The Widows of the Gasworks:
Gendered Path Dependency and the Early Dutch
Welfare State
Marian Van Der Klein
Abstract: In this article the focus is on the gender implications of the early Dutch welfare state and the status of male and female breadwinners and their dependents in social insurance and poor relief in the Netherlands around 1900. Barbara Nelson’s model of two gendered channels (workmen’s compensation acts versus mothers’ aid laws) in early U.S. social policy serves as a point of reference. The author concludes that, compared to the United States, social policy in the Netherlands did not start as extremely gendered. The assumption of Dutch second-wave feminists that the Dutch welfare state was male breadwinner–minded from the very beginning does not hold water.
At a time when many European historians appear to be retreating into detailed analyses of small-scale, local issues, Anglo-American feminist experts on the early welfare state continue to cherish big ideas. In the last decade, Theda Skocpol, Sonya Michel, Seth Koven, Linda Gordon, Denise Riley, and Barbara Nelson have all bravely written grand narratives concerning gender and the development of social policy (Skocpol 1992; Koven and Michel 1993; Gordon 1990; [End Page 1] Riley 1988; Nelson 1990). Their ideas are well conceived, inspiring, and appealing. Maternalism, male and female visions of welfare, and discourse analysis based on rights and needs have become intrinsic elements in international thinking on and interpretation of social politics.
It is very tempting for a young European scholar such as myself to incorporate those big ideas in my research on the history of gender and the dawning of the Dutch welfare state. 1 This article, for example, draws inspiration primarily from Barbara Nelson’s (1990) classic article on the development of two channels in the early U.S. welfare state. Nelson compares the first workmen’s compensation acts (WCAs) with the mothers’ aid laws (MALs) passed in various states. The WCAs were designed for white men employed in heavy industry, such as metal-working, mining, and railroads, whereas the main beneficiaries of MALs were envisioned as white widows with children. The former could claim rights on the basis of their paid labor, but the latter were dependent on favors bestowed on them as caregivers. Although the mostly male victims of industrial accidents were treated as autonomous, independent, individual breadwinners, the mothers were seen as parts of families and treated in a condescending, paternalistic manner. According to Nelson, these small beginnings helped determine the ultimate shape of the American welfare state. In other words, she postulates a gendered path dependency.
Nelson’s pioneering study received wide acclaim, both in the United States and beyond (I first encountered it in a reference made by Bussemaker [1991]). The model she sketched was so insightful and suggestive that quite a few European social scientists began to discover two channels at the origins of their welfare states as well. I wanted to see whether a similar pattern could be discerned in the history of the Netherlands. 2 I was particularly interested in seeing whether the kind of gendered path dependency she sketches out for the United States could account for the primacy of the male breadwinner principle in postwar Dutch society. 3
Nelson’s model was very appealing to second-wave Dutch feminist scholars seeking to advance women’s economic (in)dependence (Knijn 1994, 83–86). In their view, since the late nineteenth century the state had gently but firmly pushed women toward the sink without giving them any rights there, whereas the men who went out to work each day were well taken care of (see, for example, Sevenhuijsen 1978). Thus they were inclined to assume that the welfare state had been oriented toward the male breadwinner from the outset. In this article I will investigate this assumption more closely, 4 focusing on the gender implications of early Dutch welfare state and the status of male and female breadwinners and their dependents in social insurance [End Page 2] and poor relief around 1900. I begin with a local case of three widows in Rotterdam on the eve of the Dutch Workmen’s Compensation Act of 1901. I then briefly outline Dutch legislation at the time and go on to examine its underlying notions and how such legislation was put into practice. Nelson’s model and her findings in the United States during roughly the same period will be my point of reference throughout.
The Early Dutch Welfare State:
A Brief Overview
In the Netherlands before 1900 welfare was provided by all kinds of institutions—except the state. For those who could afford to insure themselves, there was a voluntary insurance market that included mutual societies (Genabeek 1999). Factory schemes served the happy few who were lucky enough to work with a big company owned by a so-called social employer (Klein 2000). Poor relief (organized by private organizations, the church, the municipality, not by the state) was the last resort. During the nineteenth century the Dutch government took an extremely passive approach to the social security of its citizens (Genabeek 1999, 358). But around 1890, the pressures of advancing industrialization prompted the state to abandon its laissez-faire stance.
Like the Americans, the Dutch began their state social insurance system with a WCA, which passed through Parliament in 1901 and took effect in 1903 (Staatsblad van het Koninkrijk der Nederlanden 1901). The act protected workers in a number of ways: those who were injured on the job were entitled to compensation for medical costs, loss of earnings, or, in the worst-case scenario, funeral costs, with a pension being paid to their surviving dependents. The state compelled employers to insure their employees and finance the benefits. Unlike later social insurance schemes, to which workers contributed, this one placed the burden entirely on the employers.
Under the Dutch WCA, women were covered both as workers (directly) and as widows (indirectly); notably, there was no separate provision for mothers—no mothers’ aid or mothers’ pensions. (Here is one difference between the Dutch and the American situation, and it is mentioned by Theda Skocpol [1992] when she sets out her theory of a maternalistic America and a paternalistic Europe.) Before 1940, the only Dutch provision explicitly addressed to mothers was moederschapszorg (maternity care, literally translated as “care for motherhood”), which was incorporated into the Health Insurance Act of 1929. The act called for medical assistance by a doctor or a midwife at the delivery and twelve weeks’ paid maternity leave for (married) women workers. The law valued women not so much as care givers but primarily as workers. [End Page 3]
It is interesting to note that the Dutch Labor Act of 1889 already prohibited women from carrying out paid work within four weeks of giving birth. In other words, it was to take forty years before compensation was provided for this gendered loss of income. All the ILO conventions and the feminist lobby notwithstanding, the Netherlands was a laggard in this respect compared to the rest of Europe. Italy had provided maternity benefits for workers as far back as 1908 through the casa di maternità (Bock 1994). 5 In Germany, which served as a role model for both America and the Netherlands in terms of social insurance, new health insurance legislation in 1883 enabled women who had given birth to claim any medical costs involved and entitled them to financial compensation for the obligatory rest period. Nevertheless, the Netherlands was still well ahead of the United States with paid maternity leave.
