Migration and the critique of ‘state thought’: Abdelmalek Sayad as a political theorist

This article argues for reading the Algerian-French sociologist Abdelmalek Sayad (1933–1998) as a political theorist of migration. Various contributions have recently called to move away from the court-like assessment of claims by host states and foreigners and to engage more frankly with empirical work more attentive to concrete experiences and power relations. I contend that Sayad’s sociological work constitutes a substantial empirical and normative resource for ethical and political theory of migration, pointing to the persistence of ‘state thought’ and presenting original normative perspectives on emigration, inclusion in democracy, naturalization or postcolonial relationships. Such a reading of Sayad from a political theory perspective would then constitute a prime example of the cross-fertilization of empirical and normative approaches.

The ethics of migration has constituted an independent field since the 1980s, asking normative questions about the scope of justice and democracy, i.e. about the fairness and legitimacy of the exclusion of newcomers from territory and citizenship (Fine and Ypi, 2016;Sager, 2016a). But the limited scope of moral philosophy, mainly written from the perspective of a host state assessing the claims of immigrants (Miller, 2016;Wellman, 2019), has recently led many to argue for renewing this normative approach. The analytic method consisting in 'articulat[ing] a few broad ideals or principles to which [philosophers] can appeal in assessing the justice of particular admissions policies' (Jaggar, 2020: 91) is criticized for being epistemologically limited and politically biased (Fine, 2020;Finlayson, 2020). Instead, one should ground the conflicts of justifications in non-ideal and noninstitutional disputes, and move from 'right-and-principle-based arguments' abstracted from concrete experiences, to an approach based on 'particular experiences of migrants with named social identities' (Reed-Sandoval, 2016: 26; see also Fine, 2019). Empirical approaches should be more widely included in migration ethics to better take into consideration 'historical injustices and relationships of domination and oppression' faced by migrants (Sager, 2020: 4). These facts are too often brushed aside for the sake of generalization. Adding more context and diversifying the claims and experiences considered for normative reasoning would enhance 'intellectual humility' (Ackerly et al., 2021: 4) and offer a more realistic picture of power relations in migration discourses by including the voices of those affected by them. When philosophers are more engaged with social sciences and social scientists more aware of the normative reach of their work, studies in migration ethics become more sensitive to marginal cases and voices and to the political dimension of their epistemological background.
In political theory and in sociology in general, proposals have been made to bridge this gap between the normative and the empirical. On the one hand, theorists import empirical methods within normative reasoning (Baub€ ock, 2008;Cabrera, 2010;Doty, 2009;Gerver, 2018;Longo, 2017;Tonkiss, 2013) and insist on the relevance of their approach for more positivist inquiries (Barry, 2002;Gerring and Yesnowitz, 2006;Shapiro, 2002), especially when it comes to a highly politicized and morally loaded issue such as migration (Carens, 2018). Recent methodological studies have described how to do 'qualitative political theory' (Cabrera, 2009), political theory 'in an ethnographic key' (Longo and Zacka, 2019) or 'grounded normative theory' (Ackerly et al., 2021) that would be informed by empirical data and more inclusive of 'what people think ' (de Shalit, 2020). On the other hand, social scientists, like Andrew Abbott (2018: 172), seriously consider creating 'for sociology a normative subdiscipline equivalent to political theory in the discipline of political science'. He adds that 'this specialty would produce formal, rigorous analyses of the political and normative questions on which sociology bears, resting its arguments on a set of canonical moral texts' (Abbott, 2018: 172). Tariq Modood (2020: 32) also 'encourage[s] engagement with normative questions for the sake of the quality of science rather than draw [ing] positivist boundaries around itself'. He adds that ' [sociology] thus has an active interdisciplinary relationship with rather than a merely dependent relationship upon political theory' (Modood, 2020: 35).
Regarding migration studies more specifically, Carens (2018) has called for a two-step change in migration ethics. The first step applies a more systematic critique to the gaps and contradictions between normative commitments and their faulty or limited realization as they are transformed into policies and institutions. Second, Carens suggests moving from the moral question of 'what is right in principle' to a realist one that asks 'what is the best we can realistically hope for under this particular set of circumstances?' (Carens, 2018: 27-28). To achieve such a methodological move, I contend that we need to expand the literature considerably and rethink our approach to reading political theory materials. Specifically, we should continue to include empirical studies, but we must incorporate the normative work of social scientists as well.
To substantiate this argument, I wish to reconstruct the contribution of a sociological work to the ethics of migration. I will focus on the writings of the Algerian-French sociologist Abdelmalek Sayad (1933Sayad ( -1998. There are two reasons for choosing Sayad as a good case study for articulating sociology and normative theory. Firstly, Sayad is a figure who has long remained in the shadow of his peers, and very little of his work has been translated into English. This article therefore has a descriptive dimension to recover Sayad's ambitious but partially ignored intellectual project. Secondly, I contend that his sociological work -a sociology that could be called critical in the aftermath of Bourdieu (P erez, 2020), or normative following Modood's (2020: 30) terminology -offers rich normative, conceptual and empirical material for ethical and political theory of migration. I will show that it participates in the deconstruction of methodological nationalism and opens up normative perspectives on emigration, inclusion in democracy and postcolonial relationships.
My goal is thus to provide an interpretation of Sayad's work on migration as both an empirical and normative contribution to political theory. While Sayad is mainly read among social scientists for his work on Algerian immigration in France, political theorists rarely discuss his arguments. Thus, I make a twofold argument: firstly, Sayad's work illustrates how to enrich normative reasoning based on empirical research; secondly, reading his work while using the lens of normative theory brings original arguments to the fore regarding statist and nationalist biases, inclusion, naturalization and moralism (or politeness) in a postcolonial migration context.
