2.1 Introduction

This chapter highlights the importance of the distinction between empirical facts versus institutional facts. Physical presence on the territory is an example of the first one, while residence is an example of the second. The chapter advocates for greater awareness of this distinction (Section 2.1) and argues that, if the distinction is disregarded or blurred, serious consequences follow. Some consequences are spelled out: first, our discourses lack in clarity, which inclines us to make controversial claims. Second, when we disregard the distinction, we might not properly identify our main object of research. For example, our main object of concern may not be individuals moving across borders. It might rather be legal structures that, in fact, immobilise people. The third consequence (Section 2.5) is that the confusion between empirical and institutional facts affects how we speak about population. We should distinguish the set of living bodies on a given territory (empirical fact) from the institutional fact of population. The chapter demonstrates that, once aware of the distinction, we are better equipped to see how the legal regulations governing the institutional fact of population affects our social and constitutional identity. In this sense, how law frames population has consequences for the empirical dimension of the individuals present on the territory. The reason is that the legal regulations governing who counts as a member of the population determines who we will live with; whether, for example, we might be allowed family unification and the identity of the children that we will give birth to. In light of this, the chapter advances the claim that both whom we admit and whom we do not admit will have an impact on who we will be tomorrow in terms of our social and constitutional identity. Migration policy is, therefore, an important constitutional matter. The fourth consequence (Section 2.6) of the blurring of the distinction between empirical facts and institutional facts is the danger of overlooking the distinction between People-as-a-part (empirical fact) and People-as-a-whole (institutional fact). Populism exploits the blurring of the latter distinction: the populist framing of anti-migration policies pitching ‘them’ versus ‘us’ is a case in point. We ought to beware of the unjustified appropriation of the People-as-a-whole by the People-as-a-part grounded in the disputable assumption that one group enjoys direct epistemic access to the common good. Finally, the chapter ends by pointing out that the appreciation of these five different consequences allows us to understand in a clearer light that migration policy has constitutional impact: it impacts on the composition of the institutional fact of the population, on the empirical fact of who is likely to live within a given territory, but, ultimately, also on the composition of the institutional fact of the People (those who count as citizens) and its role in the constitutional order (what the citizens do, which rights and duties they have) determines the state’s constitutional identity. Therefore, as this chapter argues, migration law can be employed to both strengthen and dismantle constitutional democratic institutions. Migration lawyers thus have much to offer to the constitutional lawyer concerned about the contemporary resilience of the constitutional setting.

More specifically, to this volume’s central question on the relationship between restrictive migration policy, populism and democratic decay, this chapter argues that populism typically employs a debatable view of peoplehood that lays claims on (re)defining the institutional fact of the citizenry in the manner that the group prefers. In the contemporary context, restrictive migration policy emerges as a tool for separating the wheat from the chaff, deepening the divide between the set of individuals living in a given territory and the members of the polity in ways that challenge the mode of self-governance that inspires the democratic form of government. As to this volume’s other key question – what can legal resilience do to avoid the erosion of migrants’ rights? – this chapter advances the seemingly innocuous answer that, if we wish to defend these rights, an important step would be to think (and speak) clearly about certain basic conceptual matters. We need to see how law constitutes the statuses of things that, in ordinary language, are conceived as natural entities, the evident characteristics of which it would simply be vain to resist. On the contrary, entities such as national belonging, citizenship, residence, habitual dwelling, migration and population, found in the law, do not refer to natural entities or empirical facts. Instead, they constitute, as shown below, institutional facts determined by particular constitutive rules and these rules are set up in the law and could have been different.Footnote ¹ By ignoring this conceptual point, we obfuscate the crucial fact that we have designed law so that it, say, bans visa-free travel and extraterritorial asylum applications. It is not movement itself, the mere empirical fact of a human body moving in space, that keeps people away from rights. It is the law that does so and law is a question of convention, not of empirical necessity. The motivation to change the law only comes from seeing that it is our creation.

2.2 On Institutional Facts

Legal thinking concerns what we today usually call institutional facts,Footnote ² where a function is attributed to something that does not have this function in virtue of its empirical properties. Imagine that there is a rule according to which, in certain contexts, something counts as a point in, say, a board game. It is not the nature of the thing itself that makes it a point in this particular board game. Indeed, often there is not even a ‘thing’ to refer to when we employ points in board games. What makes the point such is rather the constitutive rule. This outline of the way a point in a board game works is reminiscent of how many rules work in the law. This insight into the ontology of institutional facts plays a role in our understanding of legal concepts in general and more specifically in our understanding of central concepts in migration law.

Indeed, I understand this edited volume to focus on this internal/external membership dialectics according to which the ‘inside’ and the ‘outside’ dimensions of belonging are ‘intertwined’. Linda Bosniak famously made a point about how border protection runs through the ‘internal’ dimension, away from the territorial confines, to the heart of law and precisely to how members treat physically present non-members.Footnote ³ Vladislava Stoyanova remarks how ‘concerns with protecting the boundaries of membership then can structure and affect the position of everybody’Footnote ⁴ in a call for more of a ‘sustained dialogue between the fields of migration law and constitutional law’Footnote ⁵ – a call that I have joined on many occasions and that recently seems to have gained some traction.Footnote ⁶ Only recently has the constitutional dimension of citizenship begun to be systematically studied.Footnote ⁷ Stoyanova is perfectly right in pointing out, in line with Dora Kostakopoulou’s arguments on political identification, that the ‘internal’ and the ‘external’ have a dialectical relationship: one affects the other, and vice versa.