An Illuminating Case
To grasp in detail the gendered implications of the early Dutch welfare state, let us turn to the case of the widows of the Rotterdam gasworks. On the morning of December 30, 1897, three men were found dead in the ammonia room of the local gasworks in Rotterdam. They had been suffocated by toxic fumes as they worked. All left widows, and one of them left young children. A discussion arose over who was to provide for the survivors. As employees of the municipality of Rotterdam, the men had paid contributions to ensure that their wives would receive a pension in the event of an industrial accident. But the local authorities tried to claim that the accident had resulted from the men’s own negligence and refused to pay the widows a pension. (“Binnenlandsch overzicht” 1898; “De ramp” 1898; Rutgers-Hoitsema 1899a,b; Hintzen 1899).
This local event sparked a national debate in which providers of poor relief, feminists, and advocates of a broader welfare system—men and women alike—took part, with the feminist and social democrat Marie Rutgers-Hoitsema playing a leading role. On the eve of passage of the WCA, here was a vivid example of how things could go wrong without a national social insurance scheme. The widows were forced to throw themselves on the mercy of the local poor relief, with its social control and moral supervision. If their husbands had died four years later, they would have been entitled to benefits, irrespective of their individual needs and issues of their husbands’ culpability.
Rutgers-Hoitsema, the “founder and soul” 6 of the Rotterdam Vereeniging tot Behartiging van de Belangen van de Vrouw (Association for the Promotion of Women’s Interests), had no doubts about [End Page 4] the rights of the matter. In 1889 she made strenuous efforts to ensure that the widows of the three workers received a pension. A. Krijgsman, P. H. van Delden, and Cornelis Kerkhoven had all worked at the factory for over ten years, and their widows, represented by one D. de Klerk, claimed a pension on the basis of a bylaw regulating such payments by the municipality of Rotterdam. 7 But the local council saw no grounds for such a pension. Claiming that the men had probably gone to sleep on the job and had “thus become the victims of dereliction of duty,” they provided supporting evidence in the form of statements made by the dead men’s colleagues. The incident and subsequent events were reported in the national press.
The periodical Sociaal Weekblad maintained that the local council’s decision conflicted with “the principle of the WCA bill” (“Binnenlandsch overzicht” 1898). Even if the workers had been at fault (a claim that many others convincingly contested), their next-of-kin’s entitlement to a pension should not have been forfeited. But the protests were to no avail; the widows did not receive a municipal pension. Ultimately, two groups concerned by the widows’ plight—the first consisting of Rotterdam councilors, the second of private citizens—set up a collection jointly amounting to some 2,800 guilders. Six months after the accident, in June 1898, the councilors submitted their donations to be managed by the Association for the Improvement of Poor Relief: 688 guilders for the youngest widow with two small children, with the two older widows receiving 234 guilders and 67 guilders, respectively. The concerned citizens provided approximately 1800 guilders, and formed a local committee, of which Rutgers-Hoitsema was a member. She then entered into a debate with H. G. Hintzen, chairman of the Vereeniging tot verbetering van Armenzorg te Rotterdam (Rotterdam Association for the Improvement of Poor Relief, hereafter the Rotterdam Association). Sociaal Weekblad published their letters: paternalism, entitlement, and need were key concepts in their polemic (Rutgers-Hoitsema 1899a,b; Hintzen 1899). 8
Needs Versus Rights?
Rutgers-Hoitsema first tried to persuade Hintzen of the need for the two funds to work together. In her view, the ideal solution was an annuity in the form of weekly payment of two guilders “for the two old ladies” and an allowance for the youngest widow until her youngest son had reached age twelve. Rutgers-Hoitsema felt that this arrangement best served the widows’ interests; she had determined from the Eigen Hulp mutual insurance society what amount would be needed and believed that the requisite sum (3,492 guilders) could be raised. Hintzen, however, did not share her view. The sum of money [End Page 5] collected was too small and, more important, the Rotterdam Association not only regarded it as its duty but also believed it to be highly necessary to supervise the poor women in question and oversee their expenditures. (He may have been influenced by the fact that the association was suffering from financial difficulties, but this was not mentioned in the public debate.)
Whatever the case, the debate ended, and the Rutgers-Hoitsema committee distributed its share of the money directly among the widows that October. The youngest widow, who, according to Rutgers-Hoitsema, was “shrewd and very industrious” (1899a, 54), thereupon set up a draper’s and grocer’s shop. But the Rotterdam Association, after having paid an initial sum of 10 × 10 guilders to each widow, slowed its payments to a trickle or cut them off entirely. This was a source of concern to Rutgers-Hoitsema, partly because her industrious protégée needed more business capital and domestic help. After she had asked various members of the ladies’ committee of the commission for visiting the poor for an explanation and was again referred to Hintzen—by now a useless course of action as far as she was concerned—she sought publicity.
In two articles headed “Recht of armbedeeling” (“Right or charity”), Rutgers-Hoitsema argued that all three widows were entitled to a pension, which the Rotterdam Association was currently treating as alms (emphasis in original). It started with the fact that the association did not inform the widows of the amount of money it had received in their name. It was only thanks to the newspapers and word of mouth that the widows knew that any financial donations had been made on their behalf. Moreover, the Association treated them like any other recipient, providing assistance only in cases of want and need as determined by a voluntary poor visitor. In the case of the widows, two were receiving occasional support totaling 27 guilders. Rutgers-Hoitsema felt that the widows were a special case and that the association should adopt an entirely different attitude, both morally and financially, toward them. The widows had not applied to the association for assistance, money had been made over to the Association in their name, and they were accordingly entitled to that sum. The association should represent the widows’ interests better, cease its personal supervision, consult with the widows as to how they wished to receive the money, and guarantee that their entitlements would be transferred to their heirs when they died (one of the widows was seriously ill).
Countering Rutgers-Hoitsema, Hintzen claimed that “the only 'entitlement’ that could be said to exist in this particular case, was that the widows were entitled to the money, but the manner in which and the times at which payments were to be made remained entirely [End Page 6] a matter for the Association” (Hintzen 1899, 83). This, he argued, had been determined by the generous donors, who had told the association that it could disburse the money as it saw fit. The rest of his arguments took a moralistic tone. He largely ignored the points raised by Rutgers-Hoitsema while continuing to stress the benefits of a highly paternalistic approach. The interested parties had not been informed of the extent of the amount, because otherwise “they would be tempted to claim instant payment of the money.” The widows did not have a right to the money; it was not a pension. Supervision was necessary because it was part of the association’s standard procedure, and the fact that it was now necessary to acknowledge that the business capital of the youngest widow was insufficient showed how necessary personal supervision was. The association “would have the business examined by an expert to establish whether it was viable.” What had he now been informed? That Mrs. Rutgers had taken it on herself “to advise the same widow to take an unmarried mother and her child in as lodgers? A singular state of affairs!”
