorder and change of authority relations is slow enough to create no special communicative problems due to such disorder and change. There will always be some disorder and change. Hence, the basic problem of authority is not whether it is stable, harmonious or consensual but what is excluded and included by it. Authority does not and cannot guarantee the separation of order from disorder and thereby the discovery or the imagination of a protected locus where the 'Ideal of Enlightenment' prevails. But it does communicate that in principle and in the imagination, at least, things could always have been done otherwise.
The Third Way - beyond the actor-structure dichotomy
The actor-structure dichotomy is often called into being 'in the light of a rationalized life-world in which system imperatives clash with independent communication structures' (Habermas 1989:391). This methodological clash is supposed to derive from the study of the opposed ideologies of liberalism and socialism, both of which tend to insulate the individual from the community as an actor-structure duality. The liberals tend to see the social as an essentially private association where interests are calculated and pursued by individuals independently of, and prior to, the concerns of the social community. The state here appears as a public instrument of aggregating individual interests and of protecting their free realisation in the private sphere. The socialists, in contrast, tend to conceive of the social as much more strongly communal in content, insisting that social bonds are prior to individual interests, manifesting the 'real' interests of classes in the capitalist economy. In this view, the state appears as a class agency for integrating the social in terms of an ideology of the common good, hindering the objective interests of workers from taking effect within the social community.
Jürgen Habermas, Gesta Esping-Andersen and Birte Sum alike tend to conceive of their models as a 'third way' between liberalism and socialism, the practical embodiment of which is the (Nordic) welfare state. Most famously, Anthony Giddens has put this together in terms of a political programme, which in some respects has furnished an ideology for the centre-left in its response to liberalism, particularly in Britain (Giddens 1979). The
Gender and European welfare states    135
developed welfare state, their joint argument seems to be, sets us all free as a public agent of emancipation, linking together concerns for individual autonomy in the marketplace with the requirements for social solidarity in the civic culture. Thus the problem of order is made into a methodological problem of actors versus structures. The discussion of gender is situated within the 'difference between steering problems and problems of mutual understanding ... between systemic disequilibria and lifeworld pathologies, between disturbances of material reproduction and deficiencies in the symbolic reproduction of the lifeworld' (Habermas 1987: 363). The instrumental rationality of actors is claimed to be necessary to release both men and women from the fixities of tradition, safeguard their human rights, and enlarge their economic opportunities. However, it has to be constrained by 'the abstraction of universal lifeworld structures from the particular configurations of totalities of forms of life that arise only as plural' (Habermas 1987: 344). The communicative logic of the life-world is the medium for integrating gender differences with the common good, bringing about a peaceful socialisation of the capitalist economy, reducing anarchic competition and economic waste, and protecting 'the weak', in particular women, against the 'excesses' of industrialist capitalism (see Esping-Andersen 1990).
The problem shared by such 'third way' social democratic perspectives is that it makes little sense to speak of either a political individual or a political community in these models because political community is considered opposed to the individual and because this community is identified with either the state or the family. Furthermore, the state appears as deriving from something outside the political, namely from the clash between private and social interests in society. This apolitical stance, indicative of most modern actor-structure methodologies, seems to derive from an exclusion of the political from the domain of the theoretical. It blocks any understanding of the political as an aspect of the social, composed of a set of regularised practices in the duality of structures (Easton 1953; Giddens 1979; Bourdieu 1990).