It is ancient wisdom that how you treat others reflects on who you are, and how you conceive of who you are reflects on how you treat others. Both who you admit and who you do not admit will thus reflect on who you are. We have yet to fully explore the implications of this insight in relation to contemporary democratic decay and twilight constitutionalism. If we are not ready to give in to a lesser kind of rule of law, a lesser kind of democracy, a lesser kind of citizenship, we should look carefully at the lawmaking process when it comes to the design of migration statuses and how these affect other legal positions within the legal order. One reason for this is that one’s migration status is not only about residence rights or the right to enter a country. Rather, it is connected to a whole bundle of legal entitlements and obligations – in the political realm, in family life, in the workplace and economic and social life generally. This is why so much of one’s life is dictated by one’s migration status. This is also why it is such a politically delicate matter. Another reason is that no matter what a state chooses to do in matters of sojourn and naturalisation, the choice will inevitably tell us something about that polity. Elsewhere, I have called this the constitutional-sensitivity thesis.Footnote ⁸

2.3 Like a Point in a Board Game

The Scandinavian legal realists, the work of whom I have long been interested in, understood early on that legal thinking concerns facts of a different kind than ordinary empirical facts. They understood that legal thinking operates with non-empirical facts, facts of another type that can, however, be empirically investigated. Through institutional facts we attribute to something a function that it does not have by virtue of its empirical properties. A common way through which such facts arise is through constitutive rules which are very common in law. From this perspective, the law appears to be essentially status-generative: it creates, or constitutes, different forms of statuses.

Now a status is a bit like a point in a board game: something you can win or lose; something that may be worth having, even though the thing itself (the ‘point’ or ‘points’) does not correspond to any empirical object and cannot be exchanged, substituted or replaced by an empirical object. Status is a concept of which we may offer an empirical explanation, even if it is not itself an empirical concept. Philosophers say that we exercise ‘our deontic power’ by creating different types of statuses.Footnote ⁹

The law can thus be likened to a technology that we employ to achieve a certain order in the world; for arranging or ordering the world, as opposed to merely describing it. To a certain extent, the law makes the world as it is, one might say, and the main way in which the law orders the world is by constituting and regulating various types of statuses.

Notice that a status is epistemologically speaking a kind of vox media. In itself a status is neither true, nor false. As an artefact of the human mind, it is neither good, nor bad. However, it may become good or bad depending on how well (or how poorly) it performs its function within the ecology of other statuses and legal positions it stands in relation to.

If one cannot demonstrate that the vox media allows a functional correlation between premise and conclusion in legal reasoning, the legal status may become an arbitrary tool for separating those who fall within its remit from those who fall without.Footnote ¹⁰ Ill-conceived legal statuses that result in arbitrary line-drawing suffer from lack of justification that undermines their legitimacy. Not having acceptable and satisfactory criteria for distinguishing between, say, citizens and non-citizens, makes it difficult to argue against the proposal of distributing citizenship by, for example, flipping coins or drawing the short straw. Some scholars who have argued that citizenship itself would be arbitraryFootnote ¹¹ or a tyrannical form of dominationFootnote ¹² have understood this point. Citizenship is not necessarily arbitrary in itself, but necessarily becomes an arbitrary tool of distinguishing insiders from outsiders if we cannot demonstrate that there are good reasons for the selection mechanism to be regulated in a specific way, given the nature of the polity we face. So, as Stoyanova stresses in quoting Ben Habib on the ‘fundamental right to justification’, ‘providing justifications (…) creates a space where different interests can be identified and arguments underpinning possible decisions exchanged. It also implies identification and evaluation of the empirical considerations behind the interests and the possible solutions.’Footnote ¹³ This is true in most contexts where we deliberate on collective action.

The aforementioned insight into the ontological nature of statuses – and what it implies for how statuses can be evaluated normatively – plays a role in our understanding of notions central to our discussion today. Here is how.

Once we realise that statuses are not natural kinds, we are better positioned to appreciate how law constitutes the statuses of things that, in ordinary language, are cast as entities of a natural kind, including notions such as national belonging, citizenship, residence, habitual dwelling, migration and population. It is important for us to distinguish the undeniably empirical dimension of, say, presence on territory (an empirical concept) from its normative cousins ‘entrant’, ‘residence’, ‘stay’, ‘sojourn’, ‘abode’, ‘domicile’, etc. None of the latter, found in the law, refer to empirical facts, but rather all are institutional facts determined by particular constitutive rules that are set up in the law and that could have been different as the very establishment of a constitutive rule is itself a question of convention, rather than of empirical necessity. Not to grasp the difference of the two dimensions would be to confuse fact and norm; to confuse the language of the law with its object of regulation – a mortal sin for a jurist.

If my reading of statuses as institutional facts is correct, it follows that we make certain assumptions about our object of inquiry that may be unwarranted. Contrary to geographers, biologists and doctors, we do not study migration or population as a natural kind – although human fluxes and demographics might be relevant to our understanding of the institutional facts by which law is essentially constituted (and relevant for our understanding of changes to these facts).Footnote ¹⁴ Perhaps, to start with, migration law might not be about bodies moving in space, but about the design and operations with statuses within the abstract space of legal computation. To see this more clearly, I suggest introducing a distinction between mobility and migration.

2.4 Mobility and Migration

Let us start by distinguishing migration from mobility. The main reason why this distinction should be made is that the law is the technology we use to define who counts as a non-citizen and a citizen respectively; under what conditions this is so, how long this is the case and what it takes for this to change. Nothing of this type can be predicated of mobility in general.

