The use of the concept "maternalism" is one of the most visible signs of American feminist historians growing interest in integrating a gender perspective in welfare-state research. But whether the scholars who use this concept have a common notion of gender, is difficult to say. Scholars generally tend to understand and use the concept very differently. The following quotes demonstrate clearly that maternalism has become a fertile soil for conflicting interpretations.
By maternalism historians have meant the female version of paternalism, the assumptions women reformers made about women's nature, and the policy strategies they devised to provide social protection for women's maternal responsibilities
-Kathryn Kish Sklar 
What makes maternalism more than just a women's paternalism, however, is its rootedness in the subordination of women.
-Linda Gordon 
Even though this divergent use has not stopped American scholars from using the concept of maternalism in their welfare-state analysis, it is important to clarify its meaning. How is the concept used and for what purposes, and why are American scholars defining maternalism differently? What importance does the concept have for attempts to analyze the early welfare state?
In order to answer these questions, I will analyze the ways in which American historians define and use maternalism. It is important to emphasize that in this Chapter, I aim to explore some aspects of the debate rather than to even out the various views defended or to resolve the disagreements between them. However, I hope to set the various definitions in context by presenting the scholars more broadly in terms of publications and major ideas. It seems natural to start with Theda Skocpol because she is the one that most thoroughly has promoted the necessity of employing gender in social political analysis. Moreover, Skocpol claims that use of gender also leads to a rewriting of welfare-state history in general. Using Skocpol's ideas as a starting point, I present and analyze the debate that has taken off between American scholars. The analysis developed in this Chapter will be a point of departure for the last Chapter when I discuss the relationship between maternalism as a historical concept and gender as a category of historical analysis. The second part of Chapter three contains an examination of the possible consequences of using maternalism as a historical concept.
It is necessary to see maternalism, which is a theoretical term, in its historical context: the development of welfare states in the period 1890s - 1920s. The end of the nineteenth century was a historical turning point for many countries in the Western World. New ideas about the role of the state alongside new social needs due to the development of an industrial society resulted in increased state activity and commitment to social welfare. Political mobilization and organization in society also led to growing pressure on the state to form alternatives to stigmatizing poor reliefs. However, a characteristic feature of the emerging social security systems in the Western World was its private-public mix. Even though the state initiated new social reforms, it still relied on and co-operated with private welfare producers, such as philanthropic and voluntary organizations.
In the 1880s, Germany was the first country to introduce compulsory national insurance against sickness, accident, disability and old-age. Sweden and France followed soon with a voluntary sickness insurance in 1891 and 1898, respectively (see Table 1-3, page 29-32). While the German social insurance system originally was designed for industrial workers and workers in trade and transport, it was already in 1911 extended to cover white-collar workers. However, most of the welfare programs that emerged in the period 1890s -1920s, were aimed at needy people with low incomes. Both the Norwegian Sickness Act of 1909, and the British National Insurance Act two years later, were designed for blue- and white-collar workers above a certain income level. The labor question was also central when France in the same period began to form their social security programs. Although the French insurance programs were voluntary in the beginning, they soon came to accompany the international trend towards compulsory social programs.
Social welfare and the role of the state was also on the political agenda in the United States (see Table 2, page 30-31). But until well into the twentieth century, the US government did not undertake elaborate social welfare programs. This means that the Civil War Pension of 1862 for long was the only federal welfare program, consuming 34% of the federal budget, and that the US was a social political "laggard" compared to many European countries. On the other hand, 38 states enacted Workmen's compensation laws between 1911 and 1919. Compared to Europe, the federal government in the United States came to play a limited role in promoting social welfare in this period. It was not until the Great Depression in the 1930s that the US government came up with a contributory insurance program for retired wage earners and their dependants.
USA: -a welfare state "late-comer"?
Since the United States never developed a national health insurance program, like many European countries did in the twentieth century, the American welfare state is often viewed as "less" developed than the welfare state systems of Europe. The USA can besides be described as a "laggard" on the road towards the welfare state, due to the fact that the United States in a European perspective also was the last country to develop a social security system. To present the United States as a "welfare state late-comer" indicates an emphasis on the social political development in relation to only a few social programs on a national level. The German historian Gisela Bock describes this as follows:
The reforms by which the development of welfare states is usually studied and which are compared with each other normally relate to a limited range of "social problems": sickness, disablement, old age, unemployment, labour protection and the introduction of progressive income taxes. This literature deals at best marginally with politics to do with motherhood, fatherhood, childbearing, and childraising.
Bock indicates by this that welfare policies not aimed at the (male) wage earner have often been left out of the historical picture painted by welfare-state scholars. Scholars' focus on men and the problems of members of the male labor force may also be a result of the fact that the labor question was a central theme at this time. In particular, the countries where the early welfare policies were based on social insurance arrangements emphasized the wage worker on the formal labor market. However, if we look at the United States, the latter explanation does not ring true. Although a social security system based on insurance arrangements was not passed by Congress before 1935, it does not necessarily mean that the federal government or the government of the states played a minimal role in promoting social welfare before 1935. In other words, it is possible that scholars focusing on the labor market have underestimated or obscured the degree to which the United States and other countries developed social reforms aimed at other social groups such as women and children (see Table 2, page 30-31). These social reforms compose the social, political, and historical context in which maternalism operates.
social reforms aimed at women and children in the United States, 1890s -1920s
Even though the United States never developed a national health insurance program, many American states in the early twentieth century passed welfare programs that were aimed directly at women and particularly at mothers. In the period 1911-1928, 44 American states passed Mothers' Pension programs which provided cash benefits to poor single mothers and widows with children. In addition to these pension programs, many states in the same period developed protective labor laws to limit working hours and to establish a minimum wage for women workers. One of the first "Women's hour laws" was passed in Oregon in 1908, and by 1921 41 American states had laws restricting daily labor for women. Regulations for working women were justified as protection for those who were, or might become, mothers. Women and children were a group which needed protection because they were not able to protect themselves. Thus the court in Oregon justified protection of working women:
That woman's physical structure and the performance of maternal functions place her at a disadvantage in the struggle for subsistence is obvious. This is especially true when the burdens of motherhood is upon her..... 53
While both labor protective laws and Mother's Pension programs were welfare arrangements on state level, the US federal government in 1912 showed its concern for the welfare of mothers and children by establishing the National Children's Bureau as a part of the federal Department of Commerce and Labor. This publicly-financed bureau was established to collect and disseminate information concerning the welfare of children and mothers. As elsewhere in the Western World, this was motivated by response to the high infant mortality and falling birth rates. Although the Children's Bureau was a research agency, it became the first female stronghold in the federal government, mainly headed by professional women.
