Citizenship and mobility of the poor: Sweden during the 19th century

Abstract

The distinctive features of the modern form of citizenship include, among other things, that it is both internally inclusive and externally exclusive; that it establishes legal equality so that membership of the state supersedes all other memberships and allegiances; and that it defines membership as independent of residence. These characteristics largely evolved during the “long nineteenth century”, the period between the French Revolution and the First World War. Similar to that of today, the historical situation in which citizenship evolved was one of intensified mobility. With the example of Sweden, this article finds that citizenship, in both its internal and external dimensions, was elaborated partly as a way to manage the mobility of the migrant poor. The contours of citizenship emerged as authorities aimed to control and direct the movement of the poor, which preceded control efforts. This has implications for our understanding of citizenship as well as of the state, and highlights the agency of migrants.

Modern statehood is often defined by reference to its bounded territory, over which its monopolization of violence and its various administrative powers are exercised. But the modern state is also a membership organization, defined by the institution and the principle of citizenship. What is distinctive about the modern form of citizenship is, among other things, that it is both internally inclusive and externally exclusive. This means that it, on the one hand, establishes legal equality and makes sure that membership of the state supersedes all other memberships and allegiances. On the other, it defines membership as independent of residence, making the distinction between citizen and foreigner paramount [Bosniak 2007; Brubaker 1992; Fahrmeir 2007; Gosewinkel 2021; Hindess 2000; Walzer 1983: chap. 2].

These processes of citizenship largely evolved during the “long nineteenth century”, that is, the period between the French Revolution and the First World War [Hobsbawm 1962, 1975, 1987; Buzan and Lawson 2013]. This was an era marked by intensified human mobility and migration [Osterhammel 2014: chap. 4]. A series of macro-level developments had led to a sharp increase in the movement of the poor and destitute: for instance, the population explosion, the proletarianization of the countryside, the beginning of industrialization and the development of railways and other means of transport. Poor people took to the road or were forcibly uprooted. In Sweden (as in so many other European countries), large numbers of people emigrated to North America, while others moved to neighboring countries such as Denmark, Norway and Germany. Yet others migrated domestically, to urban centers and elsewhere, in search of work and subsistence [Bohlin and Eurenius 2010; Carlsson 1961: 448–463; Sundbärg 1910, 1911]. For authorities, the mobility of the poor brought a renewed awareness of and concern with what Michel Foucault has called the “problem of circulation”, that is, the need to distinguish between wanted and unwanted mobility, and to find ways to stimulate the former and repress the latter [Foucault 2007, 2008].

In this article, I want to discuss how these two issues (citizenship and the mobility of the poor) were historically connected. The argument is that citizenship was worked out partly as a response to the mobility of the poor. Citizenship was in this sense employed as a governmental technology for managing populations [Rose and Miller 1992; Cruikshank 1999; Hindess 2000; [author ref]). This distinguishes my approach from the dominating view of citizenship as the outcome of struggles for political rights and inclusion [e.g. Östberg 2021]. It seems to me that a fuller account can be attained if such “heroic” stories are complemented by the likes of mine, which emphasizes the practical, mundane and sometimes repressive usages of citizenship.

Empirically, there is a relatively broad historical research on the control of the mobility of the poor in Sweden. This research makes up a large part of my data. Some of it takes a far-reaching perspective, as mine does here, while other scholarship delimits its scope to a short time span or a single locality, which is then explored in detail. Research engages with issues such as passport laws [Lövgren 2018], vagrancy [Edman 2008; Johnsson 2016; Rosander 1978], master and servant relations, labor and unemployment regulations [Harnesk 1990; Junestav 2008; Olofsson 1996; Uppenberg 2018; Wallentin 1982], immigration and emigration [Bohlin and Eurenius, 2010; Hammar 1964; Karadja and Prawitz 2019; Retsö 2016; Svanberg and Tydén 2005; Willerslev 1983], poor relief and social policy [Åström 2008; Berggren and Nilsson 1965; Jordansson 2008; Montgomery 1951, Qvarsell 2008; Swärd 2008; Vikström 2006], prisons and forced labor [Nilsson 1999; Petersson 1983] and ethnically motivated exclusionary measures toward certain groups [Ericsson 2012, 2016; Dahlstedt, Härnbro and Vesterberg 2019]. But I do not know of any previous study of Sweden that has explicitly engaged with the interrelation between the mobility of the poor and the development of citizenship.

This study is delimited in the sense that it is mostly concerned with domestic movement regulations. In the second half of the 19th century, there was also substantial emigration from among similar social strata, which I focus on here. Between 1821 and 1930, 1.2 million people left Sweden, out of a population that amounted to 5.2 million in 1900. The majority went to the United States, but there was also a considerable amount of emigration to neighboring countries like Denmark, Norway and Germany [Stråth 2012: 293–294]. Here, I will only refer to emigration briefly, and will instead concentrate on internal movement within Sweden.

The outline is as follows. First, I will discuss the relative invisibility of citizenship in state theory. I will then turn to explicating my theoretical approach to the topic, specifically emphasizing the writings of Michel Foucault and Rogers Brubaker. The third section provides a contextualization of Sweden. After that I will discuss the relevant processes in Sweden, referred to here as “internal inclusiveness” and “external exclusiveness”. The paper ends with a summary.

Citizenship in state theory

Citizenship does not figure prominently in most works on state theory.¹ By this, I mean that the shaping of the modern legal institution of citizenship as linked to statehood has not been given much attention. In the early 1990s, Rogers Brubaker noted this weakness in the literature, which I would say, with some exceptions,² holds true today:

Conceiving the modern state as a territorial organization and the state system as a system of territorial states, political sociology has for the most part neglected citizenship and membership. It has made too little of the fact that the state is a membership association as well as a territorial organization; that the state constitutes itself, and delimits the field of its personal jurisdiction, by constituting its citizenry; and that political territory, as we know it today—bounded territory, within a system of territorial states, to which access is controlled by the state—presupposes membership, presupposes some way of assigning persons to states, and distinguishing those who enjoy free access to a particular state territory from those who do not. The emergence of the institution of citizenship therefore marks a crucial moment in the development of the infrastructure of the modern state and state system. [Brubaker 1992: 72]