By mobility I mean a movement of a body in space. Such a movement may take place across a border between two states, or across an interstate border. It can also take place where there are no borders or within a borderland. Mobility is an empirical phenomenon that occurs in the world; it is best investigated using empirical methods. Mobility also raises interesting normative issues, but they do not necessarily coincide with the normative issues raised by migration.

By migration I mean a change in a person’s legal position vis-à-vis one or more state(s) or supranational organisations. Migration thus concerns the legal positions (rights and obligations) that regulate the individual’s right to stay in a country, to be protected from deportation, to be reunited with family members to name only some of the many legal positions that determine a person’s migration status. Other such legal positions concern the individual’s working life and socio-economic social life in general. According to this definition, status civitatis is a ‘migration status’. In fact, citizenship is a status that positions a person vis-à-vis one or several political communities and may contain many different rights and obligations. In addition, this status usually also regulates such things as residence, protection against deportation and the like.Footnote ¹⁵

The distinction that I am suggesting is radically different from the way migration studies usually think of migration.Footnote ¹⁶ I want to take a step away from the common understanding. The reason is due to the ontological tenet that a legal status is not an empirical fact. If this is so and the law constitutes a person’s status as an insider or outsider, we should reconsider what we think we know about migration in contemporary social science. We might see things in a new light.

Allow me a metaphor, one that the Scandinavian legal realists sometimes used themselves: the metaphor of money.Footnote ¹⁷ The relationship between mobility and migration is similar to that between paper and money. There is paper that counts as money, but not all paper is money, and all money certainly is not paper. Similarly, we find mobility that counts as migration, but not all types of mobility count as such. The law perceives a difference between a person stepping out of a house and the same person jumping the fence at the Spanish Melilla enclave. Both are empirically identified as movements of bodies in space. Consider also that migration is not necessarily grounded in mobility. Sometimes, physical movement of people is legally relevant in the sense that it provides reasons for a change in one’s migration status, but a migrant is not someone who necessarily moves or crosses a border. One can remain perfectly static and yet become a migrant. It is important to understand that mobility, or the physical movement, may trigger a change in migration status, but mobility is not required for such a change to happen. This means that whatever migration is, it is not grounded in physical movement because physical movement is insufficient to explain migration.

From this viewpoint, migration is not identical to movement across borders. Sometimes a person’s physical location is irrelevant for determining his or her migration status. Migration thus does not have to take place in the physical space, but it can occur as a result of a change in personal status (marriage, divorce, childbirth, adoption, etc.). Migration status can also change due to changes in the law. Consider the case in which a new ground for loss of a status is introduced. In the case of denaturalisation, for instance, we have seen a number of EU countries that have adopted a new ground for loss of citizenship often for naturalised individuals believed to be involved in terrorism; these individuals may be citizens one day and non-nationals the next, without moving across borders. Status can also change due to changes in the territorial jurisdiction (e.g. Brexit). Status can be modified as a result of the passage of time (e.g. at 18 years, after 5 years, after 24 November 2015) or territorial tampering (e.g. Tampa).

It is also worth mentioning here that the very word migration etymologically does not mean movement in space, but rather changed status. Migration in the sense of travel or moving across borders is not migrare in Latin, but peregrinari. Now it could be objected that the etymological root which gives the word migratio in Latin also gives ameibein in Greek which means to change or exchange. But it would be wrong to assume that migration, therefore, refers to mobility or change in general. Instead, it has to do with a change in legal status such as the right of some individuals (Latini) to become Roman citizens through the institute of ius migrandi. Ius migrandi was never a right to free movement, but the right to naturalisation as a Roman citizen.Footnote ¹⁸ It was not until modern times that the Latin verb migrare began to be used to describe a movement in space.

The distinction between migration and mobility has a number of consequences. Let me mention five.

2.5 Consequences

A first consequence of the distinction made above is that our discourse lacks in clarity. As social scientists we often fall prey to confusing the natural kind with the institutional fact: we speak as if residence, sojourn, entrance, family membership and the like would be natural kinds; a confusion that arises because we wrongly assume to have direct epistemic access to the entity at hand. We presume that our prelegal conception of, say, entrance determines our legal uses of the term that only in appearance are the same. We blur the empirical and the institutional features in unfortunate ways. This lack of clarity inclines us to make claims that are questionable. Many have claimed that international law would have failed in governing migration and many have pointed to the fact that the state would be incapable in a globalised world to govern migration flows. Regardless of the empirical foothold of such claims, it is rather obvious, that, even in situations where sustained cross-border movement is observed, this says little about any failure in the exercise of the ability to govern migration as here defined. Law might not efficiently shape movements, but it is quite effective in ascribing or denying rights to individuals. The very conception of this volume aspires to show that the law is able to curtail rights (or for that matter, to engender them). Also, when it is affirmed that ‘most asylum seekers are not able to depart and reach European States. The possibilities for movement and flight have been increasingly suppressed’Footnote ¹⁹ we obfuscate the crucial fact that it is the law – our law – that we have designed in such a way that bans visa-free travel and prohibits asylum applications to be filed with the embassy. By keeping the distinction between movement and migration in mind it becomes evident that it is not the cost of air tickets or the lack of road access to the airport that makes prospective asylum-seekers ‘unable’ to file their claims.