We believe new methodologies of gender are needed to put theory and epistemology in their proper relationship. We must accept that 'any form of unity, articulation, and hierarchization that may exist between the various regions and levels [of society] will be the result of a contingent and pragmatic construction, and not an essential connection that can be recognized' (Laclau 1990: 186; Connolly 1991). The social sciences, like all other metapractices, stand in a distinctive discursive space established by society as a set of regularised practices. We see outlines of such an alternative methodology in both sociological and political research. In sociology, it is most prominent, perhaps, in Bourdieu's cultural theory of action which asks, as we did earlier in this chapter, 'How do regular patterns of conduct occur over time without being the product either of some abstract external structure or of subjective intentions?' (Swartz 1997: 95). The argument is that both individual dispositions and social structures are shaped in conventional fields of social
136    Gender divisions of power
interaction and that it is therefore the discursive construction of the social as 'habitus' and 'field' which reveals how the social can exist both inside and outside of individuals, both in their minds and in 'things' (ibid.). Individual subjectivity and societal objectivity, in such a conception, appear as properties of their mutual interpenetrations in practice. In this understanding the argument that social structures have subjective consequences is not incompatible with saying that the social world is constructed by individual actors.
We experience the same movement from individual preferences and social structures to regularised practices in the new institutionalism political research. The new institutionalists 'begin with the belief that political and administrative phenomena can not be adequately described, explained or predicted without considering the structuring quality and the independent effects of institutions (the meso-level)' (Olsen 1992: 7). This belief reflects an 'empirically based prejudice, an assertion that what we observe in the world is inconsistent with the ways in which contemporary theories ask us to think, that the organization of political life makes a difference' (ibid.: 1). On such a practical conception 'political democracy depends not only on economic and social conditions but also on the design of political institutions' (March and Olsen 1989: 17). The latter shows the articulation of conventional practices within a given field as 'collections of standard operating procedures and structures that define and defend values, norms, interests, identities, and beliefs' (ibid.).
We would suggest that the theory of culture and the new institutionalism are combined together in a new approach to politics and policy, which identifies the political not with government or the state but rather with practices of governance. We understand the political as fields or networks of relatively autonomous practices or institutions for authoritatively allocating values. This indicates how disputes and struggles over value differences can cross all established boundaries between public and private, national and international, and state and civil society. Such disputes and struggles constitute subjects through exclusions establishing a domain of de-authorised subjects erased from view (Butler and Scott 1992: 13). As such they also guarantee that these excluded and de-authorised subjects in their practices are the embodiments of political potentials for continuous rearticulations of authority relations.
When one gives up the old metapractices of defining the individual and the political in terms of the private and the state, it becomes possible to put forward the argument that the personal and the public - far from being antithetical to each other - are, in fact, articulated with each other politically. To sum up, the debate on gender through notions of emancipation, social capital, the caring state and the like simply models the political after the modern opposition between 'rational man' and 'normative society'. This opposition will most likely create more problems than solutions when studying how gender equality can be created, sustained and expanded. It does not
Gender and European welfare states    137
only prevent us from discussing the difference between a commonality which is oriented towards the good and the attainment of mutual interests (social solidarity) and one which is oriented towards rights and the solving of common concerns (political solidarity). It also directs our attention away from critically reflecting upon the tendencies in the modern state to make experts rather than 'ordinary' citizens the meta-principle of democracy and of the authoritative allocation of society's valued things. In contrast, our notion of democratically regularised practices as grounded in the mutual acceptance and recognition of difference allows for analyses of democratic political communities which are 'weak' enough to make room for the powers of both elites and lay-actors, and 'thin' enough to provide space for their various conceptions of the good (Rawls 1995, 1997).