The activity within the Children's Bureau led to a growing federal interest in maternal and child welfare which culminated with the passing of the Sheppard-Towner Maternity and Infancy Act in 1921. The act, which provided federal funding for maternal and infant health programs, was formed in the hope of reducing the high infant mortality rates that the Children's Bureau had documented. By its federal funding, the Sheppard-Towner Act was the first federal program for social welfare in the United States. But, in contrast to Mothers' Pensions programs, it never survived the Great Depression in the 1930s. The Congress decided already in 1929 to shut down federal education programs.
Table 1: Social Policies in Germany and Britain 1890s - 1920s: Year and Type of First Reform (see note 48)
Sickness GERMANY 1883 BRITAIN 1911 Accident Unemployment compulsory insurance for compulsory insurance for Old-age pension workers, female workers manual workers, including Labor protection could obtain three weeks maternity benefits for Maternity/children maternity leave 1884 the wives of covered compulsory insurance for workers (1914 maternity workers 1914 benefit paid directly to governmental employment the mother) 1887 the provision 1889 Workmen's Compensation compulsory insurance for Act 1911 compulsory workers 1924 insurance for workers in Maternity benefits a limited number of extended to non-employed trades 1908 the Old wives of insured husbands Age Pensions Act insurance for people with low income 1909 minimum wages in selected industries, also included women in some of the lowest paid occupations 1908 the Children Act legislation concerning the treatment of children by the law in both criminal and civil cases 1918 the Maternity & Child Act encouraged development of local maternal clinics and services
Table 2: Social Policies in Norway and the United States 1890s -1920s: Year and Type of First Reform (see note 48)
NORWAY UNITED STATES
Sickness 1909 compulsory *1862 the Civil War Accident insurance for workers Pensions for Union Unemployment Old-age above a certain income soldiers who were pension Labor level, including wounded, or, if they had protection maternity benefits for died in battle, to their wives of covered workers survivors in 1915, paid directly to 1911-1911 workmen's the mother 1894 compensation laws passed compulsory insurance with in 38 states 1920-1932 state and employees' in 2 more states funding 1906 1935 Social Security Act voluntary/unions with contributory insurance state contribution program for retired wage 1936 compulsory and earners and their universal means-tested dependants 1890s system 1892 Women's hour laws passed "fabrikktilsynsloven" in 8 states 1900-09 New prohibited factory-work or improved women's hour for children under 12 laws passed in 13 states years old and prevented 1909-17 New or improved women from working in the women's hour laws passed mining-industry and from in 39 states work in general for a period of six weeks after giving birth. The law also limited working-hours for children and young people between 12 and 18.Table 2 continues
Labor protection NORWAY 1915 UNITED STATES 1912-23 Maternity/children granted small maternity Minimum-wage laws passed allowances to poor single in 15 states 1918-32 mothers 1919 Mothers' Women's hour laws in 2 pension introduced in more states; 12 states Oslo, cash benefits to makes improvements single mothers 1911-19 Mothers' pension laws passed in 40 states, cash benefits to single or widowed mothers 1923-28 Mothers' pensions in 4 more states. 1911 the Children's Bureau a research agency on child welfare 1921 the Sheppard-Towner Act provided federal funding for maternal and infant health education programs
Table 3: Social Policies in France and Sweden 1890s-1920s: Year and Type of First Reform (see note 48)
Sickness FRANCE 1898 voluntary SWEDEN 1891 voluntary Accident Unemployment insurance for workers in private insurance with Old-age pension industry and trade 1893 state funds 1901 Maternity/children free medical assistance voluntary/private to poor pregnant women insurance 1934 1911 compulsory old-age voluntary /unions and disability insurance insurance with state 1909 the Engerand Act a funds 1913 universal maternity protection law income-tested system that guaranteed jobs to 1912 compulsory maternity women who stayed away leave in six weeks (was from work up to 8 weeks not put into effect before and after giving before 1931) birth 1913 the Strauss Act prescribed a mandatory maternity leave of four weeks after childbirth and two weeks later, included domestic workers
As Table 1-3 shows, the United States was not alone in developing social welfare reforms aimed at mothers and children in the period 1890s-1920s. Maternity benefits were included in the Norwegian and the British health insurance acts from 1909 and 1911, respectively. Countries like France and Sweden also enacted laws concerning maternity, but in contrast to Norway and Britain, these programs were formed separate from the insurance programs. In other words, national variations were many, and governmental concern for the welfare of mothers and children has to be seen as an international trend in this period.
However, the American welfare system stands out in comparison to the European systems. While social reforms directed towards women and children for long made up the major part of public welfare programs in the US, maternal reforms constituted only one part of a broader social security system in the European countries. This difference is crucial in order to understand why researchers focusing on social security based on wages have seen the USA as a laggard, and why American feminist scholars are now focusing on mothering (maternalism) as a source of women's political empowerment in the Progressive Era.
maternalism: a characterization of early social policies and their initiators
Theda Skocpol uses the term "maternalist policy" to characterize the early social policies that emerged in the Progressive Era and benefited some women as mothers. Women's hour laws, minimum wage laws, Mothers' pensions, the establish of the Children's Bureau and the passing of the Sheppard-Towner Act are thus described as maternalist legislation. Moreover, Skocpol also uses the term to identify different welfare systems such as the American and the British. The latter is especially clear in her article from 1991, with co-author Gretchen Ritter.
During the period when major European nations, including Britain, were launching paternalist versions of the modern welfare state, the United States was tentatively experimenting with what might be called a maternalist welfare state.
Skocpol's characterization of the British and the American welfare regimes is derived from two major observations. Firstly she focuses on the ways in which the welfare systems are organized. Whereas the early British welfare state featured regulations and benefits for workers and low-income dependent people, direct ties were established between public authorities and women as workers, mothers or widows in the US. Even though also Britain ended up offering support to mothers and widows, women's social rights were mainly decided by their ties to wage earning males. Thus the principle of direct and indirect channelling of social benefits to women is the ground on which Skocpol builds her sharp distinction between maternalist and paternalist welfare policy.