A second consequence of suppressing the distinction between migration and mobility is that we might misconceive the very object of inquiry. Our main object of research may not be individuals moving across borders. Our main object of research perhaps is better understood as the different set of legal positions, attributed to certain groups of individuals – where the physical movement in some cases may trigger a change in status, but this is just one among many triggers – relative to what a state may legitimately do to these individuals. The matters that fall within this realm are not always linked to movement either: it may concern legal positions relative to an individual’s work life, to housing, welfare, education, family life and more. The delimitation of migration law ratione materiae seems hard to settle a priori; it is a bit of a moving target. It may even be the case that migration law deals much with institutional legal structures that immobilise people. This point is becoming increasingly clear to those migration scholars who have noticed that contemporary border control policies often aim to immobilise people rather than manage movement. Are we convinced that migration law regulates the position of bodies in space rather than, say, how far in time individuals stand from their inclusion in certain institutional structures? What if it is not movement itself, this mere empirical fact, that keeps people away from institutional structures? What if it is the law that keeps individuals away from accessing rights? A poet like Berthold Brecht saw this point clearly: in Reisen im Exil he ironically wonders whether the most noble part of Man, after all, is not the passport.

In the contemporary migration debate, we are afflicted by a form of pensée unique according to which migration would be a movement across borders. We constantly talk about boundaries, borderlands, liminal spaces, frontiers, frontlines, streams, channels, flows, fluxes, etc. This obsession about space obfuscates a quite evident point for the lawyer: legal orders ultimately determine who is to be considered a migrant, how, when and where this is the case; a legal order is itself a legal construct, not a location in space. If law is an instrument, a technology we employ to determine statuses, state migration agencies identify migrants of a certain kind once they set out to look for them. If a visa is hard to get, entry into a country is more likely to be irregular. So, a country’s ‘stock’ of irregular migrants depends on its visa policy. Make visas harder to get and irregular migration rises; when a state makes legal migration harder, there will be more irregular presence, other things being equal. That is why some think illegal migration is a phantom illness.Footnote ²⁰ It is a product of the laws that combat it. Combatting illegality would be somewhat like fighting your own fist basically.

If this is right, migration crisis can be policy-made or fabricated. Mobility does not per se create a ‘migration crisis’, the law does. This is so for a deeper reason than the reasons usually given when we discuss fabricated emergencies, like those mentioned by Spijkerboer for instance,Footnote ²¹ that is that the ‘emergencies’ may be caused by suboptimal internal policy choices and not by external forces or events. The deeper reason underpinning a ‘migration crisis’ is rather the law itself that keeps people away from the reach of institutional structures: states engage in buck-passing so as to avoid responsibility for certain individuals. To see this, we have to understand that mobility is not migration; mobility is an empirical fact; migration is not; migration is an institutional fact construed by the law. In a sense, the Scandinavian realists would have said that it is a little like magic.

2.6 On Population Design and the Magic Wand of the Law

The design and management of migration statuses play an important part in shaping one of the state’s ‘constitutive elements’, namely its population.Footnote ²² Thus this section argues that migration and citizenship laws enable population design through the design of institutional facts: by the magic of pen-strokes we modify the number and social characteristics of the individuals we count as migrants or citizens.

The aforementioned infelicitous confusion between empirical and institutional facts affects how we speak about ‘population’. We are often quite unclear, when speaking about population whether we are referring to the population as a natural kind, the mere demographic instance, the set of human living bodies, or to the population as a legal construct, an object of policy design and outcome of census. Notice that population in the first sense (here: population⁽¹⁾) and in the second (here: population⁽²⁾) may differ significantly. Population⁽¹⁾ may both outnumber and, vice versa, be less numerous than population⁽²⁾. The latter may include expats within its scope. At the same time, it may exclude certain individuals on administrative grounds (e.g. unregistered births) by making them invisible in the eyes of the law.Footnote ²³

The forms that migration statuses take, the technicalities regulating these, and how they relate to other legal positions (rights and obligations) within the legal order forge the institutional design of the citizenry and hence bear heavily on the question of constitutional identity. Migration and citizenship policy, directly or indirectly, determine the extension and social composition of the population⁽²⁾ in a political community; they attribute to the individual significant bundles of rights and duties that co-determine that individual’s position within society and within the power-setting of the constitutional framework. In this sense, these areas of law contribute to the construction of the position that a given individual occupies vis-à-vis other individuals and their collective endeavours (i.e., the State). Since migration law typically governs access to residence rights and these are prerequisites for naturalisation in constitutional frameworks – where political rights, in the form of voting rights, are most often reserved for fully capacitated nationals – migration law in combination with citizenship policy, electoral laws and constitutional regulations of the role of the citizenry determine who counts as a member of the People. How we regulate access to, and the content of citizenship determines how the demos, or set of citizens, is composed. In other words, the legal positions reserved exclusively for citizens determine the share of political power that citizens enjoy in the constitutional order: how, when and where they may use political power. The composition of the People (who counts as a citizen) and the citizen’s role in the constitutional order (which rights/duties are reserved for citizens) determines the state’s constitutional identity.