There are indeed epistemological benefits in considering the normativity that surfaces in an empirical body of work and among the empirical findings for normative reasoning. I follow Abbott (2018) in his ambition to build a common canon for normative and empirical research, encompassing the ethics and sociology of migration. I also agree with Ackerly et al. (2021: 5), who identify four main functions that empirically informed theory should fill: it should enrich normative arguments, contextualize and amend these arguments according to empirical data and include dominated voices and respect their contribution as opposed to distorting their meaning for theoretical purposes. These commitments are at work at two levels in this article.
At the first level, Sayad's work with Algerian immigrants in France represents well the effectiveness and value of this methodology. Precisely, Sayad aims 'to investigate and bear witness', to stand in 'active solidarity with those he was taking as his object', (Bourdieu, 2006: xii) and to critically expand the language and normative justifications of a fairer inclusion of migrants based on his qualitative research, which spanned more than three decades. Sayad's sensitivity to the power relations that surface in his semi-structured interviews and his quest to 'give [migrants] their voice and restore the coherence and complexity of their discourse' (Noiriel, 2006: 106) are exemplary.
At a second level, I consider Sayad a crucial informant who offers 'a very personal attempt to come to terms with the brutal realities of Algeria's colonization, military occupation, political repression, and its continued social, economic, and cultural dependence on France' (Massey, 2006: 113-114). In reading his work as a first-hand account of the problems he discusses, as 'he was the phenomenon itself' (Bourdieu and Wacquant, 2000: 117), I follow the commitment to expand our normative repertoires, revise normative claims based on the empirical material that Sayad's work and interviews provide and reassess a largely excluded contribution to migration studies. However, my goal throughout is to remain sensitive to Sayad's authentic experience in order to take his normative ambition to foster democratic inclusion seriously.
First, I must briefly give some context to Sayad's life, work and methods, and explain the normative role he attributes to sociologists (first section). I then argue that Sayad helps identify state thought and nationalist logics and offers important correctives to ethical approaches; his work shows the limitations of any studyeven a philosophical one -that considers only the perspective of the host society and neglects emigration (second section). In the third section, I elaborate on the main themes that illustrate Sayad's contribution to political theory and migration ethics, namely his critique of the moralization of migration through his analysis of politeness, naturalization and integration. I reconstruct in the fourth section his argument for democratic inclusion and voting rights for foreigners. I show how his attention to social impediments to participation avoids the legalist take on enfranchisement. I finally argue that the connection he draws between differentiated voting rights and postcolonial remains could nourish a growing field of research on postcolonial relationships in migration ethics.

The normative position of a critical sociologist
From the early 1960s to the end of the 1990s, Sayad wrote and published a vast number of texts on the dynamics of migration between Algeria and France. Sayad had developed an ambitious research programme to address the 'emigration-immigration' relationships between France and Algeria, i.e. the social and political causes, types, transformations and effects of emigration in Algeria and immigration in France. Pierre Bourdieu, a long-lasting friend and colleague with whom he wrote his first book in 1964 and edited posthumously his main work (the only one translated into English), The Suffering of the Immigrant (Sayad, 2004). Entitled more explicitly in French as The Double Absence: From the Illusions of the Emigrant to the Sufferings of the Immigrant, the book synthesizes Sayad's work on the dual experience of emigration and immigration. He analyses the political and social difficulties that typify leaving one's country yet remaining half present in the immigration community. He considers Algerian migration to France as exemplary of the power relations at stake in migration, noting that colonial, racial, religious and linguistic forms of domination persist between a recently independent country and its former colonizer. In the other edited volumes of his papers, he focuses on the illusion of temporary migration (Sayad, 2006a), the political positions of different generations of migrants (Sayad, 2006b), the social uses of culture (Sayad, 2014a) and issues related to school and immigration (Sayad, 2014b).
For Sayad, migration is a 'total social fact' that sociology is the best equipped to understand. 1 He argues against what he considers as generalizations in philosophy, law and history. Such approaches would not adequately take into consideration the social conditions of migration, they remain partially blind to social discrimination despite formal rights and they neglect the specific details of a colonial and postcolonial relationship such as exists between France and Algeria. One cannot discuss the reasons for migrating and how they are assessed by a state, the principles guiding migration policies or the political rhetoric mobilized for or against non-citizens without the sociological gaze unveiling the objectivity of the social conditions that produce these phenomena. 2 However, without clearly acknowledging it, Sayad does mobilize different disciplines. He historicizes Algerian migratory flows, distinguishing three discrete 'ages'. 3 He studies the evolution of legal categories and constraints that shape the mobility and presence of non-citizens in France. He elaborates on the normative articulation between democracy and exclusion and the scope of a right to have rights. I believe Sayad could be read at the crossroads of these disciplines to fully uncover the significance of his work. More specifically, there is a distinctive normative aspect in his sociology of migration that, I argue, is relevant for political theorists. In fact, contrary to what strict empirical sociologists would recommend (Besbris and Khan, 2017), Sayad does engage with normative questioning and attempts to 'justify the mission of the engaged and public sociologist who can remedy the forms of social domination' (Noiriel, 2006: 105). First of all, then, there is a politics of his sociology inherent to his positionality. Sayad positions himself as a sociologist who gives a voice to the political 'mutes' that are migrants (Sayad, 1990: 7). 4 Doing this, he aims at describing and, hence, reshaping the power asymmetries between nationals and foreigners, citizens and non-citizens. As Daniel Chernilo (2014: 350) explains: the interests of less powerful actors ought to be favoured against those of more powerful ones. The sociologist can be seen as the reflexive amplifier with the help of which subordinate actors get their interest advanced wherever and whenever this is needed.
In the opposition between dominant and dominated, nationals and foreigners, the critical sociologist grants the latter a normative privilege. This should, in short, balance power and thereby profoundly disrupt the social order.