Sociaal Weekblad does not relate how the story ended. It seems likely that the pressure of public opinion and an improvement in its financial situation—8,000 guilders of unrestricted donations had meanwhile been received—prompted the Rotterdam Association to become slightly more generous in its contributions to the widows. Nevertheless, it retained its powerful position. Rutgers-Hoitsema could do what she liked, but the association simply refused to transmit the money directly to the interested parties without further supervision.
Nelson, Gordon, Skocpol, and the Dutch Discourse
If we look at this case in the light of Nelson’s model, we can discern a parallel with the American discourse on social welfare. Hintzen, the representative of public assistance, does indeed cite need as a criterion for assistance, adopts a highly paternalistic attitude, and sees the association as providing favors rather than entitlements. Rutgers-Hoitsema, on the other hand, with her plea for the right to benefits, is very much in line with many contemporaries who urged a general system of social insurance and, more particularly, a WCA. In the Netherlands, too, a clash was taking place between proponents of the new insurance and advocates of the old poor relief system.
In the run-up to the Dutch WCA, social insurance was often contrasted with poor relief. The first was heralded as a modern solution to social issues and stood for all that was desirable, whereas the second was castigated as an old-fashioned system that had ceased to function properly. According to contemporary reports, in the late [End Page 7] nineteenth century the burden on the municipal poor relief organizations in the Netherlands was steadily increasing; in fact, they could no longer cope. The problem could be remedied by an insurance system paid for by employers or possibly even by workers (Schwitters 1991). It was high time that employers footed the bill for the consequences of accidents that occurred in their factories. They could no longer expect society to bear the burdens of industrialization. Moreover, others claimed, social insurance would put employees (and their surviving dependents) in a much stronger position in the struggle for subsistence, because they would no longer be dependent on the favors of poor relief but could rely on accrued entitlements. Instead of being needy, they would become rightful claimants, exactly as Rutgers-Hoitsema had sought to achieve in the case of the widows.
Rutgers-Hoitsema lobbied for the interests of the widows while trying to avoid paternalism. She aimed for a system of rights rather than favors, but it would be wrong to say that she paid no attention to the question of need. She believed, for instance, that the Rotterdam Association had not calculated the widows’ need for assistance carefully enough: “After looking into the matter, the Association came to the conclusion that there was no immediate need, whereas I came to the conclusion that need did exist and that the three widows occasionally suffered want” (Rutgers-Hoitsema 1899a, 54). Later she also considered the youngest widow’s situation in detail, including turnover, rent, business costs and other factors:
[At] present this woman’s hands are completely tied: she can order no yarn, because the factory requires that she order a certain number of bales or skeins at once, which she cannot afford; her supply of stockings of the most common size is exhausted and cannot be replaced; she buys the cotton to make the shirts and aprons customers order in an ordinary shop . . . and pays exactly what I would pay for it . . . all this . . . for lack of business capital. (Rutgers-Hoitsema 1899a, 54)
In the meantime, the Rotterdam Association was paying her 3 guilders a week,
and in order to collect those three guilders the widow must go every week at a stipulated time to the Association building, which is by no means nearby. To this end she must find someone and pay them to mind her shop and look after her children while she is away on this errand. . . . It is hardly surprising that the Association concluded that enterprises that she set up did not on the whole flourish. (Rutgers-Hoitsema 1899a, 54)
A discussion even started between Rutgers-Hoitsema and Hintzen about what the youngest widow actually wanted. It is interesting to [End Page 8] note that throughout this argument about her needs, the woman in question remained nameless and was never given a chance to speak for herself.
On close examination it becomes clear that Rutgers-Hoitsema does not distinguish between rights and needs in her arguments, but mingles them. In this respect, incidentally, she showed herself to be a worthy and pragmatic exponent of the approach currently advocated by American philosopher Nancy Fraser (1990). This combination of rights and needs arguments appears to have been an effective strategy in her debate with the Rotterdam Association on the eve of the Dutch WCA. In using whatever arguments came to hand on this particular issue, Rutgers-Hoitsema’s approach was similar to that uncovered by Linda Gordon (1992) in her study of male and female leaders of the early U.S. welfare movement. Although not setting out to create a gender dichotomy, Gordon does discern distinctly male and female visions of welfare. Women, according to Gordon, used “the rhetorics of rights . . . but more often in combination with the notion that needs themselves were a claim on the polity” (Gordon 1992, 32).
Rutgers-Hoitsema and the movement from which she emerged also resembled American women reformers in another way. Many U.S. women who were involved in social work supported social insurance but rarely participated in its formulation; similarly, the Dutch women’s reform movement supported the WCA but was not very much involved in its development. The difference that Gordon notes between advocates of public insurance and advocates of public assistance has to do with “assumptions about the nature of poverty and how to help the poor.” Champions of insurance (mostly men) lobbied for money and jobs. Advocates of social welfare (men and women) had a more complex and, according to Gordon, feminist view. In the debate between Hintzen and Rutgers-Hoitsema the roles appear to be reversed. She advocates insurance and calls for money and jobs, and he argues that supervision is always required in such a broad and complex context. In fact it is Hintzen who exhibits a much more traditional mentality.
Hintzen is not the only Dutchman of the period to speak in terms of need, nor is Rutgers-Hoitsema the only woman to invoke rights. To return to the debate on the WCA, many needs-related arguments may be found in the plea made by Liberal MP Eduard Fokker for giving domestic servants equal legal status. Servants had a high risk of accidents and were poorly paid (certainly as compared to their foreign counterparts), which meant that they could not be expected to insure themselves; the state should accordingly require their employers to do so on their behalf (Handelingen 1899–1900). Conversely, the feminist periodical Evolutie tended to speak in terms of rights [End Page 9] when it came to this question. I am particularly intrigued by the fact that its editors, Wilhelmina Drucker and Theodora Haver, both well-known activists, approvingly cite the following argument by Melvil van Lijnden, another Liberal MP: “We are all citizens of the same State, we all pay tax and thus enjoy equal rights. All should have accident insurance” (Evolutie 1900). As Drucker and Haver reasoned, this included servants and housewives. They thus claimed social insurance for women around 1900 on the basis of citizenship, rather than on the basis of paid labor or motherhood.