PATERNALIST social benefits indirectly to women as MATERNALIST social benefits directly to dependants of male worker women as workers, single mothers and widows
Besides being devised by male politicians, bureaucrats, and trade unionists, paternalist measures such as those that dominated British social policy during the early 1900s attempted to shore up the working condition of all workers in ways that reinforced male trade unions, and attempted to channel public benefits to women and children through male wage-earning capacities. In contrast, early US labor regulations were not only devised and implemented primarily by female professionals and women's groups, they also applied directly to women.
Secondly, Skocpol emphasizes the political processes and political actors that initiated and shaped the welfare state development. She says:
In Britain, male bureaucrats and party leaders designed policies "for the good" of the male wage-workers and their dependants. Meanwhile, in the United States, early social policies were championed by elite middle-class women "for the good" of less privileged women.
These two last quotes indicate that Skocpol defines maternalist policies as policies formulated by women for women in particular. Thus the term "maternalist" also becomes a label for the women who promoted maternalist welfare policies. Organizations that pushed for a maternalist line of development, according to Skocpol, are the National Congress of Mothers (1897), General Federation of Women's Clubs (1890), the National Consumers' League (1899), and the National Women's Trade Union League (1903). After 1912 the maternalist line was also influenced by the Children's Bureau. Paternalist policies are, in contrast to maternalist policies, identified as policies articulated by men for men and their dependants.
Skocpol's use of maternalism, as an opposition to paternalism, has been strongly supported by the historian Kathryn Kish Sklar. In an article from 1993, Sklar maintains that:
By maternalim historians have meant the female version of paternalism, the assumptions women reformers made about women's nature, and the policy strategies they devised to provide social protection for women's maternal responsibilities. 
But Sklar fails (as Theda Skocpol fails) to give a more exact specification of the term.
criticism of previous welfare-state research
In the article from 1991, Skocpol points out that the distinctiveness of American social policy prior to the New Deal, was its maternalist character. Even though the maternalist line dominated the shape of the American welfare state, Skocpol does not preclude the existence of several lines of welfare development in the post-Civil War period. In her recent book Protecting Soldiers and Mothers (1992) she claims that there was at the same time a paternalist as well as a maternalist line of welfare development in the United States. However, the paternalist line never became influential because the country never developed a strong labor movement. The Workmen's Compensation was from Skocpol's point of view the only victory of the paternalist approach (see Table 2, page 30). But, says Skocpol, this provision was an exception.
With a focus on the "maternalist" character of the American welfare system prior the New Deal, Skocpol argues that the United States did not join the mainstream evolution of western European social progress. Accordingly, the USA was not a welfare-state "laggard", argues Skocpol. Such a statement implies that the American experience is exceptional and it revives the old interest in American exceptionalism. However, in contrast to previous research on this issue, Skocpol does not use "American exeptionalism" meaning the country's uniquely strong liberal values. Social policies in this period were unique because they were created and worked out for the benefit of women, argues Skocpol. Social progress in the US was identified with the strength of gender and not with the fortunes of the working class, as in Western European countries.
Previous comparative research on the development of welfare states has primarily relied upon theoretical approaches stressing the effects of socio-economic modernization, national values and ideologies, or demands by working class organizations. Interpretations stressing the strength of the working class and political parties have marked our understanding of the welfare state development. Skocpol, on the other hand, explains that theories stressing the strength of labor are insufficient to explain social policies aimed at mothers and female workers as opposed to industrial workers and their dependants.
Gender identities and relationships are simply not treated as analytically central in theories that derive political conflicts and outcome straightforwardly from balances of power between capitalists and organized wage earners. 
The political forces shaping the early patterns of social provision in the US were not grounded in conflicts between capitalists and industrial workers but in ethnic and gender identities. This, in Skocpol's view, made the American experience exceptional.
In order to find explanations that take the American maternalist experience into account, Skocpol argues in her latest book that we have to find an explanatory approach which differs from existing theories of welfare state development. Existing theories are insufficient because they all share problematic assumptions about the evolutionary nature of the welfare state and the socio-economic roots of political processes. In arguing for an alternative approach, she says:
Yet only by taking processes of state formation and patterns of political organization seriously, and notice that these intersect in varied ways with economic and social transformations, can we break with the progressive notion of social policies as aspects of societal evolution. 
Skocpol wants to explore how social and political factors combine to affect the social identities and group capacities involved in policy making. By viewing the polity as the primary locus for action, we are able to understand political activities, whether they are carried out by politicians or social groups, as conditioned by the institutional configurations of governments and political parties.
Skocpol's polity-centered analysis is a continuation of her previous state-centered approach, which excluded those political forces she now tries to integrate; grassroots-activism, non-institutional political activity, voluntary social work etc. Thus "polity" becomes a broader term than "the state". By asking why maternalist forces promoting social policies for mothers and women workers were more effective than paternalist forces in US politics, she turns to the structure of the polity, the "fit" between the organizational capacities of maternalist and paternalist forces, and the opportunities offered by US political institutions.
Generally speaking, Skocpol criticizes previous theories for ignoring gender dimensions of politics which not only means that they overlook social policies targeted on mothers and women workers but that they also "fail to notice the contributions of female-dominated modes of politics, some of which are not dependent on action through parties, elections, trade unions, or official bureaucrats". She implies that unless we can bring the politics of gender into our comparative analysis, we will not be able to explain patterns of politics and policy that were especially important in the US.
Skocpol's interpretation of early social policy stands in contrast to previous welfare-state scholarship. Her rethinking can be summarized by the three following elements: first, the United States was not a welfare state "latecomer" because the country already in the early twentieth century formed social welfare programs aimed at women and children. Second, these welfare programs were initiated and shaped by female reformers, who by their maternalist activities attended to the interests of all women. Third, this means that gender, and not class, became the major force in the making of the American welfare state.
This intention to rewrite the histories of the early welfare states raises many complex questions with regard to feminist scholarship of the welfare state. Especially Skocpol's attempt to present women as a unified political force by defining all women as potential mothers, has caused debate among feminist scholars. The US historian Linda Gordon in particular has presented a thorough criticism of Theda Skocpol's latest book.