Crucially, precisely because legal statuses are institutional facts and not natural kinds, we can change our population⁽²⁾ – the number and social identity of the individuals we treat as migrants or citizens, within and beyond our territory by the magic of pen-strokes. For example, EU citizens may become third country nationals by this kind of institutional expedient. In this way, the EU just lost about sixty-six million EU citizens on ‘Brexit day’ in January 2020. Of course, the magic bullet will not make the actual bodies of people disappear, nor will it, in itself, make individuals appear (population⁽¹⁾). It is not real magic after all. But the legal status can both appear, disappear, and vary at the pace of any given regulatory change. As the legal status is transformed, with each new law, regulation or doctrinal revirement, the population⁽²⁾ may change and what we may ‘legally’ do to certain individuals can thus change significantly. What we may do to individuals also impacts on who they will live with (e.g. family members) and the children they will give birth to. It follows that not only do legal frameworks impact the design of the population⁽²⁾,t they also have important consequences for the volume and the social composition of population⁽¹⁾.

If by population design we mean the design of population⁽²⁾ that impacts on population⁽¹⁾, typical ways of engaging in population design have been used to fabricate citizens or to export the poor or otherwise unwanted. Throughout history, the volume and composition of the population was often thought of as a source of strength (or the lack thereof as a weakness). During the course of history, rules concerning, say, the acquisition of citizenship have been modified to achieve objectives such as drafting men to armies, avoiding responsibility for certain socio-economic groups or getting rid of more or less challenging minorities. Think of it as eugenics by passport distribution. Popular political tricks in population design have included increasing the state’s population ‘fictively’ to gain power, an eternal temptation for electoral district designers, competing in the ‘battle for brains’ by opening the ranks of the citizenry, cherry-picking a preferred élite by fast-tracking on the migration route, intervening in other countries through one’s diaspora, imposing protection on one’s citizens abroad, like Russia did in Georgia for instance, to meddle with foreign policy, incentivising migration to solve unemployment, like Italy or the Philippines have done, or simply exporting poor people, a temptation that Gulf states have been looking into to avoid taking responsibility for guest-workers in the region.

It is important to see that to the extent that the number of individuals influences the distribution of the political power among different social groups, the dimension and composition of the population⁽¹⁾ will affect the political distribution of power in a society. Rules concerning citizenship can thus be used for the purpose of shifting power from one group to another. As early as in the fifth century BC, Gorgias from Leontinoi, the famous pre-Socratic sophist, joked about states that bow to the pleasure of fabricating citizens out of thin air. Some 2500 years later we keep on using the same trick. Citizenship policy may thus be seen as an indirect a way of gerrymandering and migration law as a tool that allows states to engage in population design, therefore, using population⁽²⁾ to modify population⁽¹⁾, often in view of obtaining political advantages.

A contemporary example of such a shift in power is the persecution of illegal migrants that takes place in India’s Assam region.Footnote ²⁴ The State claims to suffer a migration crisis due to the uncontrolled entrance and stay of Bangladeshi unauthorised migrants. Instead of, for instance, finding and prosecuting the individuals that the State believes to be ‘illegal migrants’ that it wants to deport to Bangladesh, Assam has chosen to demand that its residents, many of whom are illiterate, prove that they are indeed citizens. This implies a shift of the burden of proof from the State to the individual. Those who fail to prove their citizenship are placed in custody in view of deportation. Technically, the case of Assam is a case of quasi-loss of citizenship: the validity of the status is declared void ex tunc. There is little evidence to suggest that most of those whose citizenship is declared void would be anything else than Indians. But there is quite significant evidence to suggest that those whose citizenship is annulled are overwhelmingly Muslims. This fact alone exacerbates the sectarian tension between Hindus and Muslims in the region. In Assam, the dividing lines of politics largely mimic those between religious identities: the party system reflects social identities grounded in religion. So, if many Muslims are denationalised, the electorate’s composition changes to the effect that the political party system is impacted. It is like disenfranchising one’s political opponent thereby ridiculing the rationale of the democratic form of government. Basically, the effect obtained through a shift in the burden of proof – a technicality in itself – combined with a particularly sensitive legal status (status civitatis) is to silence particular political movements. All this is nothing short of manipulation of the constitutional identity.

The Assam case shows very well how a procedural technicality in the area of migration law can impact the political system. The effect of this manipulation distorts the constitutional settlement and by doing so denigrates the value of the citizenship also for the citizens who were not deprived of their citizenship or subjected to deportation. This is possible because citizenship is a status that offers ‘bundle-rights’,Footnote ²⁵ a whole set of rights concerning residence, family life, work life, and, last but not least, political participation.

The Assam case points to the fact that how you design migration statuses reflects who you are. It is important to notice that this is not only true where the policy design choices are of disputable kind. Rather it is true no matter what policy choice is made. To be clear, we should qualify the assertion according to which citizenship policy, in combination with migration policy, ‘allow states to choose populations’.Footnote ²⁶ At the micro or individual level, it is not so: one could object that citizenship is generally a birth-right and states do not ‘choose’ their future nationals, at least, in so far as no particular birth-control regimes are practised to this end. If the claim is read in connection with migration law, it looks more convincing: states may cherry-pick certain individuals and design categories to the effect of not allowing others to become part of the legally recognised population. However, what is important to notice is that, although it may be debateable whether states ‘choose’ their populations in reference to the individual level, this claim holds at the macro-level: states certainly engage in population design and crafting of their citizenry at the macro-level. Many have shown the causal connection between citizenship regimes, allocation of public goods and global inequality.Footnote ²⁷

Also, it is important to see that both whom you admit and whom you do not admit will reflect who you will be tomorrow. The constitutional identity of the State is determined inter alios by the choices in these policy fields. This makes migration policy a lofty constitutional matter, albeit sometimes in ways that are not immediately visible. Precisely because migration policy impacts on the constitutional bedrock of the state, there is a particular reason why the populist conception of the People ought to be of interest to both migration and constitutional lawyers. It is important to understand the causal triangulation between democratic decay, authoritarian populism and restriction of rights for migrants. One expression of this triangulation is the resurgence of the synecdoche that characterises populism.