In Sayad's case, some biographical elements further explain this position (Jammet, 2013). Sayad was born in Algeria in 1933. He first worked as a teacher in Algiers but eventually entered university in 1957, which was rare at the time for Algerian Muslims (Jammet, 2013: 47). During the Algerian war for independence, he wrote about the necessity for both independence and reconciliation and defended the figure of the liberal. In 1958, he met Pierre Bourdieu, who had decided to stay after his military service to teach and write on the sociology of Algeria. Sayad was a student of Bourdieu's, and they began to engage in fieldwork together. Unable to obtain work following independence, Sayad left Algeria in 1963 and settled in Paris. 'Ten years of illness, precarious employment, lack of resources, lack of publication, questions and doubts' followed (Jammet, 2013: 124) before Sayad found a position as a professional sociologist and began his work on migration. Himself an immigrant, an Algerian in France, a marginalized person in the profession and a specialist in a peripheral topic at the time, Sayad's research and arguments are inherently linked to his personal life. As Bourdieu and Wacquant (2000: 117, original emphasis) argued: All this Sayad knew or discovered because he was more than a scholar of immigration: he was the phenomenon itself. [. . .] the brute facts of imperial oppression, chain migration, community dislocation and fractured acculturation were constantly with him because they were within him: they were his entrails, his eyes, his soul.
His positionality as an immigrant in France and as a French-Algerian binational with a lived experience of colonization necessarily involves a specific standpoint that is reflected in his writings. From an empirical perspective, there is an epistemic advantage to having a privileged access to interviewees through cultural and linguistic proximity. This appears clearly in his subtle and meticulous transcriptions and translations of his interviews, which are always present in his books. From a normative perspective, Sayad's position makes him more sensitive to the 'suffering of the immigrant', to the unequal political dynamics between Algeria and France and to the need for foreigners to be empowered. Sayad also worked at several stages of his career for the Algerian and French governments, advising the former on agrarian policies and the latter on immigration and schooling policies, which gave him a first-hand perspective on the political and institutional dynamics governing migration and integration (P erez, 2020). I would therefore characterize Sayad's ambition as epistemic, socio-political and ethical: epistemic in the sense that it includes the perspective of the dominated, which represents a whole social world that had been ignored or excluded previously; socio-political because Sayad gives social actors the opportunity to reappropriate their social trajectory; and ethical as the socio-analysis is also a 'self-socioanalysis'. This final element signifies the circularity between the trajectory and position of the sociologist, the tools and results of the sociological analysis and the mobilization of these results to objectify one's social condition as a researcher facing particular interviewees. In other words, politics, power, domination and the possibilities of emancipatory knowledge and practice are always at play during the research process. Thus, it is essential to fully grasp the normativity of both the analytical claims made by Sayad and the critical sociology methodology, which assumes that an analytical and objective approach is impossible within the social world as it is comprised of hierarchies and power relations. As Bourdieu (1991: 127-128)  In Bourdieu's sense of an unveiling of the social world's conflicting and competitive structures, and of the resources of the actors to interact in them, sociology is at once sociological and political, empirical and normative, neutral and engaged. Sayad's work is thus normative and deconstructive. It uncovers the classifying and excluding structures on which the social order is based and by which it is justified. The nation is one of these structures, both constituted by the state and constitutive of it. Sayad's approach could therefore be summed up as a critique or deconstruction of nationalist logics.

Deconstructing nationalist logics
According to Sayad, nationalism constitutes a 'doxa' or 'state thought', i.e. the unconscious acceptance of officially accepted categories and legitimating discourses produced by state institutions (Sayad, 2004). We think in terms of labels and classifications, 'structuring structures' (Bourdieu, 1994: 13;Loyal, 2017) produced by nation-states that naturalize (political) borders and (social) boundaries. Naturalizing distinctions entrench hierarchies between those who are rightly and legitimately here and those who, one way or another, have disrupted the national order by crossing borders, settling in and claiming rights against a state that is not originally theirs. Against the cognitive and social background of 'state thought', immigration appears as 'tainted with the idea of being at "fault", with the idea of anomaly and anomie. The immigrant presence is always marked by its incompleteness: it is an at-fault presence that is in itself guilty' (Sayad, 2004: 283). Sayad's critical sociology aims precisely at uncovering and (hence) denouncing the nationalist doxa and its tireless work of classification (and therefore exclusion), making informed resistance possible: 'Immigration is undeniably a subversive factor to the extent that it reveals in broad daylight the hidden truth and the deepest foundations of the social and political order we describe as national' (Sayad, 2004: 280).
'State thought' encompasses various normative precommitments, or logics, that still infuse some political theories of migration. Elsewhere, I have identified a logic of membership (giving unquestioned priority to those already there) and a logic of consent (foreigners must consent a priori to an indisputable set of principles which they have had no part in establishing and have no say in) that we find in communitarian (Walzer, 1983), republican (Pettit, 1997) and liberal-nationalist (Miller, 2005) defences of self-determination and the right for states to control their own borders (Boudou, 2018). Consider Wellman's argument that 'every legitimate state has the right to close its doors to all potential immigrants' based on freedom of association, similar to a golf club controlling membership (Wellman, 2008). This kind of reasoning is antithetical to what Sayad writes, not so much due to an alternative formulation of the principle of self-determination, but because the framing of the discussion actively attempts to depoliticize immigration. The club metaphor, for instance, is so inaccurate that it is irrelevant in any discussion of migration and, what is more, it belies linguistic and cognitive biases that construct migration as a demand to enter an already constituted club (Sayad, 1990: 12-13). Transforming a complex political process into a club-like assessment of entry claims may be conceptually productive, but it constrains the empirical appraisal of inclusion and exclusion (Mills, 2005) and depoliticizes migration. As Sayad powerfully claims, echoing the realist critique against moralism in political theory (Rossi and Sleat, 2014): The immigrant is a paradigmatic example of a type of object we would like to deal with in purely ethical terms. The most pernicious way of subverting immigration by ensuring that it is the subject to the most total domination possible is to depolicize it. And the best way to depoliticize a social problem is to technicalize it or absorb it completely into the field of ethics. (Sayad, 2004: 224) One could argue that the club is merely a metaphor, but Sayad pays particular attention to language, as it results from 'state thought'. Moral philosophers describe abstract rights to enter, leave or exclude as common sense language; however, common sense is neither neutral nor natural and, in fact, perpetuates the negation of the political nature of migration to make it as unproblematic as entrance into a golf club.