Beyond the dichotomy of power and freedom
The idea of equality of the sexes seems intrinsically related to the outline of a political democratic practice in which 'I would accept and recognize that you make a real and significant difference, if you could accept that I do so as well' (Bang and Dyrberg 2000). Such interpersonal relations, meanings and symbols in the social are more directly based on experiences with, and images of, self- and co-governance than on exchange relations or shared mutual obligations (Strong 1994). They establish political rather than social 'habits of the heart' which spring less from trust in authorities than from trust in the capacities of lay-actors for political decision and action. The importance of such political 'capital accumulation' for democracy simply disappears from sight within the framework of modern social science. This neglect of authoritative mechanisms of inclusion and exclusion hangs together with a failure to appreciate the problem of order in (or from) disorder and the play of political individuality and commonality in the social, as discussed above. It seems important to employ the idea of the political as the making and implementing of difference, and hence to bring authority back in as the twin pair of power-knowledge in the social. The political, in this conception, is a condition of both hierarchy and self- and co-governance. We see remarkable convergences between the theory of culture and the new institutionalism with regard to these issues. Factually, they both encourage examinations of how hierarchy and domination can persist and reproduce from one generation to the next without powerful resistance and conscious acknowledgement on the part of lay-actors. Normatively, they lead us to ask how lay-actors can be related to authorities and other experts in ways that allow for self- and co-governance in both directions. Bourdieu's sociology reveals the social world as one of conflict and struggle overvalued resources and definitions that are hierarchically ordered. This notion of the social as a web of interweaving fields of struggle over various kinds of valued resources should be of considerable interest to the new institutionalists. For they are brought up within a tradition according to which 'political science is reaching towards
138    Gender divisions of power
an understanding of the very things that men consider most vital: their differences over ... the authoritative allocation of values' (Easton 1953: 50, italics added). This tradition, where politics appears as the 'interweaving of power and the authoritative allocation of values' (ibid.: 133), seems very much in line with a sociological tradition which 'does not oppose one value judgment to another but takes account of the fact that the reference to a value hierarchy is obj ectively inscribed in practices and in particular inscribed in the struggle over this hierarchy itself and is expressed in the antagonistic value judgments' (Bourdieu in Swartz 1997: 25).
The kind of power-knowledge intrinsic to the operation of political authority in the social thus indicates that gender should be studied as regularised practices that have reflection over, and involvement in, politics and policy as their content. Authority should be regarded as a communicated message, which applies to getting things expressed and done in practice. It is a means of settling disputes and struggles over the distribution of values. Authority thus indicates that whereas the acceptance and recognition of political difference are intrinsic to the balancing of uneven gender relations stability and consensus are not. The prospects for gender equality are far more dependent upon unceasing dispute, pluralisation and change than upon reaching normative agreement and consensus. Gender relations are neither logically nor practically connected with the problem of social order; they do not concern the employment of state power as either an instrument of preference aggregation or a medium of normative integration; and they do not reveal the consequences that administrative domination may have for economic effectiveness and social support. They principally concern the ability to carve out spaces of self- and co-governance in the various contexts of everyday life. This ability to convert ideas and resources into political identities is exercised in and through the institutions or regularised practices that are the issue of political life, no matter whether they operate 'below', 'above' or 'within' the state. It manifests a socio-political ontology of potentials enabling persons and groups to construct meaningful accounts of what they themselves and others are doing and to employ these for the exclusion of what are regarded unacceptable alternatives, identities or meanings.
Acknowledgement
Many thanks to Simon Duncan for valuable comments on an earlier draft.
Notes
1 In Hirdman's approach, in contrast to that taken here, culture and social order are not conceptualised as relatively autonomous and the concept is based more on the assumption of coherence. Gender inequality is seen always as a structural basis of the 'gender system', whereas here we leave space for equalisation processes. The term 'arrangement' is also preferred here to that of 'contract': the idea of a societal contract was developed in theoretical thinking and is based
Gender and European welfare states    139
on the assumption of free and equal citizens entering the social arena; the 'arrangement' approach is a more open sociological and analytical concept which leaves more space for the analysis of social and gender inequality.
2   For criticism of the idea about economic rationality in the labour market behaviour of women see Duncan and Edwards (1999). Women also may use the political and institutional framework of the welfare state in new ways which deviate from the original aims of these policies.
3   This does not mean, however, that the dividing line between traditional and more equal orientations to the question of gender equality is identical with the split between women and men, for there are different social groups with differing interests and orientations among both sexes. See also Mósesdóttir (1995).