2.7 The Synecdoche of the Populist People

The conception of People in populism is a synecdoche, that is, a particular figure of speech belonging to the category of metonymy in which one refers a part for a whole or vice versa. The particular form of this synecdoche is the microcosm. A microcosm uses a part of something to refer to the entirety.

Neil Walker has rightly pointed out that the political notion of ‘the People’ may always be a synecdoche of sorts. It typically does not refer to all persons; often it does not include reference to disenfranchised categories. All political movements operate with some notion of who are to be sociologically representative of ‘the People’; and ‘the People’ is clearly not synonymous with population⁽¹⁾, a mere empirical concept. However, the way populism employs the synecdoche differs from non-populist uses. One hypothesis is that it does so quantitatively in indicating a stricter use of the figure of speech: fewer members of the microcosm are taken to represent the macrocosm. A more convincing hypothesis is that perhaps the populist use of the synecdoche is characterised by a certain unawareness: the figure of speech would be used without awareness of the fact that it is being used. Consider, a contrario, how the synecdoche is employed more ordinarily in constitutional theory: here, the synecdoche lacks the pretence of describing the world, which is evident from the distinction made between populus and multitudo, to the effect that the theory is self-aware of the employment of the figure of speech to a much higher degree.

To illustrate the point, let us start with the obvious: ‘people’ is an abstract term, with no clear, single referent. It is does not refer to a natural kind. The collective subject, that is, the ‘people’, is distinct from the individuals composing it (i.e., population⁽¹⁾) and it exists only in our discourse.Footnote ²⁸ As Kelsen famously pointed out, ‘People’ is always a fictio that obscures the fact that the individuals in a society have multiple and diverging interests and attitudes. The will of the people, as Albert Weale has recently shown, or perhaps reminded us about, is a myth – and a particularly uninstructive myth today given the pluralism and complexity of contemporary societies.Footnote ²⁹ To make the same point in the words of a scholar who has reflected much on populism, let me quote Ernesto Laclau: ‘People is therefore always ‘a unity – a homogeneity – out of an irreducible heterogeneity.’Footnote ³⁰

What is then the ‘homogeneity’ that populists impute to ‘the people’? If we assume that Jan-Werner Müller has correctly diagnosed contemporary populism as ‘an exclusionary form of identity politics’Footnote ³¹, we would still need to know what creates this homogeneity besides the ‘national contents’ that will vary from country to country. How can we recognise the construction of homogeneity among the numerous national differences in the construction of collective identities?

Here philosophy can help. As Valentina Pazé has recently pointed out, we know well from the history of political thought what ‘people’ means in populism,Footnote ³² that is, what the said homogeneity consists of. It consists of an appropriation of the whole by a part; and not just any part but a specific part that needs to be socially recognisable.

Notice that the word demos, in the singular, had two meanings that were not clearly distinguishable in Greek.Footnote ³³ The ‘people’ was a term used to indicate broadly the set of citizens of a polis holding political rights (in classical Athens, all free and native-born adult males) but it was also used less broadly, to mean not the (political) citizenry as a whole, but its humbler members: the peasants, the sailors, the manual laborers in general. This explains the fact that Aristotle was able to define democracy not so much as government ‘of the many’, but as government ‘of the poor’.Footnote ³⁴ The very term employed – demos – obscures the distinction between the ‘People-as-a-whole’ (all members of a political community) and the ‘People-as-a-part’ (members of a class, i.e., the humbler members of society). Yet, it is indisputable that the objects denoted are two (all free and native-born adult males and the humbler members), although the word (demos) may be one. Indeed, not even the most radical anti-élitist populist would deny that the élite has political power, rather the opposite. Indeed, in populism, it is the political power of the élite that is decried.

In Book IV of the Politics Aristotle identifies different types of democracy. One type is ‘demagogic democracy’ that has many points in common with tyranny. This form of democracy has three main characteristics: (i) supremacy of the demos over the law, (ii) a direct emotional relationship between the leader and the masses, fuelled by a spirit of retaliation against the aristocratic minority, (iii) the use of ‘People’ as synonymous with ‘People-as-a-Part’: populus as plebs. The synecdoche of populism is that particular figure of speech that takes one for another, that is, that identifies demos in the second sense with demos in the first sense. This can be also conveyed with the Latin terms (perhaps more common in constitutional and political theory): it is that figure of speech which identifies populus (the constitutional notion of peoplehood) with plebs (i.e., a section or part, that is socially identifiable, most typically as the humbler members of society). Notice that demos in the second sense of plebs or People-as-a-part is never identical to populus or demos in the first sense which is by necessity composed by a larger number of individuals. The synecdoche of populism, therefore, consists in making the uncritical audience believe that the whole People would want what is really only what a section of it wants, namely a part that is both numerically and socially inferior to the whole. In the synecdoche of populism, the People-as-a-part (or demos in the second sense or plebs) is (i) numerically limited (fewer in number), (ii) bears social marks identifiable as ‘non-élite’ (which may come in a wide variety and is likely to be subject to social change) and (iii) projects itself as a ‘whole’, as a collective actor regardless of whether this is justifiable. Aristotle says that ‘the demos governs as a ‘whole’, ‘the people’ becomes a monarch, and ‘the many’ have power in their hands ‘not as individuals, but collectively’.Footnote ³⁵ This is the holistic conception of the people as an undifferentiated whole that recurs in the populist rhetoric.