Sayad uncovers a logic of interiority, which defines a political community as naturally closed and autonomous. This logic renders immigration necessarily problematic and the immigrant a perpetual suspect (Sayad, 2004: 45, 206). Migrations merely appear as incidental phenomena that may be acceptable or legitimate if they serve the interests of the community -hence the rhetoric of costs and benefits of immigration that Sayad has underlined, and the construction of migrants as free-riders benefiting from a society to which they have not contributed (Sayad, 2004: 76-80). This kind of normative logic plays a significant role in liberal theories stemming from a contractual perspective, such as Rawls' ideal assumption of a society as 'a closed system isolated from other societies' (Rawls, 1999: 7). The conception of 'associative ownership' that grounds the rights of citizens to exclude newcomers on the basis of their participation in common institutions (Pevnick, 2001), or Michael Blake's defence of exclusion based on the right not to support the protection of would-be citizens' additional rights (Blake, 2014), rely on this construction of migration as a secondary phenomenon, in the sense of happening after the fact, and a static phenomenon. For Blake, what gives citizens the right to refuse entry is the objective sharing of reciprocal obligations, rights and duties, thus the existence of a legal community that citizens alone can decide to extend to newcomers. Citizens are not obliged to realize the rights of all people. Following Sayad's critique, such arguments are based on the consideration of citizens' and states' decisions whether or not to share their rights with others, which is inherently statist and moralistic: Ethics and politics complement one another and collude in converting the rights possessed by this category of subjects (who do not have the right to have rights because they are not nationals) into duties, or into the other party's obligations towards them. Rather than recognizing our partner's rights, we are careful to describe and represent them to him as duties we take upon ourselves, as acts of generosity or unilateral largesse. (Sayad, 2004: 224) The problem is that a legal community, its institutions and obligations, is not like a cake that shrinks in size as more and more guests join. On the contrary, they function as a living language: the more people who speak it, the more beneficial it is for everyone. As John Dewey writes well in another context, 'sharing a good or value in a way which makes it social in quality is not identical with dividing up a material thing into physical parts. To partake is to take part, to play a role' (Dewey and Tufts, 1936: 383-385). The state thought and nationalist logic that Sayad uncovers tend to regularly ascribe migrants with latecomer status, always taking something from citizens who are already established. Since mobility blurs material (political) borders, migrants are perceived as disrupting the principle of the constitution of national groups.
Under this logic, Sayad shows, migration is mostly described as temporary, while it happens that migrants eventually settle in. In fact, migration calls into question the distinction between the temporary and the permanent, creating what Sayad (2004: 125) termed the 'impossible ubiquity' of the emigrant-immigrant, who is neither absent from the emigration community, nor present in the immigration one. This permanent ubiquity of migrants -their perceived incongruous and temporary presence -justifies the inequality of status between citizens and (resident) foreigners, which further maintains the latter in a situation of legal and political domination (discrimination, non-representation). This makes it impossible for them to challenge social domination (vexation, humiliation, invisibilization), and 'this social discrimination, in turn, legitimizes the exclusion of immigrants from the body politic' (Saada, 2000: 41).
This vicious circle between fact and law also occurs regarding the very designation of immigrants. Legally speaking, a foreigner does not have the nationality of his or her country of residence, and an immigrant was born abroad and came to settle. Although it is legally possible to be a national immigrant, or a foreigner without having migrated, the category of foreigner invades all spheres of social life, stigmatizing all those who do not correspond to the national 'type' (accent, name, skin colour, religion, etc.). Moreover, this stigma passes from one generation to the next, as those who remain 'enfants d'immigr es' (children of immigrants) can testify: 'they are "immigrants" who have not emigrated from anywhere' (Sayad, 2004: 291). Regardless of legal status, the nationalist vision of membership necessarily leaves a nativist imprint on the political community. The normative literature on the inclusion of immigrants does not clearly address these intergenerational migration stigmas fuelling daily racism, or the sociological impediments to immigrants exercising their rights, such as the right to vote (I will revisit this issue in the 'Democratic inclusion in a postcolonial context' section, below). Arguably, many normative theorists tend to adhere to 'state-based legalistic mechanisms' (Sager, 2017: 46) that focus on formal equality. Sayad's diachronic approach and the place he gives to migrants' family narratives shed light on these blind spots.
The nativist narrative for justifying borders in liberal democracies has also been the main target of non-nationalist approaches to migration (Abizadeh, 2012;Carens, 2013;Oberman, 2016). Theorists also moved away from the dismissal of migration as a relevant issue for theories of justice (Carens, 1987), and from realist doctrines that enclose democratic standards of legitimacy within the nation-state (Abizadeh, 2008;Benhabib, 2004). However, I contend that Sayad's work offers important correctives, as we have begun to see. From a methodological perspective, his analysis of 'state thought' and immigration paved the way for the critique of methodological nationalism (Wimmer and Glick Schiller, 2003;Sager, 2021) and the development of a transnational understanding of migration (FitzGerald, 2000;Martiniello, 2013). His whole empirical work shows the limitations of any study -even a philosophical one -that considers only the perspective of the host society and neglects emigration. Sayad insists on speaking of emigrationimmigration to highlight that any migratory movement involves a complex relationship between the society one leaves and the society one enters. Sayad analyses this relationship through time, identifying different ages of emigration and the 'disruptive effects of emigration' (Sayad, 2004: 116). Algerian peasants going to France on a mission for the village at home constitutes the first age (until the 1950s). Migrants of the second age (up until 1962) are more urban and individualistic. They do not relate as closely as before to their emigrant communities and identify more openly with their migrant status. This individual 'adventure' (Sayad, 2004: 40) brings them closer to the French community. Migrants of the third age become settlers, migrating with their families and slowly cutting ties with their communities of origin. They still suffer, however, from being 'torn between two "times," between two countries and between two conditions' (Sayad, 2004: 58). The sociology of emigration-immigration is useful in better understanding the before and after of the migrant's journey, i.e. how migration takes form in a specific life-world and shapes the dispositions and opportunities of integration in a new one. Considering emigration and immigration together may sound obvious, but the fact that emigration is rather understudied, compared to immigration, is itself a symptom of a focus on the receiving (Northern) states: Immigration results in a presence, and emigration finds expression in an absence. A presence makes itself felt; an absence is noted, and that is all. A presence can be adjusted, regulated, controlled and managed. An absence is masked, compensated for and denied. These differences in status determine the differences in the discourses that can be applied to both presence (immigration), which is amenable to discourse, and absence (emigration), of which there is nothing to be said except that it has to be supplemented. (Sayad, 2004: 120).