Dutch thinking in this era turns out to be somewhat more diverse than what Nelson and Gordon found in the American context. The gender dimensions of the Dutch discourse on poor relief and social insurance resemble more closely what Theda Skocpol discovered in her research on early U.S. social policy (Skocpol 1993, 157–170). She encountered female advocates of universal rights and male advocates of working men’s insurance “just as concerned as . . . female reformers to separate the 'deserving’ from the 'undeserving’ poor” (Skocpol 1993, 173). In two articles in the journal Contention, Gordon and Skocpol waged a grim battle over their characterizations of early U.S. social policy (Gordon 1993; Skocpol 1993). Skocpol accuses Gordon of often being dead wrong about the ideas, as well as the groups that lay behind social insurance writings and female welfare activism, and Gordon attributes their different conclusions to a differing sensitivity to power relations in terms of race, class, and gender and to a political versus a social-historical method of research.
My knowledge of U.S. history is too limited to add anything meaningful to their disagreements, but I would suggest that perhaps this debate about gendering discourses has to do with a more general problem with big ideas in the history of gender and welfare states: they almost inevitably appear to be formulated in terms of dichotomy. Nelson’s and Gordon’s theses are examples of this, but Skocpol’s own maternalistic-paternalistic model follows the same pattern. Even though Gordon (1992) warns against dichotomizing and Skocpol prefers not to bifurcate gender identities into male and female (Skocpol 1993, 168,169), their studies can be and have been interpreted as analyses of male and female as a binary set of oppositions. 9
Gender being such a peremptory category, it sometimes tends to obscure rather than clarify our understanding of a complex reality. Thus women’s history experts fall into the same trap as everybody else when dealing with gender: thinking in opposites. This makes it difficult to avoid caricaturing, a tendency that feminist scholars abhor when they detect it in other people’s writings. In historical research, sharply delineated theories of opposites do not on the whole stand up very well. Distinctions inevitably turn out to be less sharp [End Page 10] than presumed. This becomes apparent as we resume the search for two possibly gendered channels in early Dutch welfare state legislation. We will begin with the issue of mothers’ pensions.
Mothers, Survivors, Social Insurance, and Poor Relief
Did the early Dutch welfare state have two channels? This question is not as straightforward as it might appear. Like the United States, it began with a WCA, but although it had provisions for women, there was, as mentioned, no counterpart to American mothers’ aid or pensions. Yet this did not make the Netherlands entirely dissimilar to the U.S. Let us look a bit more closely at the American “original.” As Ann Orloff (1991) has pointed out, the phrase “mothers’ aid” is in fact somewhat misleading because the pensions did not benefit mothers in general, but rather those who had lost the support of their (male) breadwinners—most often, white widows(International Labour Office 1936, 694). 10 Moreover, the pensions were awarded to women not in their capacity as caregivers, but to replace the support male breadwinners had provided for their children. Reformers hoped that this form of aid would encourage women to remain at home as caregivers rather than opting for paid work outside the home, but benefits were on the whole insufficient (Orloff 1991), and most women ended up supplementing them through some form of wage-earning. 11 Many, Eileen Boris has explained in detail (1994, 115, 116), turned to industrial homework, which they could combine with their child-caring responsibilities.
Mothers’ pensions were not part of the early Dutch welfare state, but widows’ pensions were (WCA clause 23 to 27). From 1903 on, a small category of mothers could claim the survivors’ benefits that had been incorporated into the Dutch WCA. The widow of an insured worker killed in an industrial accident was entitled to a maximum of 30 percent of his former wage. This was considerably lower than the 70 percent received by workers who had been completely disabled, but in addition, any legitimate child received 15 percent of his or her parent’s daily wage on an unconditional basis, and at least two thirds of the widows listed in the accident statistics had children. Benefits for widows and surviving children were not means-tested, but those for all other survivors were; their entitlements based on the extent to which the dead worker had contributed to their maintenance. However, the total pension paid to a widow and orphans could not exceed 60 percent (WCA clause 27).
Whether this was enough for a living is doubtful. Nevertheless, the act did function as a safety net for certain Dutch women. The WCA covered some 27 percent of the registered male workforce in [End Page 11] the Netherlands, which between 1899 and 1909 amounted to an average of 433,000 men. Slightly over half of those were married—some 247,000. 12 This meant that an equivalent number of women were entitled to widows’ insurance under the act. Although fatal accidents were few, between 1903 and 1915 the number of widows (both with and without children) receiving pensions under the act increased steadily, from 141 to 1,683. Widows would occasionally remarry, whereupon they would receive a lump-sum payment—a sort of marriage premium—equivalent to twice the value of their annual pension. Children’s entitlement ceased when they reached the age of sixteen, and of course when dependents died. But that did not offset the 140 or so widows who swelled the ranks of beneficiaries every year. A similar trend can be seen in the share of benefits as a whole paid to survivors. Throughout the entire period up to 1930, somewhat more than a tenth of the compensation paid by the Dutch Social Insurance Bank—one of the main implementing bodies of the WCA—went to survivors. 13
This small group of Dutch white widows occupied the same channel as workers/breadwinners (both male and female) who were covered by the WCA, that is, the insurance channel. 14 The state organized their insurance, and it obliged employers to contribute to the WCA. Though the widows’ pension was an indirect entitlement—a benefit to which women were entitled not so much as mothers but as wives of male workers in dangerous professions—it was a right nevertheless; widows were not dependent on either the state or charity for favors. As far as the state was concerned, their claims were as legitimate as the claims of workers, and individuals in both categories received predetermined benefits for a predetermined period. The implementation and appeal procedures remained the same whether a claimant under the Dutch WCA was a worker, a widow, an orphan, or (more rarely) a widower.
In this respect it is interesting to note that the Dutch government—unlike its great social-insurance example, Germany—explicitly kept open the possibility that there were men who might incur financial difficulties should their wives pass away. Along with widows, children, parents, grandparents, and parents-in-law, the WCA also considered widowers to be next-of-kin of the deceased. Clearly, then, the architects of this first form of Dutch social insurance for survivors were not operating purely on the male breadwinner principle. I will return to this later.