So, let us recap what we noticed so far. Just like ‘population’ refers to two different things, so ‘People’ does. As clarified in the beginning of the chapter, ‘population’ may indicate all living human individuals within a determined spatio-temporal framework which is an empirical fact (population⁽¹⁾). Population may also indicate all persons recognised as such within a jurisdiction which is an institutional fact (population⁽²⁾). Also ‘People’ may indicate an empirical fact or an institutional fact. ‘People’, on the one hand, may indicate all (quasi-) social citizens within a given social setting (People-as-a-part), which is an empirical sociological fact.Footnote ³⁶ ‘People’, on the other hand, can also refer to all political citizens or members of the polity within a given constitutional setting (People-as-a-whole) which is an institutional fact.

Now, the populist use of the term ‘People’ is grounded in the view that the People-as-a-Part actually is, or embodies, the People-as-a-whole. When I say ‘embody the sources of all legitimate government’ I refer to a status-generative activity: a group in society (the People-as-a-part) – justifiably or not – claim to (re)define the institutional fact of the citizenry (the People-as-a-whole) in the manner that the group prefers. As such this synecdoche is not very surprising: it is a population design matter, and we have seen that any collective body by necessity engages in such design. What makes this synecdoche ‘populist’, however, is that it is unaware or oblivious of the fact that the People-as-a-part, which is an empirical social fact, and the People-as-a-whole, which is an institutional fact, are of a different nature and, thus, cannot be made to coincide. When the populist speaks in the name of the institutional fact of the People-as-a-whole to make the claim that a determinate interest, or course of action, is what is desired by ‘the People’, it is important to notice that this latter ‘People’ is not the same as the People-as-a-whole. This is why the synecdoche boils down to saying something along the lines of ‘this is what my social peers and I desire’. This may be a perfectly legitimate claim to make, were it not for its unfounded disguise as something that it cannot be: the interest of the ‘People-as-a-whole’ or bonum commune. Indeed, it is quite another matter to say that ‘this is what my social peers and I desire’ and to say ‘this is what is best for you all’.

Now, the people invoked as ‘the source of all legitimate government’ – the People that, I would argue, constitutionalists rely on, or intend to rely on – is the People-as-a-whole, not the People-as-a-part. In other words, for the populist and for the constitutionalist ‘the People’ actually means two different things. The semantic uses obey different rules. In the history of political thought, we can clearly see that there are two ideas in play here. First, there is the political-legal notion of the ‘People-as-a-whole’ that we may call ‘the People of constitution’ taken as the basis of legitimate power. Second, there is the sociological notion of ‘People-as-a-part’ usually meaning the ‘common people’,Footnote ³⁷ therefore, the same as plebs or demos in the second sense (namely the humbler members of society).

These two notions coexist and are at times conflated. What relation the ‘People-as-a-part’ should have to the ‘People-as-a-whole’ is a question that triggers political constitutional disagreement. At the risk of oversimplifying, the relationship between these two forms of people can either be one of identity or one of non-identity. If one believes they should be identical, one assumes the populist synecdoche: the People-as-a-part-who-should-actually-be-the-People-as-a-whole. When one assumes this synecdoche, it makes sense to use dichotomous terms to describe the political landscape. Think of the clash between a ‘people’ and a ‘non-people’, the underdogs and the custodians of an order founded on privilege. In political terms, populists will therefore typically think that you are either with or against us – allowing shades of grey is essentially perverse.

If one believes that the relationship between the two aforementioned forms of ‘People’ is one of non-identity, the ‘People-as-a-whole’ and ‘People-as-a-part’ remain separate: this will imply, for example, a distinction between the People of the State, for instance, all those having the nationality of a given country, and the people of the society that may include resident non-nationals. One way to understand this distinction is to say that the society (the People as a sociological, empirical category) is different from, or does not coincide with the People as the aforementioned Kelsenian fictio which is an institutional fact.

In sum, the populist use of ‘People’ consists in an appropriation of the whole by a part. We stand before a populist rhetoric when we face a synecdoche where the People-as-a-part claims to embody the source of all legitimate government, that is, the People-as-a-whole. The marker of a populist use of the term ‘People’ is the belief in the direct epistemic access of one group to what another group needs, wants, desires, etc. What is populist in the synecdoche is to claim that what the People-as-a-whole needs is known to the People-as-a-part.

Can there be circumstances in which such a claim is warranted? Perhaps. However, this may depend on the context. A caveat is needed here. I am not saying that there can be a justified identification of the People-as-a-whole by the People-as-a-part: it is always a conceptual error. But there can perhaps be a legitimate identification of the common interests by the People-as-a-part. As has been pointed out by others,Footnote ³⁸ the (possibly) legitimate identification of the interests of the whole by a part of the society is typical for the construction of the people as pouvoir constituant. Think about the age of revolutions when the idea of the people became a powerful drive for the worst-off in society (so only a part of it) who could use this populist synecdoche to raise issues that may very well have been justified. Think of the Levellers in England, or the more radical wing of the American and French revolutionaries for whom the idea of the ‘People-as-a-whole’ was understood to coincide with the ‘People-as-a-part’. It was not obvious that this appropriation would come across as justified in relation to the peuple constitué. While potentially apt to convey the intensity of revolutionary mobilisation, the populist view of the People might be ill-suited to make sense of the ordinary democratic tensions between collective subjects such as parties, unions, interest groups and the like who identify with a set of shared rules.Footnote ³⁹ Whether the synecdoche is politically justified or not will hence depend on the context.