Such oversights have been persistent in political theory (Sager, 2018). Emigration is mainly considered as a 'brain drain' problem (Broke and Blake, 2015;Sager, 2016b). But even in cases of refusal to restrict emigration because of brain drain, the problem lies in the limited vision of what emigration entails and the enactment of the cost and benefit rhetoric identified by Sayad: 'Immigration is meaningful, and intelligible to the political understanding, only if it is a source of "benefits" or, at the very least, only if the "costs" imputable to it do not outweigh the "benefits" it may procure' (Sayad, 2004: 76). In other words, most of the ethics of migration is limited to the question of immigration, with some defence of an abstract right to leave and considerations of its costs. Furthermore, what explains migrants' loss or retention of rights in their country of emigration are the bilateral relations between the countries of emigration and immigration. This relationship is necessarily tainted by domination, which partly explains the migration dynamics but is denied in the name of sovereign equality of states. This is especially true in a postcolonial setting. Simply put, Sayad pays attention to the gap between de jure bilateralism and de facto domination that reveals unequal power relations at the level of nations as well as of individuals. It follows that considerations of justice must take into account the social and political losses that emigration involves in assessing the claim to immigrate -the 'break' with a whole society that Sayad (2004: 88) mentions -as well as the political dynamics between the countries of emigration-immigration.
Sayad's first main contribution relates to the rhetoric and politics of nationalism infused in both state practices and in theories that encompass statist conceptions of migration and focus solely on host states. As Bourdieu notes in his introduction to Sayad's book, 'analysts approach "immigration" -the word says it all -from the point of view of the host society, which looks at the "immigrant" problem only insofar as "immigrants" cause it problems' (Sayad, 2004: xiii). Sayad aims to unmask these conceptual frames that distort the objectivity of a social phenomenon. This approach is also well illustrated in his critique of the moralization of migration, which, I will argue, is of crucial importance for political theorists.

Politeness and politics
Sayad introduces a fine-grained problematization of the moralization of migration and helps normative theorists look more closely at the practices rather than assessing the principles governing them. Here, moralization means the concealment of political and social interests behind the facade of neutrality and disinterestedness, and the disguising of relationships of power as relationships of benevolence. This aspect of his work is particularly useful for bridging the empirical analysis of daily interactions between the state and foreigners and the normative critique of the (un) fairness of the moral expectations implicitly formulated in these interactions. This has been heavily developed in studies on humanitarianism and the governmentality of refugees (Fassin, 2012;Rozakou, 2012;Vandevoordt, 2019) but remains marginal in the field of migration justice.
The normative literature on naturalization has, for instance, focused on citizenship tests, which allow the state to distribute citizenship according to individual competencies and efforts to embrace the political and national culture. Briefly, the literature discusses whether or not such tests are fair and effective depending on the justifications and criteria for testing (Baub€ ock and Joppke, 2010; de Vries, 2018; Mason, 2014;van Oers, 2020). However, as these policies were somewhat newly implemented in Europe, few have considered how such tests reveal -rather than creating from scratch -a symbolic order based on the natural and legitimate hierarchy between citizens and non-citizens. 5 By looking at the practical, discursive and symbolic aspects of naturalization, Sayad's work offers a valuable framework to analyse the informal expectations of liberal democratic states, which generally constitute moral demands of gratitude and conformity with linguistic, cultural and social norms. These expectations are moral rather than political, expressing a more systematic moralization of migration and integration, as I detail below when discussing Sayad's conception of politeness. But the objective and legalistic rhetoric of tests veils this moralization. The French case of the 'Republican Integration Contract' epitomizes this well (Boudou, 2018: 126-131;Boudou and Busekist, 2018;Brunstetter, 2012). Since such a contract cannot bind the state, and because foreigners who sign the contract cannot avoid subsequent arbitrary assessments of their 'proper' integration, this contract functions more like a symbolic oath. Specifically, the signatory party must pledge allegiance to a pre-established set of values rather than agree on mutual terms.
Consider two further practical cases. In 2016, the former French Minister of the Interior and head of the Foundation of French Islam (Fondation de l'Islam de France), Jean-Pierre Chev enement, advised 'discretion' in the public space for Muslims, because of the 'difficult context' after the 2015 terrorist attacks. The same year, a Muslim woman refused to shake hands with the representative of state during her naturalization ceremony. As an immediate consequence, later confirmed by the Conseil d'Etat, the state refused to naturalize her, arguing a 'failed assimilation' to the French community, 'in a symbolic place and at a symbolic time' (R epublique Franc¸aise, 2018). What is at stake here is what Sayad analyses as the interference of politeness in politics.