In classic realist international relations theory, citizens are only conceived of as an element of state power. To Hans J. Morgenthau, it is the relative size of the population that matters, along with its national character and morale—for instance the willingness to sacrifice one’s life for the nation [Morgenthau 1948: 91–104]. In historical sociology, too, citizenship is understood instrumentally, usually in relation to the evolving state’s extraction capacity. Citizens then figure in their capacities as taxpayers, laborers and soldiers, and in the case of women, as reproducers of the nation [Mann 1984, 1993; Tilly 1990; Yuval-Davis 1997]. This literature emphasizes that the capacity to extract the needed resources from the population has to do with centralization but also with what Michael Mann calls “infrastructural power”: “the capacity of the state to actually penetrate civil society, and to implement logistically political decisions throughout the realm” [1984: 189]. The growing importance of statistics, censuses and other aspects of “information capacity” during the 19th century should be seen against this background [Brambor et al. 2020]. It was through them, and through the dissemination of standardized forms of language and measurements, that citizens became increasingly “legible”, and thus governable, for the state in this period [cf. Gosewinkel 2021: 18–19; Scott 1998].

The notion of territory, in contrast to citizenship, is at the center of most definitions of the state. In Max Weber’s famous definition, the state is “the form of human community that (successfully) lays claim to the monopoly of legitimate physical violence within a particular territory—and this idea of ‘territory’ is an essential defining feature” [Weber 2004: 33]. There has been a tendency in social theory, traditionally, to assume and not problematize what territory is, as has been argued forcefully by Agnew [1994] and Brenner et al. [2003], for instance. But later scholarship has gone further by unpacking the concept and exploring its various economic and social meanings [Harvey 2006; Poulantzas 2003] as well as its historical, philosophical and political dimensions [Brenner et al. 2003; Elden 2013; Taylor 2003]. For example, Henri Lefebvre has argued that state territory is not “just there” but is continuously produced, as a combination of physical, social and mental space [Lefebvre 2009].

While the notion of territory has thus been explored and historicized in the literature, the same cannot be said about citizenship. Scholars have instead tended to ignore it, or to treat it as secondary to territory. Benjamin de Carvalho comments that “the historicity of the subjects of the state is still largely unexplored, or, even worse, now subsumed into the historical process of territorialization: overshadowed, so to speak, by the emergent hegemony of territoriality” [2016: 58]. The assumption has generally been that the boundaries of citizenship have neatly coincided with those of the territories concerned, from which has followed a rather unproblematized notion of citizenship. One example, from the international relations literature, is the argument that territory and population together make up the state’s “physical base” [Buzan 1991].

I believe that citizenship needs to be understood separately from territory, and that the histories of the two need to be treated as partly distinct. I will also argue that, although the question of citizenship goes unnoticed in Buzan and Lawson’s argument for the significance of the 19th century for understanding the state and the state system [2013], modern citizenship went through important developments in the period between the French Revolution and the outbreak of the First World War. These developments were, I argue, partly precipitated by increased migration, especially of the poor. In this sense, I concur with those who have argued that the mobility of the poor constitutes a particular kind of non-centered agency [cf. Papadoupoulos and Tsianos 2013].

Modern citizenship in the government of populations

Existing historical accounts of citizenship are quite varied in scope and focus. For instance, T. H. Marshall’s account centers on the rights that individuals hold in relation to the state; Marshall’s influential, although much-debated, argument is that modern history has progressively expanded to cover first merely civil, then political and finally also social rights [Marshall 1950]. Comparative national model accounts instead focus on contrasting citizenship regimes whose differences are explained by reference to national traditions, historical experiences and other material and immaterial circumstances [Brubaker 1992; Favell 1998; Jensen 2019; Gosewinkel 2021]. Finally, intellectual history accounts typically focus on the equality and self-government aspects of citizenship, the ideational roots of which are traced over long time spans, from Greek and Roman citizenship models through the American and French revolutions to modern versions of citizenship [Bellamy 2015; Pocock 1995]. The reason why these accounts differ is that their authors focus on different elements of citizenship. Modern citizenship is at once a legal status of rights, a principle of democratic self-rule and an idea about collective identity and membership [Cohen 1999; Joppke 2010].

Complementing rather than challenging these accounts, I will argue that citizenship, as a particular form of membership, has also been used as a “technology” [Cruikshank 1999] for more mundanely managing the mobility of the poor. This means that I will locate modern citizenship in the larger history of managing populations, which has been uncovered by Michel Foucault, among others (see first subsection below). To explicate the specificities of citizenship in the relevant historical period, I turn to Rogers Brubaker (see second subsection below).

Managing the circulation of people

We saw above that much state theory has emphasized that state authorities have regarded the population as a resource to be exploited. But several authors have noted that the relationship between the state and its population has become more complex over time. People cease to be merely resources to exploit and subjects that shall be made to obey, and gradually acquire more complex roles in the eyes of the state [Poggi 2003; Skinner 2008; Scott 1998]. Michel Foucault famously argued that the evolution of the modern state since the 16th century has been accompanied by the ever-increasing importance of the population to state-governing ambitions. The population is, on the one hand, targeted as a collection of individuals, who can be separated, trained and disciplined into existing norms. On the other, it is targeted as a biological, living entity that follows its own laws and regularities concerning nativity, mortality, fertility and so on. The latter, especially, demonstrates that the state often needs to use softer means of intervention, and not only rely on coercion [Foucault 2007].

A subtheme within the broader one of population management is the government if circulation. This governmental imperative of arranging people and things in space has its own history. Foucault quotes Guillaume de la Perrière, who wrote in 1567 that “Government is the right disposition of things that one arranges so as to lead them to a suitable end” [Ibid.: 98]. Since then, the issue of “organizing circulation, eliminating its dangerous elements, making a division between good and bad circulation, and maximizing the good circulation by diminishing the bad” has become a core governmental concern [Ibid.: 18]. The precise character of that concern has varied over time with transformations in knowledge. In the mercantile era, when people were seen as the “wealth of the nation”, efforts were taken to increase the population and to enhance its productivity and performance. For those who left the country without proper authorization, punishments were harsh. In this era, another governmental objective was to populate and provide labor for the colonies. “Empire”, said Napoleon’s foreign minister de Talleyrand in 1797, “is the art of putting men in their place” [Pagden 2001: 10]. Vulnerable segments of the population were often forcibly relocated, not only to provide labor for colonies, but also to rid colonial centers of unwanted elements [Linebaugh and Rediker 2000]. The unwanted layers of the population often comprised ethnic or religious minorities, but also the sick and the destitute.