References
Anttonen, A. 1997. The welfare state and social citizenship. In Kauppinen, K. and
Gordon, T. (eds) Unresolved Dilemmas: Women, Work and the Family in the United States,
Europe and the former Soviet Union, pp. 9-32. Ashgate, Aldershot. Bang, H.P. 1998. David Easton's postmodern images. Political Theory, 26, 3, 281-316. Bang, H.P. and Dyrberg, T.B. 2000. Governance, self-representation and the
democratic imagination. In Saward, M. (ed.) Democratic Innovations. Routledge,
London. Bang, H.P. and Sarensen, E. 1999. The Everyday Maker: a new challenge to democratic
governance. Administrative Theory and Praxis, 21, 3, 325-342. Bang, H.P, Dyrberg, T.B. and Hansen, A.D. 1997. Elite eller folkestyre - demokrati
fra oven eller neden.fournal Grus, 51, 5-31. Beck, U. 1994. Reflexive Modernisation. Polity Press, Cambridge. Becker, G.S. 1993. Nobel Lecture: The economic way of looking at behaviour. Political
Economy, 101,3,385-409. Berg, A.M. 1989. Part-time employment - a response to economic crisis? In Rosenberg,
S. (ed.) The State and the Labor Market, pp. 221-231. Plenum Press, New York. Blau, F.D. and Ferber, M.A. 1986. The Economics of Women, Men, and Work. Prentice-Hall, New Jersey. Bourdieu, P. 1990. The Logic of Practice. Polity Press, Cambridge. Bourdieu, P. and Passeron, J.-C. 1997. Reproduction in Education, Society and Culture.
Sage, London. Bourdieu, P. et al. 1991. The Craft of Sociology - Epistemological Preliminaries. Walter de
Gruyter, Berlin. Brooks, A. 1997. Postfeminisms - Feminism, Cultural Theory and Cultural Forms. Routledge,
London and New York. Bussemaker, J. and van Kersbergen, K. 1994. Gender and welfare states: some
theoretical reflections. In Sainsbury, D. (ed.) Gendering Welfare States, pp. 8-25.
Sage, London. Butler, J. 1990. Gender Trouble. Routledge, New York. Butler, J. and Scott, J.W. (eds) 1992. Feminists Theorise the Political. Routledge, New
York. Cohen, J.L. and Arato, A. 1994. Civil Society and Political Theory. MIT Press, Cambridge. Connell, R. 1987. Gender and Power: Society, the Person and Sexual Politics. Polity Press,
Cambridge. Connolly, W.E. 1991. Identity/Difference. Cornell University Press, Ithaca.
140    Gender divisions of power
Connolly, W.E. 1995. The Ethos of Pluralization. University of Minnesota Press,
Minneapolis. Crompton, R. 1998. Explaining women's employment patterns: "orientations to work"
revisited. British Journal ofSociology, 49, 118-136. De Certeau, M. 1997. The Capture of Speech and other Political Writings. University of
Minnesota Press, Minneapolis. Duncan, S.S. and Edwards, R. 1997. Lone mothers and paid work: rational economic
man or gendered moral rationalities? Feminist Economics, 3, 2, 29-61. Duncan, S.S. and Edwards, R. 1999. Lone Mothers, Paid Work and Gendered Moral
Rationalities. Macmillan, London. Easton, D. 1953. The Political System. University of Chicago Press, Chicago. Easton, D. 1955. A theoretical approach to authority. Office of Naval Research, Report
17, 1-59. Esping-Andersen, G. 1990. The Three Worlds of Welfare Capitalism. Polity Press,
Cambridge. Etzioni, A. (ed.) 1995. New Communitarian Thinking. University of Virginia Press,
Virginia. European Commission Network on Childcare 1996./I Review of Services for Young Children
in the European Union, 1990-1995. Brussels. Eurostat 1991. Earnings — Industry and Services. Luxembourg. Foucault, M. 1980. Power/Knowledge: Selected Interviews and other Writings 1972-1977.