Sayad pursues Bourdieu's analysis, according to which 'the concessions of politeness always contain political concessions'. Politeness, like the harmless act of shaking hands, means: the public testimonies of recognition which every group expects of its members (especially at moments of co-option) [. . .]. Because, as in gift exchange, the exchange is an end in itself, the tribute demanded by the group generally comes down to a matter of trifles, that is, to symbolic rituals (rites of passage, the ceremonials of etiquette, etc.), formalities and formalisms which 'cost nothing' to perform and seem such 'natural' things to demand [. . .] that abstention amounts to a refusal or a challenge. (Bourdieu, 1977: 95) In the same way that 'discretion' is recommended, as an etiquette handbook would advise against talking politics at dinner, informal requirements become norms, and thus criteria for judgement. Rituals of naturalization encapsulate this political structure of politeness as a series of normative expectations towards the foreigner: politeness translates into civility, which transforms (or not) into citizenship and nationality. Civility, and its opposite, the 'incivility' of 'bad' citizens (Gayet-Viaud, 2017), is the good attitude par excellence of the (future) deserving member, along with decency, respect for tacit codes of conduct (i.e. acknowledgement of the value and potential superiority of local-national mores) and gratitude to the state which alone decides if a privilege is granted. Considering these expectations, naturalization can only remain incomplete, as the process of legal acquisition of nationality appears contrary to the ideal of 'natural' membership. Sayad accurately exposed this double bind: one must become a national, but the very fact of becoming a national (and not of being one) renders this naturalization artificial. Naturalization is an impossible but necessary ritual, which reveals the politics of its ritualization and the artificiality of nationality (Sayad, 2004: ch. 11; see also Badenhoop, 2017;Fassin and Mazouz, 2007;Monforte et al., 2019). 6 Sayad completes his analysis of naturalization and 'rigged rules of civility' (Anderson, 2010: 98) with a critique of integration in the French political context, where integration had become the epitome of republicanism (Favell, 2001;Laborde, 2008). Favell (2001: 3-4) aptly renders what integration meant in the 1980s in France: the 'problem of integration' concerns more than just the integration of new ethnic minorities. It is as much a general question about the 'glue' of a particular society-in each case here [France and Britain], a nation-across its wider cultural, regional and class divisions. [. . .] France emphasizing the universalist idea of integration, of transforming immigrants into full French citoyens.
The French conception of integration, hence, is different from the Anglo-American understanding of it as cultural blending or 'the mixing of segregated groups' (Young, 2000: 197). Both do, however, 'tend to leave the dominant group relatively undisturbed while requiring significant changes from members of the excluded groups' (Young, 2000: 216). Sayad (2004: ch. 10) addresses integration as a contested and polysemic notion that still structures the ideal of French republicanism regarding immigrants. 7 More importantly, he argues that integration cannot be wanted, declared or expected. Integration 'can be achieved only as a side effect of actions undertaken for different purposes' (Sayad, 2004: 223). For integration to have some reality, it must happen without direct desire or participation; similar to wanting to fall asleep or wanting to forget something, the more we actively try the more we fail. Sayad shows the perversity that moral philosophers have also well analysed (Fitzgerald, 1998;Walker, 1988): it is never possible to articulate precisely how much gratitude should be expressed and what amount of civility and politeness should be shown to testify of one's integration. They depend entirely on subjective appreciations, relative to moral feelings and attitudes that cannot be formally required, but only truthfully and spontaneously experienced. Indeed, no one can be forced to show gratitude without ruining gratitude itself. Such a gesture has proper value when it is sincere and voluntary. In other words, such expectations create an unpayable debt for foreigners.
To illustrate this substitution of politeness (broadly conceptualized) for politics, Sayad uses home metaphors, often ironically: It is rude, contrary to good manners and social morality to interfere in the internal affairs of the household in which one is received. (Sayad, 2006a: 166) [Political neutrality] prohibits those who are not local (i.e. non-nationals) from intervening in the political life of the owners [maıtre de c eans]. (Sayad, 2006b: 13) Wanting immigrants to be effectively 'at home' and not 'as if they were home', that is to say, falsely at home because they are in reality 'at someone else's place', 'at our place' [chez nous], requires that we give them the means and reasons to feel at home or, better still, that we accept that they conquer these same means and reasons, the right to vote being the condition and the conclusion of all this. (Sayad, 2006b: 31-32) Rather than 'right of residence' [. . .] it is significant that implicit and sometimes explicit reference is made to 'duties': the duty of assistance, the duty of humanity, the moral obligation, the sense of hospitality, the honour of France. (Sayad, 2014a: 172) This list of metaphors illustrates Sayad's profound criticism of 'welcome' as an ideology, against a large part of the political and sociological literature of the beginning of the 20th century. Migration has long been perceived mainly from the perspective of the immigration society and, more problematically, as a privilege granted by the sovereign.
We see again how nationalist logics reduce the rights of individuals to the goodwill of the 'host' society. Sayad underlines the hypocrisy and good feelings at work in the conception of a society as a family on the one hand, and the welcome into this family as an act of hospitality on the other. Discourses of generosity, disinterested love, benevolence and empathy prevail where, in fact, a distinctly political process is happening. This moralization of immigration hides the interests (material and symbolic), the competition and the conflict behind a veil of a naı¨ve morality of gift-giving, which triggers the expectations of gratitude, politeness and good manners. The moralization of immigration remains salient in the recent revival of the rhetoric of hospitality, particularly towards refugees (Boudou, 2021).