The 19th century was the era of “the great ‘confinement’ of paupers and vagabonds”, in which individuals were enclosed in specific institutions to be trained and disciplined [Foucault 1977: 141]. But it was also an era in which other, softer, ways of managing mobility were developed and linked to many different governmental objectives. The ambition was not (just) to forcibly move people and things around but also to, more gently, steer and guide processes. Reasons for this might include a desire to enhance trade, improve public hygiene by opening up disease-ridden areas or allow people to move to search for work, but also to fight off the influx of beggars, vagrants, criminals and so on. The overriding goal was that “of maximizing the positive elements, for which one provides the best possible circulation, and of minimizing what is risky and inconvenient, like theft and disease, while knowing that they will never be completely suppressed” [Foucault 2007: 18–19]. Liberal government, from this perspective, is not about guaranteeing freedoms as abstract ideals, but about organizing and managing freedoms, including by defining their limits. Freedoms (to travel, to own property, to take up employment, to buy and sell etc.) are necessary to various governmental objectives, but they must be overseen so that detrimental consequences of those freedoms are avoided. This means that concerns with security, in a broad sense, are a necessary correlate to freedom. In producing freedoms for subjects to enjoy, one must at the same time protect both individuals and societies from the risks that arise as part of the same process. One needs to make sure, for instance, that the freedom of economic processes does not put workers at excessive risk, while at the same time the freedom of workers does not endanger production. The liberal “art of government” is therefore continuously engaged in arbitrating the limits between freedom and security [Foucault 2008: 51–73].

As will be shown below, when we study Sweden in the 19th century, it is possible to discern these twin processes at work, both establishing freedoms and securing them. In this period, many new freedoms were introduced that required new mobilities—for instance, freedom to travel, to work, to trade, to set up businesses. At the same time, another set of measures were introduced in order to limit and guide these movements, both harsh and direct ones (e.g. punishments for vagrancy) and much more indirect ones (e.g. campaigns to convince people not to emigrate). But to grasp the specifics of citizenship in this management of mobility, Foucault’s writings do not suffice, and I have therefore turned to Brubaker.

Citizenship and the mobility of the poor

Rogers Brubaker’s seminal Citizenship and Nationhood in France and Germany [1992] has been of great importance for citizenship and nationalism studies. Most enduring, while also much criticized, has been his argument that the two countries significantly differ in their relationship to newcomers: France’s citizenship regime is “civic” in orientation, while Germany’s is “ethnic”. But here I will instead focus on a less-remembered part of his argument, which—moreover—I have not found as clearly expressed in any other, more recent scholarship [ibid: chap. 3]. The greatest advantage for me is that it allows me to connect the control over internal movements of the poor to state-making, by means of its effect on the evolvement of citizenship.

According to Brubaker, the state should not only be defined in terms of its delimited territory and its monopoly of violence. Just as fundamentally, the state is also a membership organization. It is organized around a particular modern conception of citizenship as simultaneously internally inclusive and externally exclusive. It is thus “hard-on-the-outside and soft-on-the-inside” [Bosniak 2007: 2451].

The shift toward internal inclusiveness was accomplished through the evolvement of legal equality, which put people in direct relation to the state. Legal equality necessitates that the state is the only source of legitimate law, which, furthermore, covers the whole territory. This was established as part of the process toward internal state sovereignty, and signified a radical break with the medieval period, when legal orders were multiple and overlapping. Moreover, legal equality means that the relation between the individual and the state is direct, and not mediated by intermediate membership organizations such as guilds or estates. There were many steps along the way to the establishment of legal equality. Of great importance were the establishment of economic freedoms: freedom of trade, of enterprise and of labor—including the mobility of labor. “To be sure, citizenship presupposes legal equality and legal equality was realized in the economic domain … The result was a unitary, homogenous space, within which all persons were formally free and equal economic actors” [Brubaker 1992: 61–62]. The ideal of legal equality was certainly never perfectly realized in practice as, for instance, gender, race and property were of continued importance [Fahrmeir 2007]. But the dismantling of domestic passport controls was one of several milestones attained in striving for internal inclusiveness.

The increasingly militarized and interventionist state of the nineteenth century necessitated a precise and bureaucratic delimitation of membership, which citizenship provided [Gosewinkel 2021: 19]. Modern citizenship not only required the establishment of inclusiveness and equality among members, but also the possibility of excluding others [Brubaker 1992: 64–72]. The second main shift was therefore toward external exclusiveness. A principal factor, Brubaker points out in his discussion of Germany, was the movement of the poor, and how this was to be handled under conditions of liberal economic integration. The poor had been the responsibility of local authorities for as long as anyone could remember. As the poor were becoming increasingly mobile, the local authorities faced a new situation which called for new regulations, that would be more restrictive in character.

The distinction between one’s “own” poor and poor others had first been made in relation to beggars in the late 15th and 16th century; local beggars were allowed but those belonging elsewhere were prohibited and barred. From the 16th century onwards, the most important “ours/yours distinction” concerned poor relief claimants. Municipalities were responsible for supporting their own poor, which incentivized them to define membership more restrictively, to send back poor relief claimants from elsewhere, and to bar new potential poor relief claimants from entering. This in turn led to frictions between local authorities as they all tried to externalize responsibility, with domestic deportations and inter-municipal negotiations as results. When the state was striving for internal sovereignty, such friction could not be allowed, since it put wider peace and stability at risk [Brubaker 1992: 64–65.

In the early 19th century, ideas of economic liberalism were becoming influential, and the state moved towards freedom of movement and of occupation. For these and other reasons, peasants became more mobile, and in this situation it became very difficult to sustain the autonomy of municipal poor relief. At that point the central state stepped in—not by taking over provisioning for the poor but by acting as a regulator. This involved legally defining the relationship that localities had with their poor as well as, crucially, when and to whom the right of legal domicile should be extended [Brubaker 1992: 64–65; see also Brown and Oates 1987]. Over time, this led to a perception that mobility across national borders needed to be controlled, too. As Leo Lucassen has written on the case of Germany: “it seems not too far-fetched to assume a causal relationship between the emergence of the (albeit embryonic) welfare state and legislation to control the entrance of immigrant workers on to the national labour market” [Lucassen, 2003: 188].