Pantheon Books, New York. Giddens, A. 1979. Central Problems in Social Theory. Macmillan, London. Giddens, A. 1992. The Transformation of Intimacy. Stanford University Press, Stanford. Ginn, J. and Arber, S. 1998. How does part-time work lead to low pension income? In
O'Reilly.J. and Fagan, C. (eds) Part-time Prospects, pp. 156-174. Routledge, London. Gunnell, J.D. 1998. The Orders of Discourse, Philosophy, Social Science, and Politics. Rowman
& Littlefield Publishers, Inc., Lanham. Gustafsson, S. andjacobsson, R. 1985. Trends in female labour force participation in
Sweden. Journal of Labour Economics, 3, 256-274. Haavio-Mannila, E. 1985. The state, the family and the position of women in the
Nordic countries and Poland. In Risto, A. et al. (eds) Small States in Comparative
Perspective. Oslo. Habermas, J. 1987. The Philosophical Discourse of Modernity. Polity Press, Cambridge. Habermas, J. 1989. The Theory of'Communicative Action,Vol. 2. Polity Press, Cambridge. Halier, M. and Höllinger, F. 1994. Female employment and the change of gender
roles: the confluctual relationship between participation and attitudes in
international comparison. International Sociology (Journal of the International Sociological
Association), 9, 87-112. Heiligers, P. 1992. Gender and changing perspectives of labour and care. Paper presented at
the First European Conference of Sociology, August 1992. Held, D. 1996. Models of Democracy, 2nd edn. Polity Press, Cambridge. Hemes, H.M. 1987. Welfare State and Women Power: Essays in State Feminism. Norwegian
University Press, Oslo. Hirdman, Y. 1988. Genussystemet - reflexioner kring kvinnors socialaunderordning.
Kvinnovetenskaplig Tidskrifl, 3, 49-63. Hirdman, Y. 1990. Genussystemet. In Statens Offentliga Utredningar, Demokrati och
Makt i Sverige. SOU, Stockholm. Hirst, P. 1994. Associative Democracy: New Forms of Economic and Social Governance. Polity
Press, Cambridge.
Gender and European welfare states    141
Hoist, E. 2000. Die Stille Reserve am Arbeitsmarkt. Grösse - Zusammen - Setzung -
Vethalten. Sigma, Berlin. Honnegger, C. and Heintz, B. (eds) 1981. Listen der Ohnmacht: Zur Sozialgeschichte
weiblicher Widerstandsformen. Suhrkamp, Frankfurt am Main. Ishwaran, K. 1959. Family Life in The Netherlands. The Hague. Jensen, Per H. 1996. Komparative Velfaerdssystemer: Kvinders reproduktionsstrategier mellem
familien, velfaerdsstaten og arbejdsmarkedet. Nyt fra Samfundsvidenskaberne,
Copenhagen. Kamerman, S.B. and Kahn, A.J. 1981. Child Care, Family Benefits, and Working Parents.
Columbia University Press, New York. Kaufmann, F.X. 1995. Zukunft der Familie im Vereinten Deutschland. Schriftenreihe des
Bundeskanzleramtes, Vol. 16. Beck, München. Keane, J. 1991. Democracy and the Idea of the Left. In McLellan, D. and Sayers, S.