Linda Gordon, one of the leading feminist historians of the nineteenth and twentieth centuries America, did already in the 1980s what Theda Skocpol is trying to do in the 1990s: employing the category "gender" in the analysis of the American welfare state. Like Skocpol, Gordon also stresses the importance of organized women as promoters and initiators of maternal and child welfare in the Progressive Era. This indicates that both Skocpol and Gordon have moved away from the patriarchal paradigm. In spite of their common emphasis on women as political actors and the possibilities that might represent, they tend to interpret maternalist policies quite differently. Actually, as we will see, Linda Gordon disagrees in many respect with Skocpol's dichotomous use of maternalism. By looking into the debate that developed between Skocpol and Gordon in Contention (1993), I hope to find out why they understand maternalist policies differently.
maternalist policies and the concept of "power"
Whereas Skocpol argues that the maternalist celebration of the civic value of mothering was shared "by mothers of all classes and races", Gordon claims the contrary. She says:
the maternalists were elite, dedicated to providing what they believed was good for the working class, and mainly in agreement on the necessity of shoring up the male breadwinner as head of family.
Gordon indicates that America's maternalists were captive to the country's elite and gender-traditionalism from the start. Moreover, the history of the mother's pensions, which became Aid to Families with Dependent Children (ADFC) with the Social Security Act in 1935, provides also, from Gordon's point of view, a clear evidence that women's power does not always promote all women. Even though female reformers initiated social welfare programs for women, their activity first of all reflected the interests and the values of the middle-class. Skocpol fails, according to Gordon, to see this because she neglects the concept of power. Referring directly to Protecting Soldiers and Mothers, Gordon maintains:
The book fails to create a satisfying explanation of the construction of social provision in the period because of the systematic exclusion of social-structural power relations -as class and gender- from her analysis.
While Skocpol's polity-centered claims bring social movements back into the analysis centre, they leave out social structures such as class and gender. Gordon continues:
Without any discussion of power differentiations between men and women, rich and poor, white and Black, WASPs and immigrants, the various civic organizations are reduced to pressure groups competing on a presumably level field. Moreover, her lens picks up formal organizations but not informal policy influences from social movements or shifts in popular consciousness.
To Gordon, the problems in Skocpol's interpretations are already present in the outset of the book: she fails to produce any adequate definitions of what she means by "paternalist" and "maternalist". Gordon continues: "This failure exemplifies ways in which Skocpol's approach to the influence of gender is undeveloped in relation to the theoretical level of much scholarly gender analysis today". Clearly, Gordon indicates that Skocpol's analysis is not matched by familiarity with scholarly debates on gender. Gender means "female" for Skocpol, and Gordon claims that "she produces an entirely celebratory account of the women's organizations she studies. She has no critique of maternalism".
Skocpol uses maternalism as an opposition to paternalism, without directly expressing the distinctions between the two concepts, with the exception of the structural differences mentioned above. The absence of such a specification and definition is a result of her failure to ground her concept of gender in questions of male and female power, says Gordon.
Gender is, after all, not merely a neutral or benign difference; it is a difference, or rather a set of meanings culturally constructed around sexual difference, in a context of male domination. In the entire book there is no discussion of male power in general or in its specifics -or, to put it inversely, of the fact that the forms of political power with which Skocpol is so concerned are shaped by their maleness. 
The maternalist strategy was after all a result of women's lack of political power, says Gordon, and thus the concepts of paternalism/maternalism refer to an inequity of power in relation to both gender and generation.
maternalist policies: -result of fixed differences between men and women or shared understandings of the proper family?
Identifying the influence of gender in welfare thought, we should not argue that these visions were dichotomized between men and women, claims Gordon. In an article from 1992, she emphasizes this by saying:
My purpose is not so much to distinguish male from female as it is to illustrate the importance of asking questions about gender, questions that illuminate similarity as well as difference.
The stratification of the American welfare system into the social insurance and public assistance program, often called the two-track welfare system , was, in the way Gordon sees it, a result of gender values shared by both men and women, in order to maintain the family wage system.
...male and female welfare reformers worked within substantially the same gender system, the same set of assumptions about proper family life and the proper sphere for men and women. 
By not employing gender as a male/female opposition, Gordon is able to underscore that men and women were holding similar visions of the economic structure of the proper family in which the welfare state took its form. However, while these gendered assumptions did not necessarily express antagonism between men and women, they were anything but universal: "they expressed a dominant outlook, to be sure, but one that did not fit the needs and understandings of many less privileged citizens". In other words, Gordon thinks it is false to believe that a kind of unity among women was present at this time. Women's activism was as much as men's, determined by class as much as by gender. "Specifically, this supposed unity denies that women's agency also derives from other aspects of their social position." Gordon continues:
She [Skocpol] generalizes about these "maternalists" as if they were manifestations of some universal female principle. They did share some fundamental beliefs and assumptions about proper role of government and the proper construction of families, but Skocpol identifies these commonalties no more than their differences.
By accusing Skocpol for presenting women as if they have no class identities, Gordon can be said to have taken an anti-essentialist view which became visible among US feminist scholars in the 1980s. She says: "At one time women's studies or gender scholars assumed that male/female were inevitably a binary set of opposite principles, and that women had a unique and universally similar perspective, but no longer!" Gordon tries by this to avoid the trap of false generalizations by unmasking the differences among women.
Gordon's argumentation against Skocpol corresponds in many ways with Birte Siim's criticism of Helga Hernes. Gordon is, like Siim, more critical to the idea of female alliances across social boundaries, than both Skocpol and Hernes. By stressing women's various social positions, Gordon and Siim problematize the common female experience that Skocpol and Hernes emphasize. Whereas Skocpol and Hernes see "women" as a homogenous group, different from men, Gordon and Siim stress that women's gendered identity is also necessarily determined by their socio-economic position in society. Therefore, gender becomes, in the way Gordon and Siim argue, not a category of fixed differences between men and women, but differences constructed socially and culturally.
This contrastive notion of gender explains partly why Skocpol and Gordon have different understandings of maternalist policies. Skocpol sees maternalist and paternalist policies as results of female and male political activity, respectively, whereas Gordon understands maternalist policies as a result of shared understandings of the proper family among male and female welfare reformers. According to Gordon (but contrary to Skocpol), men can also initiate and produce maternalist policies.
Linda Gordon defines the concept of "maternalism"
Although Gordon denies the idea of any kind of unity between middle-class and working-class women, she still implyes that such a unity should exist. By characterizing social insurance programs for workers as superior to public assistance programs for women and children, Gordon is asking "Why did women design inferior programs for women?". In order to understand how women, and indeed feminist women, could design and support inferior social programs for other women, she thinks the legacy of feminism has to be considered. Historians in general, and especially feminist historians, have normally used feminism to refer only to those who struggled primarily for gender equality. A broader definition is needed, in order to include the women and men who did not believe in total gender equality but who agitated for greater respect and power for women in their proper sphere. Gordon defines feminism in this way: "Feminism is a political perspective that considers women unjustly subordinated, finds that oppression to be humanly changeable, and strategies for women's advancement." 