This process, occasioned by the movement of the poor, led to the eventual codification of membership of the state in the form of citizenship legislation. “Citizenship had crystallized as a formally defined and assigned status, distinct from residence. The citizenry was externally exclusive as well as internally inclusive … [Citizenship] could serve as an instrument and object of closure” [Brubaker 1992: 71]. In Sweden, the first real citizenship legislation is from 1894, around the same time as in many other European countries [Bernitz 2012; Gosewinkel 2021: 19–20].

Contextualization: Sweden in the 19th century

During the 20th century, Sweden became politically dominated by social democracy, and in the post Second War period attained very high levels of social and economic equality. In many people’s minds, Sweden became almost a symbol for egalitarianism—an ideal for some and a warning for others. But at the end of the 19th century, Sweden was no more equal than other European countries [Bengtsson 2020]. Writes Thomas Piketty: “[U]ntil the early twentieth century Sweden was a profoundly inegalitarian country, in some respects more inegalitarian than countries elsewhere in Europe; or, rather, it was more sophisticated in organizing its inequality and more systematic in expressing its proprietarian ideology and shaping its institutional incarnation” [Piketty 2020: 185].

In 1900, the wealthiest 10% owned 86% of all private assets and earned 47% of total incomes. This is on a par with the level of inequality in notably unequal countries today (e.g. Brazil, China, Russia). Compared to other countries in the same period, Sweden was at about the same level as UK and France, and slightly more unequal than the USA. The drivers of this were multiple. The rich had become much richer because of booming industrial production, increased exports of agrarian products and more efficient agriculture. At the mid-level of the social hierarchy, the highly educated were being rewarded with much higher wages than previously and large farm owners had increased their earnings considerably. Meanwhile, and as we will return to below, the situation for the poorest layers of the population had become more acute as a result of the processes of industrialization, urbanization and the proletarianization of the countryside [Bengtsson 2020: 55–61]. In terms of social equality, Sweden appears to have been among the most patriarchal, hierarchical and status-oriented countries in northern Europe in the 19th century [Ibid.: 71–72]. In the political sphere, Sweden had an intricate system for upholding inequalities, also after the introduction of a bicameral system in 1866. It took a long time for national voting rights to be made equal. Women were excluded from voting, but so were about 80% of all men in the second half of the 19th century because their incomes were too low. At the local level, voting rights were graded according to property and income, and firms could vote, too. In quite a few municipalities, one person or company possessed more than 50% of the votes, in one case even 98%. Thus, in democratic terms, Sweden was significantly more unequal than other West European countries [Ibid.: 73–82].

There was intense concern among policy makers about “the social question”, as in many other European countries [Case 2016; Stråth 2016: chap. 2]. Macro-sociological processes such as industrialization, urbanization and demographic change had deeply affected societies, and as we have seen, had created not only new elites but also new forms of poverty, vulnerability and associated problems. The social question was understood as covering many different social ills, among them prostitution, drunkenness, degeneracy and vagrancy [Edman 2008: 131; Johnsson 2016: 61; Petersson 1983].

One factor intensifying social problems was the enormous population increase during the 19th century. There were around 187 million Europeans in 1800 compared to 400 million a century later, and in Sweden the population rose from 2.3 million to more than 5 million in the same period [Sundbärg 1910: 11, 78–79]. One effect of this change was a steep decrease in the number of landowning farmers. The pattern of young people taking over their family farm when their parents grew old was broken in the early 19th century: there were not enough farms to sustain the growing population. What followed instead was a diversification of non-propertied farm workers into different categories,³ as well as social declassification and poverty [Uppenberg 2018: 20–23].

As a consequence, poor people became increasingly mobile. When people could not support themselves in their farms and villages, they had to move in order to find employment and provide for themselves. As mentioned above, a sizable share of them chose to emigrate, especially between 1880 and 1930, but there was also considerable internal movement. A new and mobile social underclass emerged. The elites regarded them with fear, as a threat to themselves as well as to the social order at large. But from the point of view of the elites, the mobile poor also presented an opportunity, in that they could provide cheap labor. A pool of “free”, waged labor was needed when industrialization began to take off in rural locations as well as urban centers. Labor needed to be mobile and adaptable to new and shifting demands on the labor market, and the rural poor therefore also constituted a labor market resource [Stråth 2012: 251–253].

Karl Marx understood these mobile people in terms of “surplus populations”, which he connected to the exploitation of workers that followed capitalist development. The term has recently been taken up by some scholars to describe the situation of today’s “marginalized”, “wasted” or “superfluous” migrants and refugees, often with a somewhat broader meaning [Rajaram 2018; Tyner 2013; Yates 2011]. Marx discerned a few different categories, which I think are observable in my case as well [Marx (1867) 1990: 794–797]. The floating surplus population are those who have been incorporated into the labor force, full-time workers who are temporarily out of work. Marx notes that “Some of these workers emigrate; in fact, they are merely following capital, which has itself emigrated” [Ibid: 794]. The latent surplus population are those who have not yet entered into the labor force. When capitalist production took hold over agriculture, the demand for rural workers fell. Hence, many of them turned toward the urban centers, while those that stayed put received lower wages, and therefore always stood “with one foot already in the swamp of pauperism” [Ibid: 796]. The stagnant category includes those who are only employed at very irregular intervals. It consists of workers who have become redundant, especially in branches where handicraft has given way to manufacture, and manufacture to machinery. This category is particularly vulnerable to exploitative working conditions with a maximum working time and a minimum of wages, and its members are very difficult for labor leaders to organize. The final category is the lumpenproletariat, which Marx considers “the lowest sediment” of the surplus population. It includes vagabonds, paupers and prostitutes, those who are able to work but refuse to, orphans, pauper children, the “demoralized” and those who are not able to work—either because they have not managed to adapt to new circumstances or because of disabilities resulting from workplace accidents [Ibid: 797; see also Barrow 2020: 35–43].⁴ I think that many of the poor migrants that I am concerned with could be placed in one of these categories, but I want to add one more—the nomadic population. These were people of rural origin, whose labor was used in building, draining, brickmaking or railway-building. Their form of employment drew them from place to place: “when they are not on the march, they ‘camp’” [Marx (1867)1990: 818]. Many of “my” migrants were at least occasionally employed in this type of labor. They were forced by necessity to move, and their movement was subject to quite harsh controls.