Democratic inclusion in a postcolonial context
Following from these views of migration, integration and naturalization, Sayad raises several normative questions, which is rather uncommon for a sociologist: Do we have to subordinate present immigration to the future (or potential) naturalization that will complete it, and do we actually have the means to implement that policy? Do we have to select in advance the immigrants we need, and what selection criteria are to be used to prevent any damage being done to the cultural homogeneity of the nation [. . .]? (Sayad, 2004: 226) Will this passion [for equality] go as far as full inclusion, as far as the integration into the national (i.e., naturalization) of everything that is not naturally national? (Sayad, 2014a: 52) Who belongs to the nation and who doesn't? Who can belong to it and who cannot? By what means and under what conditions can one belong to it and under what conditions does the nation accept this 'acquired' membership? (Sayad, 2014a: 63) Sayad develops an argument for democratic inclusion, drawn from both his empirical material and his historical and philosophical readings (Sayad, 2014a: 188-191). He does not conceptualize the possible duties of the state towards foreigners; again, that would merely maintain the epistemic and political superiority of the host society. Instead, Sayad argues for civil rights movements allowing migrants' voices to be expressed and heard, not as a claim to integration, but as a sign of integration. The first part of the argument is negative. It consists of showing how legal-political exclusion and social exclusion legitimize each other. Sayad returns several times, both empirically and normatively, to this problem of a territorial presence without political recognition which facilitates domination. The second part of the argument is a reconstruction of the democratic norm of inclusion to better consider this gap and define the criteria of democratic legitimacy regarding equality. The challenge is not only to allow immigrants to act or speak out (voice), but to make available 'the ability to re-appropriate for oneself the possibility of constructing one's own identity and of evaluating that identity in complete autonomy' (Sayad, 2004: 288).
The demand for democratic equality and the recognition of individuals as subjects of law presuppose the opportunity to take part in the decisions that affect us, whatever our status. Without direct control over the laws and decisions that affect immigrants (citizens or not), the 'most common form of tyranny in human history' is perpetuated: 'the rule of citizens over non-citizens, of members over strangers ', as Michael Walzer (1983: 62) famously wrote. Foreigners, immigrants and noncitizens to whom the legitimacy of presence is denied find themselves outside any legal, political and moral sphere. Only violence offers a possibility of existence and recognition, argues Sayad: To a youth forbidden to speak, forbidden to participate in the most elementary civic life [. . .] there remained, at first, only the most violent violence to respond to the legal violence that excluded them from the life of the polity, [. . .] paradoxically, only violence remained to be able to exist civically. (Sayad, 2006b: 20) This parallel between Walzer and Sayad appears fruitful for the question of democratic inclusion. Without going into too much detail, at first glance Walzer stands on the opposite side to Sayad. He reaffirms the sovereignty of the state and of the citizens to decide on and allocate membership. His position constitutes the conventional view, i.e. the normative standpoint positing the moral primacy of states to decide upon their borders. He does, however, specify some limits to the arbitrariness of this decision, the duties towards refugees and the necessary political inclusion of immigrants once they have settled in the territory. The argument against the tyranny of citizens over foreign residents is close to Sayad's views, but Sayad offers resources to clear away Walzer's communitarian undertones (Carens, 1987: 264-270). To use Walzer's terminology, the privilege of the host allocating membership represents a quasi-nativist monopoly and a predominance of nationality over other goods to be distributed. This a priori excludes noncitizens and anyone who does not partake in the shared meanings, that is, a public or national culture so vague and dependent on primary socialization that it has the objective function of excluding the non-native. Membership should not be decided by nationals alone, who otherwise profit from their monopoly (the concept is also in Sayad), and may exclude de facto or de jure anyone who does not deserve, in their eyes, to join the national club (Sayad, 1990: 12;2014a: 52).
For Sayad, the solution lies in the politicization of exclusion by migrants themselves. Inclusion should not be reduced to a legal status, given or not, but may result from a more general action. As the literature on performative citizenship has shown, inclusion is achieved or actualized through acts relative to the rights and claims of would-be citizens (Bloemraad, 2018;Isin, 2017). The distinction between national and foreigner, citizen and non-citizen, is not simply a discrimination. It is a distinction of status that should be contested and redefined through a series of performative acts-as-citizens. Similarly, Sayad argues that integration should be acted upon (acting-as). It cannot result from the mere appropriation of cultural idiosyncrasies or the conspicuous acceptance of national narratives. This would be polite integration and would maintain the ascription of the foreigner as a guest yet to be integrated. In short, social equality and recognition facilitates participation (feeling oneself legitimate in taking part in politics), and participation on an equal footing with citizens is the sign of integration. Such integration fosters democratic inclusion by voicing one's interests as a noncitizen.
The right to vote is obviously an essential tool for democratic inclusion (Beckman, 2006;Lopez-Guerra, 2005;Pedroza, 2019), but it might have drawn too much attention from political theorists to the detriment of the social conditions of inclusion. Inclusion is overly limited by its legalistic understanding -an abstract right to participate, perpetuating the status quo. The sociological literature, in line with Sayad, has clearly illustrated how non-citizens tend to participate less in the political life of the host country: there is a discrepancy between democratic inclusion -which may remain formal -and integration, in the sense of Sayad. Different factors, such as low socio-economic status, lack of familiarity with the political system, unequal access to informal channels of information, approximate knowledge of the official language or fear towards state bureaucracy, impede the ability to participate (Hammar, 1994;Rudiger and Spencer, 2003). A sociologically informed framework that is sensitive to social barriers to participation must constrain the normative argument for enfranchisement. Access to political action completes a formal right to vote. Through conventional and non-conventional forms of political participation, people publicly refuse to accept a subordinate position that prevents any criticism or questioning of the national order (Sayad, 2014a: 58-60). Making oneself heard is a normal consequence of full democratic inclusion, breaking with the codes of political politeness that foreigners/guests are expected to follow.