This definition might include women who did not call themselves feminists. Nevertheless, it enables us to describe the women reformers who worked for improvement of women's conditions within the separate female sphere, using the language of gender difference instead of gender equality, without interpreting them as less "real" or "true" feminists. Does this mean that Linda Gordon also sees maternalists as feminists?
Linda Gordon thinks, despite her critical remarks, that maternalism is a useful label for an orientation among women reformers from the mid-nineteenth century to the Progressive Era. In order to use the concept, she says, we need to specify it more than Skocpol does. Gordon's definition of maternalism contains four parts: First, maternalist policy proposals contained a conviction that women reformers should function in a motherly role towards the poor. Second, this conviction was embedded in women's belief that it was their work and experience as mothers that made women uniquely able to lead in the campaign for social public provision. Third, the arguments that maternalist feminists used to justify the need for state protection was their responsibility as mothers of future generations. Arguments like these were sometimes biological, other times social, but always based on a commitment to gender differentiation. This explains why women reformers had a mixed attitude to women's economic independence: they were supporting direct payment to women at the same time as they were refusing to support permanent or universal child benefit programs, which might have undermined male-headed households. Since many female reformers adapted the family-wage assumption to their maternalist strategy there are no reasons for defining maternalism as antithetical to paternalism. As Gordon explains:
Most of the women-dominated groups agreed with these men on the basic premise that public provision should support the family wage, i.e., the principle that men alone should be able to earn enough to support a family without help from wife or children.
Actually, argues Gordon, the policies and advocates that Skocpol calls "maternalist" can just as well fit the paternalist definition, because "the maternalists were elite, dedicated to providing what they believed was good for the working class, and mainly in agreement on the necessity of shoring up the male breadwinner as head of family ". However, to Gordon maternalism was not "women's paternalism" as Kathryn Kish Sklar expressed it. Maternalism is in contrast to paternalism rooted in the subordination of women. Gordon expresses the fourth aspect of maternalism as follows: "Maternalism showed its standpoint -its view from underneath- and from there built a strategy for using the space inside a male-dominated society for an activism that partially subverted male power".
maternalist policies and the concept of "race"
Gordon argues that the faith in the family wage, in addition to being a social construction, also was a racial pattern, far more characteristic of white than black women in the early 1900s. By focusing explicitly on differences in welfare visions among black and white women activists, Gordon has shown that black women to a greater extent than white women, accepted married women's employment as a long term and widespread necessity. Although black women activists differed from whites in their approach to women's economic role, they can still cautiously be described as maternalists, says Gordon, because they stressed the importance of motherhood in the same way as their white counterparts.
The racial aspect of maternalism has also been highlighted by the political scientist Gwendolyn Mink. She claimed already in 1991 that America's maternalists were captive to the country's racism, because they linked the problem of racial order to the material and cultural quality of motherhood. In a more explicit way than Gordon, Mink has been occupied with maternalist policies and racism. She says:
Women's politics was moreover a racial politics, tying the future of the republic to uplift of the citizenry. In the main, this politics was directed towards the new immigrant population -the eastern and southern Europeans who moved into northern cities during the late nineteenth and early twentieth centuries.
Theda Skocpol is concerned with race in another meaning than Gordon and Mink. While Gordon and Mink emphasize maternalist policies with regard to respectively racial differences and racism, Skocpol highlights the ways in which maternalists used "race" rhetorically, in their welfare campaigns. In general Skocpol criticizes feminist scholars for overstating their sense of race and race anxiety in the Progressive Era by referring directly to Mink. Skocpol says:
Mink traces social policies for mothers in this period to a vague, overarching sense of "race anxiety". She pays little attention to variations. And much of her evidence consist of quotations using the word "race". Many quotes from female reformers and women's groups are taken out of context, and Mink often misunderstands the use of the word "race". Sometimes in this historical period, it was, as she suggests, used to refer to whites versus nonwhites, or northwerstern versus southeastern Europeans; but at other times it was used to refer to the human race, or all citizens, or all humankind. Women reformers often used the word in this sense when they spoke of mothers' responsibility for furthering the well-being of the race.
Skocpol's argumentation shows that she, in contrast to both Gordon and Mink, chooses to interpret women's reformers use of the word "race" in two ways, both in situations where it was used in the meaning of all human beings and where it was used in the meaning whites versus people of color. This understanding of race is matching her celebratory account of the maternalist organizations. Mink, on the other hand, says that she is using the term "race", "as it figured in the politics of industrializing America. It refers to people of color and to white people we commonly refer to today as "ethnics" ". Gordon, on her side, employs race in the meaning black/white, African-Americans versus Anglo-Americans. The major point with Gordon's emphasis on race was to show that maternalism was a highly racial construction, formulated mainly by white middle class women. On the other hand, Skocpol does not take this aspect into account.
As the discussion shows, the concept "race" invites different interpretations of maternalist policies. The debate also shows how sensitive race, as a theme, is in the United States, and how academic scholarship is influenced by this sensitiveness. Skocpol's attempt to present maternalist policies and the maternalist women as an inclusive phenomenon with regard to social classes and races, can be said to be a political reckless project in a multi-cultural country like the USA. Even though the racial feature of maternalist policies only composes one part of maternalism, it becomes a crucial one in the discussion between Gordon and Skocpol.
In the following we will see that the racial aspect is not explicit among the remaining definers and users of maternalism. While Skocpol and Gordon in general went into details in their discussion, the remaining scholars such as Seth Koven and Sonya Michel have a less conscious stand to the questions raised in Contention. This does not mean that they do not deal with the scholarship of Skocpol and Gordon. Both Koven and Michel refer to Skocpol and Gordon in their footnotes and biographies. Still, it is important to point out that they do not take critical stand on Skocpol's and Gordon's definition (or lack of definition) of maternalism.