There was a second channel of social provision in the Netherlands during this period: poor relief. Nearly always organized locally, this form of social assistance was distributed by private organizations, the church, or the municipality. Up until World War II it was a very [End Page 12] important part of Dutch welfare provisions (Leeuwen 1998, 276). Around 1900 almost 5 percent of the Dutch population received continuous support from poor relief organizations (Gerwen and Leeuwen 2000, 433). Only when one was really out of money and when networks of family, friends, and neighbors failed to provide support, could one appeal to poor relief; it was, in other words, the last resort.
Like the Dutch insurance channel, this one was also not restricted by gender. Widows and mothers in need who were ineligible for support from the WCA could turn to the relief organizations, which provided practical assistance (midwives, restoratives, and well-baby clinics) as well as small amounts of money. Poor relief also provided for large numbers of old people, including men. Ill and unemployed men were also significant beneficiaries. (Unfortunately, exact and gendered numbers about the clientele of Dutch poor relief are not available; Valk 1986; Leeuwen 1998). Because nobody could actually survive on the assistance offered, recipients—women as well as men—were advised to find jobs. At the same time, financial assistance was accompanied by social control and lady visitors. There are numerous examples of men who experienced support by charitable organizations as stigmatizing and humiliating. Whether Dutch widows with children felt the same has not been studied but is plausible, as the example of the widows of the gasworks shows.
With regard to state welfare provisions in the United States and the Netherlands in the early twentieth century, then, we can make the following comparative points. Because the Netherlands had no mothers’ aid laws, it appears at first glance to be quite different from the United States. But the two countries were similar in two significant ways. The first has to do with American WCAs. On closer examination, this “first-channel” benefit was less exclusively male than Nelson’s model implies. Through insurance for next-of-kin, widows as well as (male) workers were covered under the scheme (Orloff 1991, 263). After a fatal industrial accident, widows received their survivors’ benefits directly (without means testing), just as in the Netherlands. For this reason, Orloff (1991) tells us, in an articulate and critical reaction to Nelson, that scholars must distinguish between intentions, implementation, and administration in early U.S. social policy. In both countries the widows’ pensions through the WCA were not so much honorific as they were meant as a compensation for the flaws of progressing industrialization.
The second similarity lies in the pension practices of the United States and the Netherlands. The administration of mothers’ pensions in the United States resembles the way Dutch poor relief treated its clients. Both were noncontributory; both offered small payments and expected the widows/mothers to earn additional (or must we say, [End Page 13] primary?) income. In addition, both imposed some form of control on clients, and both (despite honorific rhetoric about motherhood as a service to the state in the United States; Michel 1999, 79–81) had a stigmatizing effect. One could say that American mothers’ aid was intended as first aid, whereas Dutch poor relief was intended as last resort, but in terms of administration, both schemes were quite similar. With regard to the gender orientation of the “second channel,” however, the Dutch one was clearly not as exclusively feminized as the one Nelson describes in the United States.
Gendered Workers in the Dutch WCA
If one focuses on the beneficiaries of workers’ benefits on the other hand, the gendered character of this group in the Dutch WCA is hard to miss. In the Netherlands, as in America, men were the main beneficiaries; wage-earning women received only 2 percent of the total benefits paid out to victims of industrial accidents. 15 This percentage bore almost no relation to the proportion of women in the labor force, both in the sectors that were actually insured under the WCA and in the force as a whole.
It is not clear why so few women in covered industries received benefits. In the first years of its administration, the WCA insured approximately 54,000 women and 560,000 men, a rough approximation of the one-to-ten ratio of women to men in the Dutch industrial sector at the time. English historian Barbara Harrison has suggested that in the British case a comparable extremely low percentage can be explained by the absence of women workers near machines (Harrison 1993, 260), but it is difficult to either support or refute this point with the Dutch evidence available. It might also be the case that women were more prudent on the shop floor, or that controlling doctors were more eager to declare them healthy or fit for the job again. In any event, women received benefits only about one-fifth as frequently as could be expected.
Overall, the ratio of female to male wage-earners was one to three, with the bulk of women employed in sectors that were not covered by the act, such as domestic service. In 1899, 96 percent of domestic workers were female, constituting 44 percent of the registered female workforce (Eijl 1994, 369, 370). Their exclusion was briefly debated in Parliament, but the situation was left unchanged (Poelstra 1996, 231–237). Nevertheless, their jobs frequently involved risky tasks that when performed by male workers would have been covered. Compare, for example, the situations of domestic servants and of window cleaners. Both quite frequently fell from windows while cleaning them, but the latter were insured whereas the former were [End Page 14] not. Workers in the agricultural sector—the largest employer of women after domestic work—were also excluded from the benefits of the act (Staatsblad 1901, clause 11). Thus the majority of Dutch women workers were not in a position to benefit from this first form of public insurance.
Yet the impact of the WCA on Dutch women as a whole was not insignificant. In thinking about the social and economic significance of a form of social provision, it is important to consider not only who actually receives benefits but also who is eligible for them. Industries with heavy female employment (like textiles, laundries, papermaking, and canning) fell under the WCA because they used mechanical power. In addition, hundreds of thousands of women, as noted, were covered indirectly through the act’s survivors’ benefits. All told, when we add together the 54,000 insured women workers and the 247,000 women who were eligible for survivors’ benefits, we arrived at approximately 300,000 women—more than half of the 570,000 men covered under the WCA. 16
Although the Dutch and American WCAs had similar gender implications, one cannot argue that the Dutch social insurance channel was so narrow that it only served working men. In comparison to the three hundred thousand women just enumerated, Nelson speaks of only a “few women beneficiaries” in connection to the U.S. WCA (Nelson 1990). 17 Nevertheless, in both societies male breadwinners were far more likely to be insured than female ones. As to benefits given directly to workers on the basis of employment, the Dutch WCA of 1901 mandated insurance of 27 percent of the registered male labor force but of only 11 percent of the registered female working population. Four to five times as many women were insured indirectly through their husbands. Thus for women the act was proportionately more important as widows’ insurance than as workers’ disability insurance.
But does all this mean that the origins of Dutch social policy were male breadwinner–minded tout court? Let me finish with a brief exploration of this question in which the intentions of the ruling Dutch liberal government and the case of Bernardus Gerhardus L. play a rather surprising role.