The poor were the object of many governmental and legal strategies. Sweden typified the Lutheran attitude toward poverty in that a sharp line was drawn between the “deserving” and the “undeserving” (lazy and/or immoral) poor [Kahl 2005]. A background condition for our discussion here is that employment was compulsory until 1885 (tjänstetvånget). All able-bodied men and women who did not own property in the form of land or economic resources had to be employed by a master from the age of 15 [Johnsson 2016: 52–83; Junestav 2008]. The individual duty to support oneself had a long history, having been laid out in the medieval laws. These established that any person who did not have sufficient means was obliged to work for the Crown, and this principle was continued in later, national legislation in the 16th century [Kjellson 1920: 167–169]. Punishments for the unemployed were severe and over time have included flogging and having one’s ears cut off. After the 16th century forced labor in service of the Crown was more common, and until 1824, so was forced conscription to the army [SOU 1923: 13–16; Nilsson 1999: 124–125; Rosander 1978: 11–12; Uppenberg 2018: 132; Johnsson 2016: 90].

A crucial legal distinction separated those who enjoyed access to legal defense (laga försvar) from those who did not and who were hence “defenseless” (försvarslösa). The term for legal defense first appeared in legislation on the nobility’s privileges in 1569 and 1617, where it originally referred to the freedom from military conscription that members of this social group enjoyed. It later took on a more general meaning, and essentially meant protection from being treated as a vagrant, which was enjoyed by members of all social estates who were engaged in any “proper” activity [Kjellson 1920: 170]. Vagrancy (lösdriveri) and defenselessness (försvarlöshet) were hence legal synonyms that justified coercive action [Johnsson 2016: 25; Uppenberg 2018: 90].

In Sweden, political debates and policies on the topic of mobile populations during the 19th century were hence drawn in two conflicting directions: on the one hand, the drive to control and secure the movement of these “dangerous” population groups, on the other, the need to satisfy the demand for labor, which tended toward greater liberties [cf. Foucault 2007].

Toward internal inclusiveness

I cannot account here for the full range of legal developments through which internal inclusiveness was accomplished, but will mention some of the most relevant ones. One concerns the role of the estates, the ständer. The four ständer—nobility, clergy, burgesses and commoners—were represented in Parliament until 1866, when representation by estate was replaced with a bicameral system. But even at the time of the 1810 parliamentary reform, the system of the four estates—which had first been introduced in the 17th century—was already considered outdated by many [Möller 2015: 21–34]. The ständer had different privileges and roles. For instance, the nobility were exempted from most taxes and held the highest offices. Their role was important, but it was not as dominant as in some other European countries, and feudalism did not really take hold in Sweden. The burgesses were town-based merchants, who monopolized business and commerce, took part in town governance and were subject to urban legal codes [Ibid.: 24].

The privileges and monopolies were gradually abolished, as legal reforms created a homogenized economic space and steps were taken toward legal equality within that space of activity. Internal customs duty had been in place since 1622, payable by anyone who transported goods for sale into cities (lilla tullen); it was abolished in 1810 [Lövgren 2018: 47]. Another old regulation that was abolished in the 19th century was the guild system. All burgess craftsmen had been organized in guilds (skrån), which controlled entry to and effectively monopolized their respective trades. The guild system was heavily criticized by a liberal opposition that became more vocal from around the 1840s. It was discontinued in 1846 for the countryside. In 1864, the more comprehensive law on freedom of commerce led to its abandonment in the towns as well; by this time, freedom of internal movement had been established [Stråth 2012: 59–60; Junestav 2008: 101].

Freedom of movement—and control of the movement of the poor

Reforms that establish freedom of exchange and commerce are necessary conditions for economic spatial integration to occur, but they are not sufficient. For such integration to be accomplished, the free movement of labor also needs to be in place [Harvey 2006: 375–376]. For legal equality to be established, people need to be able to move freely and on equal terms across the territory. If we only consider passport legislation, this was obtained in 1860. But movement was regulated in other types of legislation as well, and therefore I will also consider the vagrancy acts (försvarslöshetsstadgorna) and the master and servant acts (tjänstehjonsstadgorna) that regulated legal defense [Harnesk 1990: chap. 5; Olofsson 1996: chap. 2; Uppenberg 2018: chap. 5; Wallentin 1982: 8–11]. Such legislation was negotiated by the ständer and supported by broad swaths of the population [Lövgren 2018: 75; Johnsson 2016: 60].

Today, we tend to take for granted that we are entitled to leave our country of citizenship, as well as to move about and settle freely within it. These are also recognized as human rights in the UN Universal Declaration of Human Rights from 1948. But historically these rights have been severely limited, in Sweden as in most other countries [cf. Caestecker 2000; Fahrmeir, Faron and Weil (eds) 2003]. Passports were required for both domestic and international travel until 1860. Travel abroad was generally forbidden in 1620. The reason was the perceived need to increase the population, in accordance with mercantile thought, and particularly to retain needed able manpower, merchants and craftsmen. The nobility were exempt, since the right to study and learn abroad was among their privileges. Members of the other ständer were allowed to go abroad, for instance for the purposes of trade and learning their crafts, but to do this they needed special permission. There were tough punishments for those who traveled abroad without the proper documentation and permissions. The loss of inheritance rights was one consequence, and in the late 18th century one could be sentenced to death if found guilty of this for the third time. The prohibition particularly targeted servants, who were mobile because they did not own any land and were therefore more likely to leave. This group was not represented politically at either the national or local level, which facilitated this prohibition [Lövgren 2018: 48–51, Losman 2005: chap. 25].