The two historians Seth Koven and Sonya Michel are perhaps the scholars, besides Theda Skocpol and Linda Gordon, that most thoroughly have tried to employ the concept of maternalism in their analyses of women's welfare activism. Already in 1990 they worked out a definition of maternalism that in many ways became style-setting for the new comparative scholarship on gender and the origins of welfare states. Their central role within the scholarship was largely confirmed in 1993, when they edited a collection of essays written on the topic of maternalist politics and welfare state formation. The book Mothers of a New World contains articles from scholars such as Kathryn Kish Sklar, Molly Ladd-Taylor as well as from Koven and Michel themselves. As we will see in this part, Koven and Michel are more ideologically concerned than Theda Skocpol. They intend to see maternalism as an ideological trend in the early twentieth century. How do they suceed?
maternalism as political discourses about women, the state and society
Koven and Michel focus on individual and organized women who campaigned for the improvement of maternal and child welfare. By being "maternalist" in orientation, these women initiated welfare programs that to some degree came to serve as models for the ways in which central authorities handled the welfare needs of women and children. Through the example of women's welfare agency in Europe and the US, Koven and Michel introduce maternalism in order to explain the ideological grounds on which women's reform activity were built.
In an article co-authored with Ruth Rosen, Seth Koven refers to maternalism "as a political concept that accepts the principle of gender difference, especially women's identity as mothers, but maintains that women have a responsibility to apply their domestic and familial values to society as a whole". In the introduction of Mothers of a New World, Seth Koven and Sonya Michel define maternalism as follows:
We apply the term to ideologies that exalted women's capacity to mother and applied to society as a whole the values they attached to that role: care, nurturance, and morality. Maternalism was the central and defining core of some women's vision of themselves and of politics.
Such specifications emphasize that Koven and Michel perceive maternalism as a political discourse that operated in relation to other political discourses and in relation to a wide array of concrete social and political practices. Both of them deny, however that maternalism was a unified movement speaking with one voice. They say: "Maternalism does not refer to a specific movement per se. Nor do we use the noun "maternalist" as we might "feminist", to establish identity of a particular social actor". Feminists and not-feminists, radical, liberal or conservative, pro-and anti-suffrage: the capacious umbrella of maternalism gathered them all, argue Koven and Michel.
Not referring to a specific movement or identity, Koven and Michel's definition accepts that visions of motherhood and maternal roles may vary over time and place in relation to the social and political locations of activists and reformers. Even though Koven and Michel call attention to the changeable character of maternalism, they also stress that the maternalist rhetoric served different interests for men and women: "Although male politicians used maternalist rhetoric, it was often merely a cloak for paternalism." This attempt to limit maternalism by referring to its contrast is quite similar to the way Theda Skocpol contrasts paternalism with maternalism. Contrary to other feminist historians who have highlighted the repressive use of politics and rhetoric of motherhood by male social-reform activists, Koven and Michel along with Skocpol, try to illustrate the opportunities that maternalism offered women. They both criticize feminist historians who consider motherhood and maternalism incompatible with female emancipation. From their point of view this incompatibility has led some feminist historians to downplay women's influence on the formation of welfare states. 
maternalist policies - between private and public
Koven and Michel also criticize traditional historians for minimizing the role of women in welfare state formation. Because historians in general have underestimated the role of the voluntary organizations, they have also failed to notice the role of women. Historical studies of welfare-state formation have from their point of view:
failed to look closely at those places where women were most influential: in their localities as elected and appointed officials and as leaders and rank-and-file members of voluntary societies that addressed every conceivable social problem. In their zeal to trace origins of central government, they have minimized the role of local government, and hence of women as well.
Because voluntary organizations have merely been viewed as private training grounds for men who wanted to make careers in public office, historians have in general failed to notice that voluntary organizations had different meanings for men and women. Since few or no women were able to follow the path of careers in public office and since women also lacked the rights to vote at this time, the activity within voluntary organizations became the major arena in which they could engage in public life and social policy.
It was within the maternalist paradigm, closely linked to the traditional female sphere, that women first claimed new roles for themselves by stressing the importance of motherhood for society as a whole. Koven and Michel are claiming that women reformers challenged the boundaries between public and private by transforming motherhood from women's private responsibility into public policy. However, at the same time as they challenged the distinctions between public and private, men and women, state and civil society, they also evoked traditional images of womanliness. This apparent paradox of entering the public political arena by reinforcing the traditional female sphere of children, family, care, and nurture, is what maternalism is all about, argue Koven and Michel.
The public/private distinction is especially addressed by Seth Koven in her contribution to Mothers of a New World. Maternalism as practices and discourses involves interactions between private and public in a way that led Koven to underscore the artificiality of constructing private and public as two bipolar categories. She stresses that women moved between the private and the public in their roles as social reformers, activists, workers and consumers of welfare.
radical and conservative maternalists
Even though there were no common political strategy among women reformers besides their concern for improving maternal and child welfare, Sonya Michel argues that maternalists were divided into two general factions on the question of the state's role in protecting and aiding women and children. When some women fought for greater federal responsibility in social welfare, others insisted that maternal and child welfare should be the concern of private and philanthropic organizations and not the state. Historians have from Michel's point of view tended to focus on the radical maternalists that supported an extension of state intervention. These radical women aimed their activity at the government, and operated within the settlement movement, the National Consumers' League and the Children's Bureau. Michel calls for more attention to their counterparts, the conservative reformers
who were committed to the protection of mothers and children, but whose conservative convictions inclined them to take another political path -one that did not necessarily lead to or through government. 
The conservative fraction of the maternalist movement gathered reformers which asserted that the role of the federal government should remain minimal. With this distinction between radicals and conservatives, Sonya Michel stresses the complexities of maternalist ideologies and the numerous forms they could take. She says: "As a powerful set of metaphors, maternalism was taken up by the activists across the political spectrum".
By dividing the maternalist paradigm into two major fractions, the radical and the conservative, Michel tends to see maternalism more as a specific movement. She also uses the label "maternalists" as a characterization of women reformers who pressed for maternal and child welfare reforms. This use appears as a paradox compared with her definition in the introduction of Mothers of a New World . "Maternalism does not refer to a specific movement per se. Nor do we use the noun "maternalist" as we might "feminist", to establish identity of a particular actor".
the connection between women's power and the strength of the state
Women's maternalist policies was, according to Koven and Michel, strongest in the nations that had the weakest, least bureaucratic state. They say: "the strength and the range of women's private-sector welfare activities often varied inversely with the strength of the state". The American state and, to a lesser degree, the British state are from their point of view, "weak" states because they relied on local and private forms of welfare provision, in lack of a strong centralized government. Nations with strong centralized goverments and well-developed welfare bureaucracies, such as France and Germany, provided, on the other hand, less space for women's welfare activism to develop. Despite a weak women's movement, the French and the German state regimes resulted in more social welfare programs for women and children than the "weak-states" (see Table 1-3, page 29-32). Especially France developed many social protection laws for working women. This means that a strong women's social-action movement did not necessarily result in more benefits for women and children: "in the first group of four countries [USA and Britain] those where women's social-action movements were strongest granted the least generous state-welfare benefits for women and children before 1920". This line of reasoning is contrary to the ways in which Theda Skocpol presents the US as the "first maternalist welfare state".