Breadwinners and the WCA:
Principles and Practicality
It should be noted at the outset that around 1900 the Dutch did not conceive of breadwinners as male by definition. This becomes clear in the Bernardus Gerhardus case (Centraal Orgaan 1904, 118–121). Bernardus was a man about twenty years old. While he was busy digging out the old dike on the right bank of the River IJssel, [End Page 15] he was caught between a dump cart and the falling earth. He broke his spine and died the same day. Bernardus was a son in a family with eight children (two daughters and six sons) between seven and twenty-three years of age; before his death, he regularly passed his entire weekly wage of 8 guilders on to his father. To determine if and for which part Bernardus contributed to the family income prior to his death—to know, in other words, if Bernardus was a co-breadwinner whose wage had to be compensated by the WCA—the WCA Court of Appeals made a calculation of the family income. Note that although the court used the notion of breadwinner, it did not elucidate or clarify the term as such. Any interpretations were wrought in the course of legal practice:
The father earned around 4 guilders a week, while, when he was unemployed, tilling the land next to his house, so that his weekly income may be established at five guilders a week. The three grown-up sons each earned an equal wage . . . of 8 guilders a week. The adult daughter helped her mother in the house and worked the land, thus economizing on a maid, wherefore the daughter’s contribution may be established at a money value of 3 guilders; the other children attended school and did not contribute to the family income. (Centraal Orgaan 1904, 119)
The court remarked that it was disregarding the income of a third daughter who was a maidservant in Arnhem because she was completely independent and did not contribute to the family coffer. In total, the family income amounted up to f.32 a week, a very reasonable sum for that time. Striking in this calculation is the fact that the unpaid work of the father and daughter were officially counted; even the value of domestic work of a young woman was expressed in money. But even more important for the present inquiry is the fact that the court carefully traced the contributions of both women and men to the family income. In the first decade of the twentieth century, it was not sex or gender but sharing a household, family relations, and the value of work that qualified an individual as a (co-)breadwinner. 18
The liberal government of Pierson and his WCA minister Lely knew very well that many families during this period needed more than a father’s income to survive. Neither the Dutch as a whole nor certainly the legislators of the WCA placed blind faith in the adequacy of the male breadwinner. 19 Most men simply did not yet earn a family wage, and it was not until the 1920s that employers and workers came to an agreement on that point. In spite of the fact that women usually earned much less than men, their wages nonetheless contributed substantially to their families’ livelihood.
Moreover, this liberal government did not on principle wish to [End Page 16] distinguish between accidents that happened to men and those that happened to women. As socially oriented liberals, they were aware of the women’s emancipation movement, and they tried hard not to treat women—particularly married women—as second-class citizens. This meant that married women were eligible for the individual workers’ benefits provided by the WCA. If a married female employee of a cannery lost part of her fingers to the metal punch, regardless of the income of her husband or children, she was entitled to a fixed benefit. The same applied to the many married women who worked as cleaners in industrial plants, for example in printing. If the WCA covered the business, they were in. Thus the act and the jurisprudence it engendered reckoned with married women as separate breadwinners. According to its provisions, women were to be considered individuals, not dependents of their husbands.
This proved to be one of the first cracks in what Dutch historian Marianne Braun has called the persona-miserabilis doctrine in the Dutch context (Braun 1992)—a doctrine anchored in marital laws that since 1834 had deprived married women of their legal agency. Because of this they could not act independently within the legal system and had virtually no independent economic rights. Braun speaks of a “surreptitious undermining” (Braun 1992, 298, 299) of this doctrine under pressure of feminism throughout the twentieth century. The WCA can definitely be seen as a step in this direction.
In formulating the WCA of 1901, the Dutch government and Parliament did not take as their point of departure the notion of a small nuclear family in which the oldest man was the sole breadwinner. Clearly they had a larger familial economic network in mind. Thus it may be argued that the origins of Dutch social insurance were a bit less univocal than Dutch second-wave feminists have presumed. The WCA most certainly did not result from a well-established male-breadwinner ideology.
Conclusion
After the first decade of the twentieth century, liberal politics lost ground in the Netherlands. Under confessional governments, the male-breadwinner principle became stronger and more firmly entrenched. Over the course of the twentieth century, Dutch married women became increasingly dependent on their husbands’ incomes, as more men began to earn a family wage. Payment of dependents’ allowances and family allowances in the 1930s and 1940s, as well as postwar forms of social insurance, underscored the financial role of especially the father. The results led Dutch second-wave feminists to believe that the Dutch welfare state was rooted in a male-breadwinner ideology. [End Page 17] In that context we can begin to understand the attraction of Nelson’s model.
Nevertheless, as the historical record reveals, from the very start, Dutch men and women (workers, mothers, and widows) were covered by at least two different strands of social protection: the somewhat older system of local public assistance and the new state-created social insurance. Both strands offered support to both genders, albeit in different ways and with different merits. It is obvious that the Dutch welfare system had gender implications, but in its earliest stage of development, its pattern does not fit Nelson’s model of a strictly gendered division between the schemes and their beneficiaries.
At least a decade ago, Ann Orloff and Theda Skocpol called for a more nuanced version of Nelson’s interpretation of early U.S. social policy (Orloff 1991; Skocpol 1993). Now this modest comparison with the Netherlands indirectly calls for a similar move. In analyzing the gender implications of different channels in early welfare states, let us examine more precisely intentions, effects, administration, the supposedly indirect benefits to dependents, favors, and entitlements. Let us also take into account provisions that were not organized by the state. In my research, poor relief turned up, but it is likely that employers’ schemes and mutual societies also played a role in the development of many welfare systems. 20 The way in which these channels were connected to each other seems worth considering, with regard not only to discourse but also to the division of public (and private) expenses. It is worth noting that one of the reasons for the emergence of a Dutch WCA was that poor relief institutions did not want to and were not able to foot the bill for industrial accidents. In the Netherlands the more important nationwide social insurance became, the less important locally organized poor relief was. From a wider perspective these channels worked as communicating vessels.
Marian Van Der Klein studied history at the University of Utrecht in the Netherlands. She is writing a Ph.D. thesis on women, feminism, and social insurance in the Netherlands from 1890 to 1940 at the International Institute of Social History in Amsterdam. In 2002 she organized an international workshop on “Maternalism Reconsidered: Mothers and Method in Twentieth Century History” (volume forthcoming). Her previous work includes “Employers’ Schemes in Twente, the Netherlands: Industrial Accidents, Caring Power, and Local Breadwinning Practices around 1900,” in Business and Society. Entrepreneurs, Politics and Networks in a Historical Perspective (Rotterdam: Centre of Business History), 425–443.