(eds) Socialism and Democracy, pp. 6-17. Macmillan, Basingstoke. Knijn, T. 1994a. Social dilemmas in images of motherhood in the Netherlands. European
Journal of Women's Studies, 1, 183-206. Knijn, T. 1994b. Fish without bikes: revision of the Dutch welfare state and its
consequences for the (in)dependence of single mothers. Social Politics, 2, 83-105. Laclau, E. 1990. New Reflection on the Revolution of our Time. Verso, London. Leira, A. 1992. Welfare States and Working Mothers. Cambridge University Press,
Cambridge. Lister, R. 1995. Dilemmas in engendering citizenship. Economy and Society, 24, 1-40. Lorini, M. 1975. Trent'anni di lotte delle lavoratrici italiane. Editrice Sindacale Italiana,
Proposte No. 33, Roma. March, J.G. and Olsen, J.P. 1989. Rediscovering Institutions. Free Press, New York. March, J.G. and Olsen, J.P. 1995. Democratic Governance. Free Press, New York. Millar, J. and Warman, A. 1996. Family Obligations in Europe. Family Policy Studies
Centre 7, London. Mincer, J. 1985. Intercountry comparisons of labor force trends and of related
developments. Journal of Labor Economics, 3, 3, 1-32. Mósesdóttir, L. 1995. The state and the egalitarian, ecclesiastical and liberal regimes
of gender relations. British Journal ofSociology, 46, 4, 623-642. Moss, P. 1990. Bernepasning i de Europteiske Fallesskaber 1985-1990. Kommissionen for
de europaeiske faellesskaber. Mouffe, C. (ed.) 1992. Dimensions of Radical Democracy. Verso, London. Mouffe, C. 1993. The Return of the Political. Verso, London. Nätti,J. 1993. Temporary employment in the Nordic countries: a "trap" or a "bridge"?
Work, Employment and Society, 7, 3, 451-464. Nickel, H. 1995. Frauen im Umbruch der Gesellschaft: Die zweifache Transformation
in Deutschland und ihre ambivalenten Folgen. Aus Politik und Zeitgeschichte: Beilage
zur Wochenzeitung das Parlament, B 36-37/95, 1, September, 23-33. OECD 1988, 1996, 1998. Employment Outlook. OECD, Paris. Olsen, J.P. 1992. Analyzing institutional dynamics. LOS-senter Notat, No. 14, 1-38. Orloff, A.S. 1993. Gender and the social rights of citizenship. American Sociological
Review, 58, 303-328. Parsons, T. 1951. The Social System. Free Press, New York. Pfau-Effinger, B. 1994. The gender contract and part-time paid work by women - a
comparative perspective. Environment and Planning A, 26, 8, 1355-1376.
142    Gender divisions of power
Pfau-Effinger, B. 1996. Analyse internationaler Differenzen in der Erwerbsbeteiligung von Frauen - theoretischer Rahmen und empirische Ergebnisse. Kölner Zeitschrift für Soziologie und Sozialpsychologie, 48, 3, 462-492.
Pfau-Effinger, B. 1998a. Gender cultures and the gender arrangement - a theoretical framework for cross-national gender research. Innovation, 11,2, 147-166.
Pfau-Effinger, B. 1998b. Culture or structure as explanations for differences in part-time work in Germany, Finland and the Netherlands? In Fagan, C. and O'Reilly, J. (eds) Part-time Perspectives, pp. 177-198. Routledge, London.
Pfau-Effinger, B. 1999. Defizite der Theoriebildung zu den Grenzen wohlfahrtsstaatlicher Geschlechterpolitik. In Hradil, S. (ed.) Verhandlungen des 29. Kongresses der Deutschen Gesellschaß für Soziologie, des 16. Österreichischen Kongresses für Soziologie und des 11. Schweizerischen Kongresses fúr Soziologie "Grenzenlose Gesellschaß?" in Freiburg, Bd. 1, pp. 203-218. Sentaurus, Plenumsveranstaltungen, Freiburg.
Pfau-Effinger, B. 2000. Kultur und Frauenerwerbstätigkeit im europäischen Vergleich. Leske und Budrich, Opladen.
Pfau-Effinger, B. and Geissler, B. 1992. Institutioneller und soziokultureller Kontext der Entscheidung verheirateter Frauen für Teilzeitarbeit - Ein Beitrag zur Soziologie des Erwerbsverhaltens. Mitteilungen aus der Arbeitsmarkt- und Berufsforschung, 25, 3, 358-370.
Plantenga, J. 1996. For women only? The rise of part-time work in the Netherlands. Social Politics, 4, 3, 57-71.
Pringle, K. 1998. Children and Social Welfare in Europe. Open University Press, Buckingham.