When national enfranchisement is not immediately possible, theorists have argued for local forms of citizenship and voting rights, while the municipal right to vote has gained legitimacy for EU citizens all over Europe (Arrighi and Baub€ ock, 2017;de Shalit, 2018). However, Sayad warns normative theorists by calling into question this re-enactment of a double standard for nationals and citizens: What is the value of the distinction to which this selective, partial exercise of the right to vote refers, and what is the value of the distinction that one makes, basically, between local interests and national interests, between minor interests and major interests, between subordinate interests and supreme interests? (Sayad, 2006b: 29) For Sayad, who has the history of France and Algeria in mind, this hierarchy of voting rights parallels the colonial logic of nationalism: some people still remain subjects without being citizens, and have local but not national political autonomy. 8 He repeatedly emphasizes this connection between the colonized of yesterday and the immigrant of today. Because he focuses on the relationship between France and Algeria, he can write without further precautions that 'everything suggests that today's immigrant is the counterpart of yesterday's colonized. He is only colonized in a new way, colonized from beyond colonization' (Sayad, 2006b: 29). It is difficult to extrapolate from Sayad's 'extreme degree of selfabsorption' (Massey, 2006: 115) in the Algerian migration case study. However, Oboler (2006) draws direct parallels with the relations between Puerto Rico and the United States, and some studies in political theory have considered past colonial relationships to justify a special right to immigration for formerly colonized people (Amighetti and Nuti, 2016). Migration studies have more prominently mobilized Sayad's perspective when addressing North African migration (Cherti and Collyer, 2015;Gross-Wyrtzen and Gazzotti, 2021). It brings to the fore the relationship between the states and peoples of both the places of emigration and immigration, instead of only focusing on the immigration state faced by 'newcomers'. Sayad insists on the continuity of the phenomenon of migration, clarifying that the act of crossing a border is necessarily entangled in emigration and immigration projects that cannot be distinguished independently and are generally structured by (colonial) history. Ignoring this continuity to privilege the perspective of the host state is epistemologically problematic and, politically, it reflects the power structure between states, especially when they are the former colonizer and the former colonized state.
In law, E Tendayi Achiume (2019) has recently made a more radical proposal: postcolonial relationships still produce neo-colonial forms of power that shape a transnational political community in which former colonizing and colonized peoples are both equally entitled to have a say regarding this power. Since the members of this community retain a right to self-determination, Achiume argues that 'for some Third World persons, at least one available means of pursuing political equality and asserting sovereignty (the capacity to self-determine)-together, decolonization-may very well be migration' (Achiume, 2019(Achiume, : 1552. This goes beyond Sayad's point but follows his ambition to think of emigrationimmigration as the continuation of colonial relationships by other means. 9 This approach is of crucial importance when thinking about migration justice in two complementary ways, and it follows Sayad's methodological focus on both structures that make migration possible as well as individuals' strategies within these structures. Firstly, an understanding of decolonial migration aids in assessing the power of states over migrants in a more structural manner. Rather than an abstract consideration of the duties of states regarding migrants, even regarding migrants from their former colonies, such an understanding shows how the power of First World states is pervasive, especially in international organizations, affecting all migrants from a formerly colonized Third World (Achiume, 2019(Achiume, : 1561. Secondly, Achiume includes decolonial migration as a part of 'a more general theory' of 'de-imperial migration', i.e. 'any form of migration that is responsible to informal imperial subordination and that offers a means of countering that subordination through individual (rather than structural) means of enhancing political equality' (Achiume, 2019(Achiume, : 1565(Achiume, -1566. So, in addition to the structural move, decolonial migration here emphasizes the individual's migration paths as part of the scope of political justice. Migrants are actors of democratic flourishing rather than mere beneficiaries. Following these developments, Sayad's perspective could nourish a growing field of research on colonial remnants in migration ethics (Finlayson, 2020;Jaggar, 2020) and how to circumvent them, both methodologically and normatively.

Conclusion
I have argued that political theorists should expand their research on the ethics of migration to consider empirical studies and the normative work conducted by social scientists. Instead of making an abstract methodological case, I have shown why and how reading the sociological work of Abdelmalek Sayad offers new insight into the articulation between the social sciences and the political theory of migration. I have also maintained that Sayad's work illustrates how to enrich normative reasoning based on empirical research. Reading his work as normative theory brings original arguments to the fore regarding statist and nationalist biases that moralize migration, inclusion and citizenship. Driven by a sense of epistemic justice towards the 'mutes' who were Algerian migrants in France, Sayad deploys a critical sociology that is empirically rich and normatively sensitive to the power asymmetry that still applies today.
I followed Sayad to analyse his normative project in different areas, all centred on a critique of state thought and nationalism as a nativist set of logics and practices. Criticizing the imperatives of naturalization, politeness and integration, Sayad contributes to a better understanding of the moral expectations towards newcomers and suggests new ways to develop a grounded and de-moralized take on migration. Sayad also helps develop an argument for democratic inclusion: social equality and recognition facilitates participation on an equal footing with citizens, which should be the only sign of integration. Such integration, coupled with the right to vote for foreign residents, fosters democratic inclusion by enabling non-citizens to voice their interests. Finally, Sayad's attention to the postcolonial dynamics in migration helps decolonize political theory, adding methodological, empirical and normative findings into normative ethics and justifying stronger claims to migration in postcolonial contexts. immediate agreement between the incorporated structures, turned into practical schemes, such as those which organize temporal rhythms (for example the quite arbitrary division into hours of the school timetable), and the objective structures'. 6. Civility can also be expected from non-migrant minorities. It has the same effect of constraining dissensus (i.e. political expression that formally or substantially questions the majority's position) by patronizing rules of politeness. See in particular Anderson (2010: 98) on civility and politeness as an enabler of segregation in the United States. 7. I will distinguish this ethno-cultural integration into the nation from democratic inclusion, which is the legal-political inclusion into the demos. On the different justifications of the norm of democratic inclusion, see Baub€ ock (2017). 8. Political theorists have since developed a convincing argument regarding the possibility of having differentiated (rather than hierarchical) rights to better account for the different situations of individuals being present on given territory (sojourners, residents, members), and to take into consideration the willingness of some temporary foreign workers to have special rights for economic and legal protection while not having a path to citizenship. See in particular Song (2016) and Straehle (2019). 9. See also Sayad (1990: 17-19). On Sayad's postcolonial reception, see Boubeker (2010) and Go (2013).