Notes
This is a revision of the paper presented at the 12th Berkshire Conference on Women’s History in Storrs, Connecticut. I want to thank Agnes Andeweg, Lex Heerma van Voss, Sonya Michel and Elise van Nederveen Meerkerk for their comments. Translation by Jane Hedley-Prole.
1. I am writing a Ph.D. dissertation on women, feminism, and social insurance in the Netherlands from 1890 to 1940 at the International Institute of Social History (Amsterdam).
2. Of course, the Netherlands is a tiny country in comparison to the United States, but, as Robert Cox (1993, viii) has shown, it is worth studying comparatively.
3. The male breadwinner was so firmly embedded in the Dutch welfare system that in 1978 the Dutch state was reprimanded by the European Economic [End Page 18] Community (EEC) and granted seven years to review its social security system in terms of discriminatory clauses concerning sex/gender and marital status.
4. An extensive literature on the gendered implications of early Dutch social policy does not yet exist. In addition to my research, pioneering efforts have been made by Tjitske Akkerman (1998).
5. See also International Information Centre and Archives for the Women’s Movement (Amsterdam) Nationaal Bureau voor Vrouwenarbeid archives, 783.
6. Sociaal Weekblad 1900, 10; on Rutger-Hoitsema see also Röling (1987), Smit (1995, 1998), and Eijl (1994).
7. Because women could not represent themselves in legal matters during this period, de Klerk apparently served as their agent.
8. According to Hintzen they had already had words on this subject earlier (Hintzen 1899, 83).
9. For example by each other and by Bussemaker (1991).
10. In some states, eligibility was extended to women whose husbands were temporarily incapacitated or disabled, and even to those who had been deserted or divorced, but they were far less likely to receive benefits.
11. Child labor was also an important source of income for such families—and another target of reformers.
12. I combined figures from the accident statistics compiled pursuant to the WCA (Ongevallenstatistiek 1906) with population statistics.
13. The figures, in percentages of pensions as a whole, come from Dutch Social Insurance Bank reports Verslagen omtrent den staat der Rijksverzekeringsbank en hare werkzaamheden in het jaar (1903 to 1910, 1915, 1920, 1930, and 1940).
14. In the Netherlands, too, it was almost exclusively white women who benefited from insurance and relief. The Dutch WCA covered only working men who lived and worked within the narrow boundaries of the Netherlands in Europe. The scheme did not apply in the colonies (East and West Indies). During the first decades of the twentieth century, very few women of color lived in the (European) Netherlands.
15. The figures in this and the following two paragraphs are based on the Ongevallenstatistiek (1906). The figures given here are the average of those four years. The accident statistics were combined with data from the Thirteenth General Census (1966).
16. At most 12,000 additional men (in addition to the 560,000 workers) were covered as potential survivors of female breadwinning wives. At most 5 percent of the married women were part of the registered labor force in the Netherlands at the turn of the century: 5 percent of 247,000 covered married women in the WCA equals 12,350 potential surviving husbands.
17. Nelson does not mention widows or precise numbers of working women covered by the American WCAs.
18. The term breadwinner was indeed very broadly understood in the Dutch WCA jurisprudence: breadwinners could be men, women, or children who contributed in one way or another—be it financially or in natura—to the income of a joint household. [End Page 19]
19. During the same period, only a minority of American men were earning a family wage, but the ideology of the male breadwinner earning such a wage was stronger in the United States, in part because it supported the claims of trade unions for higher wages. In the Netherlands around 1900 the unions were not yet that strong.
20. Sonya Michel has argued we should talk about welfare systems that are at least three-tiered (Michel 2001): (1) benefits given directly to workers on the basis of employment; (2) benefits given to dependents of employees; and (3) benefits given as “gratuities” (what I call favors in this article), noncontributory. But maybe three tiers are not enough. Maybe we should first ask how many providers of welfare there were at a certain time in a certain place, and then ask again what it meant for clients, male and female, to be dependent on the one or the other.
References
Akkerman, Tjitske.1998. “Political Participation and Social Rights. The Triumph of the Breadwinner in the Netherlands.” Pp. 38–50 in Gender, Participation and Citizenship in the Netherlands, ed. Jet Bussemaker and Rianne Voet. Aldershot: Ashgate.
“Binnenlandsch overzicht.” 1898. Sociaal Weekblad April 30: 146.
Bock, Gisela. 1994 “Poverty and Mothers’ Rights in the Emerging Welfare States.” In A History of Women in the West, vol. 5: 20th Century, ed. Françoise Thebaud. Cambridge, Mass.: Belknap Press of Harvard University Press.
Boris, Eileen. 1994. Home to Work. Motherhood and the Politics of Industrial Homework in the United States. New York: Cambridge University Press.
Braun, Marianne. 1992. De prijs van de liefde. De eerste feministische golf, het huwelijksrecht en de vaderlandse geschiedenis. Amsterdam: Spinhuis.
Bussemaker, Jet. 1991. “Verplicht eigenbelang of verplicht altruïsme? Literatuur over sekse en verzorgingsstaat.” Tijdschrift voor Vrouwenstudies: 515–530.
Centraal Orgaan voor de Ongevallen-Verzekering en andere Werklieden-Verzekeringen onder redactie van Mr. Dr. J. E. Millard. 1904. 1.
Cox, Robert H. 1993. The Development of the Dutch Welfare State: From Workers’ Insurance to Universal Entitlement. Pittsburgh: University of Pittsburgh Press.
“De ramp in de gasfabriek te Rotterdam.” 1898. Sociaal Weekblad: 168–170.
Eijl, Corrie van. 1994. Het werkzame verschil. Vrouwen in de slag om arbeid 1898–1940. Hilversum: Verloren.
Evolutie. 1900. “De Ongevallenwet.” 7 June: 36.
Fraser, Nancy. 1990. “Struggle over Needs: Outline of a Socialist-Feminist Critical Theory of Late-Capitalist Political Culture.” Pp. 205–231 in Women, the State and Welfare, ed. Linda Gordon. Madison: University of Wisconsin Press. [End Page 20]
Genabeek, Joost van. 1999. Met vereende kracht risico’s verzacht. De plaats van onderlinge hulp binnen de negentiende-eeuwse particuliere regelingen van sociale zekerheid. Amsterdam: Stichting Beheer IISG.