Passports for internal travel were required from the 16th century, and went through many different reforms after that [Johnsson 2016: 100–104, 379–383]. For a very long time it was not considered self-evident that it was the role of the state to issue passports and to inspect them. For instance, the inns where travelers stayed overnight were among those engaged in control over movement. They were ordered by law to keep records of travelers and to deliver these to the Crown bailiff (kronofogden) (to the police after 1917) [Losman 2005: chap. 25]. Passports were also issued by many different authorities. The local administration office (magistraten) was the main one, and the one that common people turned to. Other examples included military commanders, university chancellors and even the inspectors of student associations [Rosander 1978: 13]. Employers—and masters—provided certificates for their servants to carry when traveling, so that they would not be mistaken for vagrants [Lövgren 2018: 95]. For a long time, the information given in passports was not uniform or standardized, especially not before 1812 [Ibid.: 149]. They usually included the person’s destination, reason for traveling, duration of the document’s validity, and some description of the traveler’s physical appearance. But some categories of travelers (for instance Jews, Roma people, peddlers) were required to complement this documentation with health certificates since there were prejudiced beliefs that they spread venereal diseases [Rosander 1978: 12]. From 1824, it was compulsory for the defenseless (unemployed) to include information on their reputation in the passports [Lövgren 2018: 85].

Passport controls have served different purposes throughout history. In times of wars and coups, apprehending spies, traitors and defectors is a core motivation. But a major purpose has always been to control the movement of the poor, and this was the primary purpose during the 19th century. The ambition was not only to prevent begging and criminality but also to defend the existing order from the perceived social threats associated with vagrancy and unemployment [Petersson 1983; Johnsson 2016: 99–102, chap. 4]. This was the objective that motivated an 1812 law on domestic passports, as well as follow-up regulations a decade later that were meant to make it more difficult for these “suspicious” people to acquire such passports. The norm was that people be employed, self-supporting and sedentary, and all mobile poor should be detected, compelled to work or sent back to the localities that were responsible for them [Lövgren 2018: 71–72, 85].

The passport was an important tool in managing internal movement, and thereby to reach the state’s objectives in other areas of legislation as well. The master and servant acts regulated work relations generally. The first of these dates from 1664, and the last from 1833 [Uppenberg 2018: 113]. The 1833 act applied particularly to maids, farm workers and statare. ⁵ It specifies that servants needed to be God-fearing, faithful, hardworking, obedient, sober and to have moral standing. The master was required to provide for the servant, including in the case of illness. The master should treat him or her (literally: “it”) with kindness and compliance, if it so deserved. If not, the servant should be treated with strictness and severity. The master had a legal right to exert corporeal punishment on servants, although this was restricted to the young in 1833. It was prohibited for servants to travel without permission from their masters or to leave before the working year had ended. If they did so, masters were permitted to bring them back home with the use of force (Tjänstehjonsstadgan 1833 § 1, 5, 7, 10, 14, 44, 50). In some parts of the country, servants were not allowed to move to other counties in search of work at all—only marriage or inherited land were legitimate reasons for moving [Harnesk 1990: 35].

The master and servant legislation expressed a patriarchal, hierarchical and preindustrial view of society. It was based on Martin Luther’s “Table of Duties” (Hustavlan), which specified, among other things, that servants should obey and be loyal to their masters, just as children should obey their parents and wives should submit themselves to their husbands. At the same time, those in superior positions were charged with duties toward their subordinates, who should be treated with fatherly love and care. The master–servant relation was not restricted to the sphere of labor and wages, as in employment under liberalism; instead, it comprised the whole of the servants’ lives, including their moral and religious education. Another contrast was that the master had an intermediary position between the state authorities and his servants, who were his protégés first and state subjects/citizens only second [Harnesk 1990: 40–48; Uppenberg 2018]. This view became outdated over the course of the 19th century, as waged labor expanded and labor market relations became more volatile and temporary. The master and servant legislation was repealed in 1926, but had already been out of use for some time [Stråth 2012: 370].

This law also regulated legal defense/defenselessness, mentioned above. Section 1 explains who was considered to have access to legal defense. This category included among others those who had legal permission to engage in the arts, in trade, studies, agriculture, crafts or shipping; property owners; those who live off their parents and relatives, and those who received poor relief. All others were required to attain employment in order not to burden society (Tjänstehjonsstadgan 1833 § 1). The law did not take into consideration whether there were any jobs available, but instead made employment the responsibility of the individual.

If one was defenseless, one could be treated as a vagrant. An 1802 law explained that “each member of society has a duty to be of benefit to that society through proper occupation, and no vagrant or idler should be suffered, either in town or countryside” (my translation).⁶ A law passed two years later provided a broad definition of vagrancy, which it divided into ten different categories.⁷ Among them were ex-convicts, who were either unemployed or had disobeyed their sentenced restrictions on visiting certain cities or locations; unemployed journeymen; discharged soldiers; foreign defectors; and all those without legal defense (försvarslösa), including Roma (zigenare) and travelers (tattare). All these categories of people were to be sentenced to work for a undefined period of time as punishment for vagrancy. Such sentences were to be passed by county governors, who represented a state and not a local or municipal authority [SOU 1923: 17]. The volume of sentenced vagrants grew steeply. The workhouses did not suffice, and vagrants continued to fill up prisons. Nilsson shows that the increase in prison inmates noticeable in this period was to a large extent attributable to vagrants. These had in most cases not committed criminal acts but were only guilty of a “status crime”. Most of them were men, but there were also women with children [Nilsson 1999: 121–123; cf. Johnsson 2016: 17–21, 77–82, 511].

Because of the practical problem of crowded prisons, the definition was narrowed in 1819, but this was soon criticized for being too lenient. A new act in 1833 returned in many ways to the earlier and harsher approach.⁸ A distinction was drawn between those who were only defenseless and those who were defenseless as well as depraved, that is, had been found guilty of certain crimes. The former were first allowed some time to find employment, and were also issued a passport to travel to a destination where that could be obtained. If this was not successful, the defenseless person was taken into public care. He or she could be sent to a workhouse, or to the newly instituted pionierkåren (later kronoarbetskåren)—a work corps modeled on the military, where he would both work and receive moral education. If none of these were available, he or she was sent to a correctional institution, where he or she would be kept apart from criminals. Those who fell into the category of depraved and defenseless people, however, were to be sent directly to correctional institutions. For both groups, the punishments were at this point not time-limited [SOU 1923: 18–19; Nilsson 1999: 104, 121–126, 200–207]. In 1846, a new regulation made punishments limited in time. Defenseless people who had committed some kind of crime were to be held for either three or four years, while all other defenseless people were held for two [SOU 1923: 20].