The "strong-state/weak-state" paradigm created by Koven and Michel has resulted in discussion within the new feminist scholarship on the welfare state. Theda Skocpol is among the scholars who has commented on Koven and Michel's argumentation about the connections between the power of women's welfare agency and the strength of the state. Skocpol says:
I do not fully agree with Koven and Michel's analysis. Their contrast between "weak" and "strong" states is too crude to get at the differences among national political systems that affected how likely women are to become politically active and
(a separate issue) in what ways women can have an impact on policy decisions. In particular, Koven and Michel fail to analyze crucial differences between the ways class and gender identities figured in the social politics of Britain and the United States between 1870s and the 1920s. Although both of these nations had "weak states" in Koven and Michel's terms, they actually had very different governmental institutions, administrative systems, and electoral and political party systems. 
In line with these critical remarks on Koven and Michel's scholarship, Skocpol is in general hostile to their way of approaching the history of welfare-state formation. But Skocpol expresses clearly that she appreciates Koven and Michel's comparative approach because it shows that
gender is not just a relation of social domination or social inequality, as the patriarchal theories emphasized. Female gender identities-which are not all the same, and which change over time -can also be sources of social solidarity, organization, and moral purpose. 
Although Skocpol mentions the constructive elements of female identities, women's common identity as "Women" becomes the most important aspect for Skocpol.
Koven and Michel employ maternalism generally as a political ideological concept in order to describe and explain women's social political activity, based on their identity and capacity as mothers. Maternalism becomes in this manner not manifested in a particular movement or organization per se. Michel tends, however, to use maternalism more as a characterization when she distinguishes between radical and conservative maternalists, and locates them within specific organizations or networks. In the following we will see that the historian Molly Ladd-Taylor has brought Michel's specification of maternalism as a movement even further.
maternalism as a social movement
Molly Ladd-Taylor is another US historian who has recently developed yet another definition of maternalism based on her research on women's child welfare activism in early twentieth-century US. In contrast to Koven and Michel's wide definition, Molly Ladd-Taylor has a more specific use of the term. In her latest book Mother-Work, Ladd-Taylor expresses explicit dissatisfaction with the way feminist historians have used the term "maternalism". While scholars such as Theda Skocpol, Seth Koven, Sonya Michel employ the term to describe practically any woman who used the language of motherhood to justify her political activities, Ladd-Taylor thinks that "such a general use conflates very different ideologies and types of organizing that relied on the rhetoric of motherhood."
In order to distinguish women activists who used the language of motherhood, Ladd-Taylor calls for a more precise definition of maternalism. She uses the term maternalism
to denote a specific ideology whose adherents hold (1) that there is a uniquely feminine value system based on care and nurturance; (2) that mothers perform a service to the state by raising citizenworkers; (3) that women are united across class, race, and nation by their common capacity for motherhood and therefore share a responsibility for the world's children; and (4) that ideally men should earn a family wage to support their "dependent" wives and children at home.
Even though this definition can refer to a wide range of political perspectives including the women who worked for women's suffrage, Ladd-Taylor argues that she does not believe that maternalists can properly be called feminists. The distinction between feminism and maternalism is crucial, according to Ladd-Taylor, because maternalists were wedded to an ideology rooted in the nineteenth-century doctrine of separate spheres and to a presumption of women's economic and social dependence on men. On the other hand, feminists stressed female individuality, political participation, and economic independence. Ladd-Taylor explains that both feminists and maternalists used the language of motherhood in the 1910s, but that their argumentation had different meanings. Whereas maternalists argued for protection of women within their traditional family role, feminists used the rhetoric of motherhood to improve women's status and to criticize the male-headed family. Moreover, feminists were, contrary to maternalists, relatively unconcerned with social welfare reforms.
According to Ladd-Taylor's definition, only leaders of the National Congress of Mothers and members of the Hull House/Children's Bureau network can be called maternalists. However, she does not treat these women under one general category "maternalists". Women within the Mother's Congress (later the Parent-Teachers' Associations) are in Ladd-Taylor's work characterized as sentimental maternalists because they were convinced that women's highest calling was marriage and childbearing. Their traditionalist thinking was clearly visible in their negative attitude towards extending government aid to working-class or wage-earning women. These women therefore limited their social welfare visions to single or widowed mothers and their children. From Ladd-Taylor's point of view, the sentimental maternalists were principally responsible for the passing of mothers' pensions legislation in the 1910s. Women within the settlement movement and the Children's Bureau saw maternal and child welfare reform as "a step towards broader government protection for male as well as female members of the working class". Ladd-Taylor characterizes them as progressive maternalists.
Through characterizing the club women and the government women as respectively sentimental and progressive maternalists, Ladd-Taylor ends up with a very specific definition of maternalism. She also operates with a sharp distinction between maternalism and feminism. The term "maternalist" refers in the way she uses it to a particular social actor who is not a feminist. Feminism and maternalism coexisted, however, in the 1910s, until the debate over the effects of the Equal Rights Amendment (ERA) in the 1920s brought about fundamental differences between the two, argues Ladd-Taylor.
Feminists associated with the National Woman's Party made individual women's right to equal opportunity in the public sphere a political priority, while their maternalist opponents strove first of all to protect women in their family role. 
According to Ladd-Taylor, pro-ERA arguments depended on the language of gender sameness and equality while the maternalists argued against the ERA by emphasizing women's difference from men. Thus Ladd-Taylor's distinction between feminism and maternalism is based on the opposition between the language of gender equality versus the language of gender difference.
maternalism as a paradigm for feminist ideology
Contrary to Ladd-Taylor's distinction between maternalism and feminism, Ann Taylor Allen has in her historical work on feminism and motherhood, defined maternalism as a paradigm for feminist ideology and action. She says that maternalism is
a feminism that takes woman's experience as mother and nurturer as the basis for interpretations of women's history, for distinctively female approaches to ethical and social questions, and for improvements in women's status.