Gerwen, Jacques van, and Marco van Leeuwen. 2000. Zoeken naar zekerheid: risico’s, preventie, verzekeringen en andere zekerheidsregeleingen in nederland 1500–2000. Deel III. De ontluikende verzorgingsstaat 1890–1940. Amsterdam: NEHA/Verbond van Verzekeraars.
Gordon, Linda. 1990. “The New Feminist Scholarship on the Welfare State.” Pp. 9–35 in Women, the State and Welfare, ed. Linda Gordon. Madison: University of Wisconsin Press.
_____. 1992. “Social Insurance and Public Assistance: The Influence of Gender in Welfare Thought in the United States, 1890–1935.” American Historical Review 97(1): 19–54.
_____. 1993. “Gender State and Society: A Debate with Theda Skocpol.” Contention 2(3): 131–157.
Handelingen der Staten Generaal. 1899–1900. II, 231–239.
Harrison, Barbara. 1993. “Are Accidents Gender Neutral? The Case of Women’s Industrial Work in Britain, 1880–1914.” Women’s History Review 2(2): 253–275.
Hintzen, G. H. 1899. “Mijnheer de Redacteur.” Sociaal Weekblad 13(7): 83–84.
International Labour Office. 1936. Studies and Reports Series M. (Social Insurance) no. 13, International Survey of Social Services vol. I. Geneva.
Klein, Marian van der. 2000. “Employers Schemes in Twente, the Netherlands: Industrial Accidents, Caring Power and Local Breadwinnning Practices around 1900.” Pp. 425–443 in Business and Society. Entrepreneurs, Politics and Networks in a Historical Perspective. Rotterdam: Centre of Business History.
Knijn, Trudie. 1994. “Fish without Bikes: Revision of the Dutch Welfare State and its Consequences for the (In)Dependence of Single Mothers.” Social Politics 1: 83–106.
Koven, Seth, and Sonya Michel, eds. 1993. Mothers of a New World: Maternalist Politics and the Origins of Welfare States. London: Routledge.
Leeuwen, Marco van. 1998. “Armenzorg 1800–1912: erfenis van de republiek.” Pp. 276–317 in Studies over zekerheidsarrangementen. Risico’s, risicobestrijding en verzekeringen in Nederland vanaf de Middeleeuwen, eds. Jacques van Gerwen and Marco van Leeuwen. Amsterdam: NEHA/Verbond van Verzekeraars.
Michel, Sonya. 1999. Children’s Interests/Mothers’ Rights: the Shaping of America’s Child Care Policy. New Haven, Conn.: Yale University Press.
_____. 2001. “The Benefits of Race and Gender: Democracy and Equity in America’s Public-Private Welfare State.” Unpublished paper.
Nelson, Barbara J. 1990. “The Origins of the Two-Channel Welfare State: Workmen’s Compensation and Mothers’ Aid.” Pp. 123–157 in Women, the State and Welfare, ed. Linda Gordon. Madison: University of Wisconsin Press.
Ongevallenstatistiek betreffende het tijdvak 1 (1903–1907). 1906. Samengesteld [End Page 21] ter voldoening aan de bepaling van art. 17 der Ongevallenwet 1901 [Industrial Accident Statistics 1903]. Amsterdam: M. van Lindenbaum.
Orloff, Ann Shola. 1991. “Gender in Early U.S. Social Policy.” Journal of Policy History 3(3): 249–282.
Poelstra, Jannie. 1996. Luiden van een andere beweging. Huishoudelijke arbeid in Nederland 1840–1920. Amsterdam: Spinhuis.
Riley, Denise. 1988. Am I that Name? Feminism and the Category of “Women” in History. London: Macmillan.
Röling, H. Q. 1987. “Hoitsema, Maria Wilhelmina Hendrika.” Pp. 61–64 in Biografisch woordenboek van het socialisme en de arbeidersbeweging in Nederland part II, ed. P. J. Meertens, Mies Campfens, Albert F. Mellink, Bob Reinalda, and Johanna Welcker. Amsterdam: Stichting Beheer IISG.
Rutgers-Hoitsema, M. W. H. 1899a. “Recht of armbedeeling.” Sociaal Weekblad 13(5): 52–54.
_____. 1899b. “Recht of armbedeeling.” Sociaal Weekblad 13(82): 94.
Schwitters, Rob J. S. 1991. De risico’s van de arbeid. Het ontstaan van de Ongevallenwet 1901 in sociologisch perspectief. Groningen: Wolters Noordhoff.
Sevenhuijsen, Selma. 1978. “Vadertje staat, moedertje thuis? Vrouwen, reproduktie en de staat.” Socialisties-Feministiese Teksten 1: 18–67.
Skocpol, Theda. 1992. Protecting Mothers and Soldiers: the Political Origins of Social Policy in the United States. Cambridge, Mass.: Belknap Press of Harvard University Press.
_____. 1993. “Soldiers, Workers and Mothers: Gendered Identities in Early U.S. Social Policy.” Contention 2(3): 157–183.
Smit, Christianne. 1995. Maria W. H. Rutgers-Hoitsema (1847–1934). Een vrijgevochten vrouw in de eerste feministische golf. Ph.D. diss., Department of Political Science, University of Amsterdam.
_____. 1998. “Een vrijgevochten vrouw. Maria W. H. Rutgers-Hoitsema.” Onvoltooid Verleden October: 51–59.
Staatsblad van het Koninkrijk der Nederlanden. 1901, January 2. “Wet houdende wettelijke verzekering van werklieden tegen geldelijke gevolgen van ongevallen in bepaalde bedrijven.” [Dutch WCA, literally: Act Regarding Compulsory Insurance of Workmen Against Financial Consequences of Accidents in Certain Businesses].
Thirteenth General Census. 1996. 13de Algemene Volkstelling 31 mei 1960. Deel 10. Beroepsbevolking. C. Vergelijking en uitkomsten van de beroeptellingen 1849–1960. W. de Haan N.V. Hilversum, pp. 28–40 (Table 1).
Valk, Loes van der. 1986. Van pauperzorg tot bestaanszekerheid. Een onderzoek naar de ontwikkeling van de armenzorg in Nederland tegen de achtergrond van de overgang naar de Algemene Bijstandswet, 1912–1965. Amsterdam: Stichting Internationaal Instituut voor Sociale Geschiedenis.