In 1860, the passport laws were repealed. As in many other European countries, from this point on until the outbreak of the First World War, people could normally leave and enter the country, and travel within the country, without passports. The reasons for this change were the liberal and market-oriented ideas that had already dismantled the guild system and established free trade and commerce. Liberal parliamentarians and commercial interest groups were upset about passports’ inefficiency and their detrimental consequences for trade [Lövgren 2018: 137–141].

At this point it is probably reasonable to say that freedom of movement for labor was established, which meant that an important step toward legal equality and internal inclusiveness had been taken. But there were important exceptions, deriving from enduring concerns about the mobility and circulation of the “deviant poor” [Althammer 2014]. The laws on defenselessness and compulsory labor were in force until 1885, and in the meantime the unemployed/defenseless could not avail themselves of the new freedom of movement. Other exceptions were former workhouse inmates, peddlers and small-scale traveling tradesmen, all of whom still needed passports [Lövgren 2018: 137–141].⁹

When defenselessness was removed from the legal vocabulary in 1885, a new law on vagrancy was brought in. Merely being unemployed was not reason enough to be considered a vagrant; the law defined such a person as somebody who roamed about from place to place, who did not have any means of subsistence, who did not seek employment and whose ways of life threatened public security, order and morality. Johan Edman underlines the durability of this vagrancy law. It was, with some changes and amendments, in force until 1965, when it was replaced by similar legislation on dangerous antisocial behavior (samhällsfarlig asocialitet). Only in 1982, with the reform of social services, was vagrancy finally removed from legislation. Edman therefore contends that for the vagrant, the long 19th century was indeed very long, lasting almost into the new millennium [Edman 2008: 132].

Toward external exclusiveness

The second dimension of modern citizenship is its external closure. In the early 19th century, the citizen–foreigner distinction was quite insignificant, but this changed, and toward the end of the century it was of key importance in defining responsibilities and duties [Caestecker 2003: 131–132]. It gradually became more significant than other social divisions (for instance, estates and guilds), which did not mean that these other identities were abolished or became completely irrelevant [Fahrmeir 2021]. The citizen–foreigner distinction, moreover, was formulated as independent of residence, so that residence was conditional on citizenship rather than the other way around [Brubaker 1992: 70]. In Brubaker’s account, an important factor that spurred this major social change was the mobility of the poor, which intensified questions and conflicts over the responsibility for those that could not support themselves.

Poor relief

Historically, begging has been an important way of alleviating poverty. Begging has been conceptualized and handled differently over the centuries, and in the 19th century great efforts were made to suppress it [Dahlstedt, Härnbro and Vesterberg 2019]. The main focus here is on the division that was drawn between those who were permitted to beg and those who were not. Already in 1698, begging had been restricted to those who were considered to belong to the locality. For “foreigners” from other places, begging was prohibited and the guilty were deported [Lövgren 2018: 69].

Another major concern was the responsibility for poor relief claimants. The poor had for many centuries been the responsibility of local church parishes [Kaspersen and Lindvall 2008]. The poor and unpropertied tended to move in search of work and livelihoods, and sometimes ended up claiming poor relief in other parishes. Conflicts arose because of the difficulties of establishing which parish was responsible for a particular pauper—the one where he or she was born or the one where he or she had moved to. In 1788 the state intervened, and it was established in law that responsibility fell on the parish where the poor person had right of domicile (hemortsrätt), which was usually where he or she was registered. The law also established that the local parish had the right to prevent the settlement of people that they deemed likely to end up claiming poor relief at some point in the future. The parishes used this right very broadly, excluding not only those who could not work, but also unmarried mothers, families with many children, ethnic minorities and other unwanted persons. This had a great impact on the conditions of the mobile poor, making them increasingly exposed to punishment for vagrancy and very vulnerable in times of famine [Ericsson 2016: 93; Johnsson 2016: 109–111; Jordansson 2008; Montgomery 1951: 40–46].

Another consequence was that municipalities avoided registering the mobile poor and thereby providing them with hemortsrätt, because they wanted to evade financial responsibilities. Since the mobile poor were not registered, they could not attain passports, and since they did not have passports, they could not move to find work without running the risk of being detained as vagrants. Anna-Brita Lövgren therefore concludes that they were the “undocumented workers” of those days [2018: 148].

There were continuous struggles between parishes over responsibility for the poor [Ericsson 2012, 2015: 48–51]. Given these difficulties, the suggestion was made to nationalize poor relief. The arguments were that this would relieve struggles between localities, ensure similar levels of support across the country and even out the costs for poor relief between parishes. The counter-arguments were that nationalization would make poor relief more costly because control would become more complicated and inefficient. The responsibility therefore stayed at the local level [Montgomery 1951: 69–70]. But legislation about vagrancy, domicile, and compulsory service complemented it at the national level.

The poor relief act of 1847 established that society did have a responsibility to provide scanty assistance to the poor. It also ended the parish right to prevent settlement. But the responsibility for poor relief still lay with the parish of domicile. The law scaled down the responsibility of masters, and instead put the responsibility with local administrations. The role of the church was also reduced, as a public poverty board (fattigvårdsstyrelse) was to be set up in every municipality. Those who applied for poor relief but were denied were now given the right to appeal. The law also expressed a view of unemployment which at the time was new and controversial: those who on account of their unemployment had previously been subject to vagrancy laws should now be offered work instead of punishment [Montgomery 1951: 99–107; Vikström 2006]. This somewhat more liberal and softer approach to poverty was repealed with the 1871 poor relief act, which again limited the responsibility of state and municipalities. Poor relief was not to be offered to able-bodied adults. It was no longer the duty of municipalities to provide jobs in difficult times, and the right to appeal was removed. Forced labor was reintroduced as a punishment for vagrancy [Kaspersen and Lindvall 2008; Montgomery 1951: 113–119; Stråth 2012: 372–375].

During the last two decades of the 19th century the workers’ movement began to organize. There were increasing demands for poor relief to be seen as a social right, retaining the dignity of the person rather than declassifying him or her. Moreover, under the influence of Bismarck’s reforms in Germany, ideas for social insurance of various kinds appeared. Laws were passed on worker protection for factory workers (1889), state-backed health insurance (1891), workplace accidents (1901) and old age pensions (1914). The poor relief act of 1918 reflected this move toward a rights-based view. Among other things, the level of assistance should no longer be “scanty”, as the law had previously stipulated, but more comprehensive. Municipalities had responsibility to set up old people’s homes and to pay for hospital care. Moreover, the practice of auctioning off the care of orphans and elderly poor to the lowest bidder was now made unlawful [Åström, 2008; Edebalk 2008; Qvarsell 2008; Stråth 2012: 375–382, 409; Swärd 2008].