Maternal feminism as a feminist ideology based its claims on women's difference from men, and not on the essential similarity between the two. According to Allen, maternal feminism or maternalism represents what the historian Karen Offen calls "relational" forms of feminism because these women stressed social contribution rather than individual rights. Whereas historians such as Molly Ladd-Taylor and others present equal-rights feminism and maternalism as conflicting, Allen claims to the contrary that these two forms of feminist ideology coexisted and interacted because
feminists in the nineteenth and early twentieth centuries did not find them contradictory or mutually exclusive, but would use arguments based both on similarity and on difference as the context and opportunity dictated. 
Motherhood has always been central in feminist discourses, says Allen.
"Motherhood" -private, public, biological, and social- was the centre of a feminist discourse that, although constantly developing, was also continuous from the first feminist writings in the late eighteenth and early nineteenth centuries until the twentieth century.
Especially the idea of motherhood as a basis for a specifically female ethic, provided a standpoint for feminist criticism and activism in the late nineteenth and early twentieth centuries. This standpoint was often expressed symbolically by the mother-child bond which "became the basis of a concept of social morality that linked the self to the other and the individual to the community".
The rhetorical use of motherhood shows, from Allen's point of view, that the social construction of gender was indeed connected to the bodily realities of sex and reproduction. However, this does not mean that the feminist standpoint was a product of a biological function and thus common for all women. The idea of motherhood as a source of ethical authority, was a cultural construction derived from the socio-economic conditions of the elite.
The ideology that made motherhood a basis of empowerment was chosen, not determined, in response to a specific historical situation, among a specific group of women. For them, it provided a way to shape their experience by devising a symbolic framework on which to understand it. Feminist ideas of motherhood were based on a familial culture that originated within the upper and middle-classes. The ascendancy of this ideology resulted from the ascendancy of middle-class women within the feminist movement.
This indicates that Allen views maternalism as a "bourgeois" product, built upon the experience of elite and middle class women within a specific historical situation. She continues: "The practice of social motherhood through education and philanthropy supported the wider claim of the middle class to cultural hegemony over the lower classes".
While historians of feminism often have identified equal-rights ideology as more progressive or more "real" feminism than maternalism, it is connected to the ways in which historians of feminism tend to approach and judge the theme. Allen says: "Historians of women have more often judged the figures to be discussed here according to present-day feminist ideology based upon equal-rights theory and the questioning of all gender differences". In other words, by evaluating previous generations of feminists on present-day ideologies, we tend to marginalize aspects of women's work and experience, especially in regard to women's role as mothers. The present-minded approach to the history of feminism can partly explain why maternalism for long has been neglected by feminist historians, argues Allen. "Therefore, we must reconstruct the intellectual, social, and political contexts in which feminists spoke in order to understand their meanings".
The numerous definitions and uses of maternalism and maternalist policy presented above, confirm the difficulty in stating anything general about these concepts. Nevertheless, the analysis enables me to summarize three major considerations about maternalism and maternalist policy, common to all disputing scholars.
Maternalist policies are first of all used as a characterization of maternal and child welfare policies that emerged in the early twentieth century, and which channelled social benefits directly to women as mothers. Second, the scholars also use maternalist policy to define the ways in which women reformers mobilized and sought power in order to improve the welfare of women and children. By being cut off from the ballot and political parties, women sought alternative means of political action outside the institutionalized political system. Put differently, women's exclusion from the formal political system forged political styles different from and in opposition to male politics. This alternative political activity found expression in the emergence of voluntarism politics such as the creation of voluntary associations, lobbyism, parades, and mass-meetings, together with practical charity work such as the foundation of social settlements. Third, the ideological principle for female reformers claiming political influence and power was, according to the debating scholars, grounded in a celebration of women's difference from men, a difference based on women's experience and responsibility as mothers, as nurturers, and as carriers of morality in society as a whole. Mothering became in this way the ideological backbone of women's welfare activism. The use of the ideology of motherhood to legitimize women's political activity, in times when political rights were reserved for men, thus is what these scholars define as maternalism. As maternalism was challenging the idea of separate spheres for men and women, female reformers reinforced the notion that women were essentially different from men and thus especially gifted for specific functions in society. By applying the idea that women possess special knowledge or moral qualities by virtue of being mothers, female reformers made mothering and reproduction the fundamental defining experience of womanhood.
Even though all scholars agree about mothering and nurturing as the basic components of gender difference, they still end up with contrastive definitions of the concept "maternalism". In the debate between Linda Gordon and Theda Skocpol we saw that different explanations of maternalism are due to different notions of gender.
Skocpol's use of gender as fixed differences between men and women explains why she contrasts maternalism to paternalism. By arguing that only women could hold maternalist visions of welfare politics, she indicates that women's social political activism was shaped by their essential femaleness. Thus mothering, nurturing, and caring become female "attributes", common to all women. This indication is also promoted by Seth Koven and Sonya Michel: "Although male politicians used maternalist rhetoric, it was often merely a cloak for paternalism". Such a statement tells us indirectly that male politicians have different intentions of politics than women by virtue of their maleness, and vice versa. Therefore Koven and Michel use gender as fixed or essential differences between men and women. Molly Ladd-Taylor, however, does not define maternalism by referring to its male opposition. But she confirms the notion of women's essential difference from men by defining maternalism as a uniquely feminine value system based on care and nurture, shared by all women. Thus Ladd-Taylor's notion of gender is solely female-centered.
Linda Gordon, on the other hand, understands gender as meanings culturally constructed around perceived differences between the sexes. The tendency to define women as different from men on the basis of their capacities as mothers and nurturers is, according to Gordon, a social and cultural construction, grounded on dominant social definitions of maleness and femaleness. This notion of gender does not preclude that also men could take up maternalist visions of politics. Ann Taylor Allen also follows this line of reasoning when she interprets female reformers focusing on women's experience as mother and nurturer as a cultural product. Nevertheless, the emphasis on the social and cultural construction of gender, does not preclude Allen from being concerned with bodily realities of sex and reproduction. This connection between biological sex and gender is of interest in the next, and last Chapter, where I highlight the relationship between maternalism as a concept and the category of gender. The major aim in Chapter three, however, is to present and discuss possible consequences of the use of maternalism as a historical concept.