Putnam, R. 1993. Making Democracy Work. Princeton University Press, New Jersey.
Putnam, R. 1995. Tuning in, tuning out: the strange disappearance of social capital in America. Political Science and Politics, 28, 4, 664-683.
RawlsJ. 1995. Reply to Habermas.Journa/ of Philosophy, 42, 3, 132-180.
Rawls,J. 1997. The idea of public reason revisited. University of Chicago Law Review, 64, 3, 765-807.
Schama, S. 1988. The Embarrassment of Riches. An Interpretation of Dutch Culture in the Golden Age. University of California Press, Berkeley, CA.
Sum, B. 1994a. Engendering democracy - the interplay between citizenship and political participation. Social Politics, 1, 3, 286-306.
Sum, B. 1994b. Kan, medborgerskab og politisk kultur. In Andersen, J. and Torpe, L. (eds) Demokrati og politisk kultur, pp. 125-159. Systime, Herning.
Siim, B. 1999. Towards a gender sensitive framework for citizenship. In Bussemaker, J. (ed.) Citizenship and Transition of European Welfare States, pp. 85-100. Routledge, London.
Strong, T.B. 1994. Jean-facques Rousseau: the Politics of the Ordinary. Sage, London.
Swartz, D. 1997. Culture and Power. University of Chicago Press, Chicago.
Veil, M. 1996. Zwischen Wunsch und Wirklichkeit: Frauen im Sozialstaat. Ein Ländervergleich zwischen Frankreich, Schweden und Deutschland. Aus Politik und Zeitgeschichte: Beilage zur Wochenzeitung das Parlament, 29-38.
Walby, S. 1990. Theorizing Patriarchy. Blackwell, Oxford.
Walby, S. 1994. Is citizenship gendered? Sociology, 28, 2, 379-395.
Yeandle, S. 1999. Gender contracts, welfare systems and "non-standard working": diversity and change in Denmark, France, Germany, Italy and the U.K. In Felstead, A. and Jewson, N. (eds) Global Trends in Flexible Labour, pp. 95-118. Macmillan, Basingstoke.
6    Gender, migration and social inequalities
The dilemmas of European citizenship
Marina Calloni and Helma Lutz
6.1 Issues: gender, contracts and citizenship
The main interest of this book is to reconsider the interaction between economy, society and culture from a gender perspective and to both describe and explain differences in how this occurs in the European Union. The aim of our chapter is to refocus this interest and to shed light on it from a different angle. By taking into account the position of migrant individuals and groups we want to indicate the persistence and the increase of new forms of social inequality and gender discrimination in the European Union.
In particular, in the present section we will reframe the social, political and ideological theory of the 'gender contract' and the structure of national constitutions, by including the perspective of'foreigners', that is non-citizens, and migrant women among them in particular. In the second section we will consider the influence of migration on the constitution of the European Union. For European unification has transformed previous definitions of migration, and of who have to be considered migrants. (Hence the 'old migration' refers to that before 1989, when the EEC became the EU; see section 6.2.) In fact, since 1989 migrants in the EU have been considered only as those individuals who are not citizens of a European member state. (The 'new migration' is thus after 1989; see section 6.2.) Migration, however, cannot be considered to be culturally and economically homogeneous. For this reason, in section 6.3, we will discuss the phenomenon of the 'feminisation' of migration. This is of fundamental importance for understanding the structure of gender roles in a trans-cultural way and the emergence of new forms of social inequalities within each European nation state. The case of domestic workers, who are mostly employed by indigenous women will be taken as an example. Finally, taking as background the new situation,of migrants in Europe, and the increase in new kinds of discrimination (including that between women), in section 6.4 we will try to indicate some perspectives which can help us to conceptualise European citizenship in a new way. This should not be based any longer on the principle of the ius sanguinis (literally the 'right of the blood'), but should be understood as a negotiable and flexible political and cultural construct, open to 'new workers and citizens'.
We will introduce our argument by questioning the theory of 'gender