Brubaker argued, in the case of Germany, that there was a clear link between the development of poor relief policy and the establishment of modern citizenship. We can see this link in Sweden too. The citizenship legislation of 1894 was elaborated in cooperation with the other Nordic countries. This cooperation was initiated because of the migration of poor people, as many poor Swedes took on seasonal work in Denmark and Norway. The dilemmas that we have previously seen at the local parish level arose here too. Conflict was building up concerning which country was responsible for the poor relief and health care of which migrant pauper. After a period of diplomatic tension, in 1888 the authorities of the three countries decided to cooperate on distributing responsibilities and regulating repatriation. In this process they realized that it was necessary to harmonize citizenship laws. In Sweden, there had been some regulations before, but it was now that firmer nationality laws developed. Among other things, these set out explicit rules for membership acquisition both through naturalization and at birth. It was the first time that ius sanguinis was established in law—which signaled a type of membership not reducible to residence in the territory [Bernitz 2012: 2–3; Ersbøll 2015: 8; Willerslev 1983: chap. 7].

Emigration

Brubaker delimited his analysis to the internal migration of the poor. But it seems to me that in Sweden at least, membership of the state was also worked out in relation to outward mobility of the same group, that is, emigration. When labor was abundant, people were starving and the mobility of vagrants was threatening, authorities did not really object to people leaving. As we saw above, emigration was legalized in 1860, around the same time as in many other European countries [Green and Weil 2007; Karadja and Prawitz 2019; Osterhammel 2014]. Emigration really took off with the failed harvests of the 1860s, and it peaked between the 1880s and 1920 [Stråth 2012: 295; Västerbro 2018]. After a while, this voluminous emigration started to be seen as problematic. This was partly because of aggressive nationalism and biological racism, which emerged toward the end of the century, and made some politicians deplore that the sons and daughters of the nation were leaving and being replaced by “suspicious” foreigners. The latter were considered of “bad blood”– Jews and Roma in particular—or as potential revolutionaries [Ericsson 2016: 164; Svanberg and Tydén 2005: chap. 20; Hammar 1964]. But the concern over emigration was also, to no small degree, due to the lack of labor in certain sectors of the economy. A series of private initiatives tried to convince potential emigrants to stay home, and an ambitious official report was commissioned (Emigrationsutredningen) [Stråth 2012: 293–309; Sundbärg1910, 1911].

The concerns raised by emigration also marked citizenship legislation. Authorities now had to specify under what conditions citizenship would be retained or lost due to emigration. Return migration was quite considerable—around 200,000 out of the 1.2 million who emigrated between 1821 and 1930 came back to Sweden [Stråth 2012: 293–294]. Citizenship rules needed to be worked out with this in mind, too. In the 1894 act it was specified that emigrants lost their citizenship after ten years abroad if they did not actively communicate that they wanted to retain it. But it was also possible to regain it if they returned and settled in Sweden. The condition for this was that they had not in the meantime acquired another citizenship. In 1909 this was amended so that even those people could regain Swedish citizenship upon resettlement in Sweden, provided that the other nationality was in a country with which Sweden had a bilateral agreement (the USA and Argentina), and provided that he or she renounced the second citizenship when becoming a Swedish citizen again [Bernitz 2012: 3]. We can therefore see that emigration too, and not only domestic migration, raised concerns which necessitated the formulation of citizenship rules that were partly independent of residence.

Summing up

In this article I have pursued the argument that the migration of the poor, and particularly the varied attempts at controlling it, was an important factor in the evolvement of modern citizenship. It can be placed within the context of the historical continuity of governmental concerns with governing the circulation of people and things. Moreover, the establishment of modern citizenship has been, and is, a vital component of the modern state, which theories of the state tend to overlook.

I have discussed the case of Sweden in relation to the two dimensions of citizenship. The first is internal inclusiveness. Over time, membership of the state became more important than previous social memberships, and various reforms established greater economic freedoms. When it came to the movement of labor, there were desires both to free it up for the benefit of industry and trade and to suppress the movement of the deviant poor. Freedom of movement was established in 1860, but there were exceptions for vagrants and for others who were perceived as suspicious or as undeserving poor.

The second dimension was external exclusiveness. We have seen that the responsibility for the poor was traditionally a concern for local authorities, and the role of the state grew when it specified rules for domicile and belonging in relation to responsibility for the mobile poor. The state also continuously worked out the distinction between the deserving and the undeserving poor, where the former were entitled to poor relief and the latter were treated as defenseless vagrants. This process also had significant international aspects. Rules on national citizenship were elaborated in cooperation with other Nordic countries, in order to settle which jurisdiction was responsible for which poor migrant. Moreover, efforts at managing emigration had effects for citizenship in the Swedish case. The modern form of citizenship, which separated membership from residence, was hence worked out both in relation to the domestic and the international movement of the poor.

My analysis has found that the movement of the poor in the 19th century preceded efforts at controlling them. As a response to various macro-sociological developments, as well as subjective motivations, the poor took to the road in large numbers. This caused a regulative frenzy among decision makers, who aimed at managing and controlling the mobility. This finding speaks to the literature on “autonomous migration”—which holds that poor migrants are actively involved in shaping institutions and social relations, since they exercise a particular form of dispersed and nonintentional agency [Mezzadra 2011; Rodrìguez 1996]. In the words of Dimitris Papadopoulos and Vassilis S. Tsianos: “The autonomy of migration approach […] means that the movement itself becomes a political movement and a social movement […] migration is [seen as] autonomous, meaning that it has the capacity to develop its own logics, its own motivation, its own trajectories that control comes later to respond to, not the other way round” [Papadopoulos and Tsianos 2013: 184]. This means that modern citizenship was not only worked out in response to the workers’ movements, or other coordinated popular movements aiming at institutional power. It was also in part a response to the autonomous mobility of the poor. With this in mind, we should therefore recognize migrants’ role in the development of modern citizenship.