This also explains a central point concerning the relationship between the populist appeal to the People and democratic decay. It is generally assumed that once we adopt the twentieth-century representative and constitutional conception of ‘party democracy’, the synecdoche seems obsolete. Its obsolescence, however, obtains only if we are sure about who is the constituted people/peuple constitué and its powers. This is something we are certain of only in the framework of accepted constitutional rules. Where such constitutional rules are the object of deep disagreements, for example, amidst a constitutional crisis, the synecdoche might resurge. This is in line with the observation by Pierre-André Taguieff that conjunctures that favour the rise of populist mobilisation include what we normally call a ‘legitimacy crisis’ of the political system.Footnote ⁴⁰ A so-called crisis of representation would count as a crisis of legitimacy. From this premise it is not far to conclude, like Mény and Surel, that ‘representation’ for populists means ‘treason’. One may think of crises of parties, of parliamentarism, of trust in institutions and such like as variants of a crisis of legitimacy. There is thus a connection between the populist synecdoche and democratic decay.

An important expression of democratic decay is lack of respect for pluralism. Democracy conceives politics as expressing a plurality of ‘parts’ which are no longer perceived as ‘factions’ that are destructive for the social body.Footnote ⁴¹ The populist view of the People relies instead on a reductionist view of politics that considers politics as struggle for power. Such a view rules out that conflict can, and perhaps should, end in a Kelsenian pursuit of compromise between forces that represent ‘partisan’ interests and opinions but do not claim to be the ‘only legitimate totality’.

Notice that this reductionist view of politics as struggle for power – that leaves a Machiavellian and perhaps Schmittian aftertaste – is not a mark of populism since it is also found in the debate over different models of democracy. In particular, the dichotomous manipulation of the political space is a typical characteristic of the ‘majoritarian’Footnote ⁴² or ‘immediate’Footnote ⁴³ forms of democracy which are based on the formally, or substantially, direct election of the head of the executive. Think of the presidential and semi-presidential systems, where the political battle culminates in an electoral face-off between two leaders. Valentina Pazé has thus asked ‘Is it a coincidence that America – the continent of populism – is also the home of presidentialism?’Footnote ⁴⁴ At the same time, the populist wave has surged through Europe precisely in a period when many parliamentary systems were ‘presidentialising’Footnote ⁴⁵ Without suggesting a causal implication, the simplistic logic of populism based on drawing a sharp line between ‘us’ and ‘them’ and on the direct relationship between the leader and the masses seems to be particularly suited for presidential systems, and particularly unfit for the institutional complexities of parliamentarism. In systems closer to ‘consensual’ and ‘mediated’ models of democracyFootnote ⁴⁶ the very institutional model translates awareness of the fact that there is a substantial difference between the empirical fact of ‘People-as-a-part’ and the institutional fact of ‘People-as-a-whole’. The constitutional setting of such models is articulated around, on one hand, the artificial character of the constitutional conception of the People and, on the other, the need to articulate and re-compose the plurality of interests that are voiced by sections of society (People-as-a-Part) through a variety of political forces that organise and articulate different visions of the world. The (sociological) People is dissolved into a multiplicity of people. In such systems, the Aristotelian distinction between People-as-a-whole and People-as-a-part is upheld.

2.8 Conclusion

Legal thinking concerns institutional facts. This ontological insight grounds the distinction between migration and mobility, only the first being an institutional fact. It is important to understand that migration statuses in the law are not necessarily grounded in mobility. Migrants and citizens may be fabricated by the law and they may disappear by the same magic. This implies that the design and management of migration statuses play an important part in shaping the ‘population’ of a country in ways relevant for legal and constitutional purposes. This substantiates the claim that migration and citizenship law enable population design. The aforementioned ontological insight also allowed me to distinguish two meanings of ‘population’. The first refers to the empirical fact of the sum of human beings within a given space and time, while the second refers to the institutional fact of the sum of persons recognised as such by the legal order. We are often oblivious of this distinction in discussing migration, border control and citizenship matters despite the fact that it so radically shapes our understanding of how these policy-areas impact the democratic constitutional setting. These policy areas directly or indirectly determine the extension and social composition of the population in a political community. They also determine the attribution to the individual of important rights and duties that co-determine that individual’s position within the society and within the power-setting of the constitutional framework. In this sense, these areas of law contribute to constructing the position that a given individual occupies vis-à-vis other individuals and their collective endeavours (state action). Since migration law typically governs access to residence rights and the latter are prerequisites for naturalisation in constitutional frameworks where political rights in the form of voting rights are most often reserved for nationals, migration law, in combination with citizenship policy and electoral laws, determine who counts as the People. The same ontological insight also allowed me to distinguish between two meanings of ‘People’ depending on whether we refer to the empirical fact of a set of members belonging to a society or to the institutional fact of the set of members of a polity within a given constitutional framework. This distinction reflects Aristotle’s distinction between the People-as-a-part and the People-as-a-whole. I also showed how populists employ these notions.

The overall picture I was able to paint shows that how we regulate access to and the content of citizenship determines how the demos, or set of citizens, is composed since the legal positions that are reserved for citizens determine the share of power that citizens enjoy in the constitutional order (i.e., how, when and where they may use political power). The composition of the institutional fact of the People, or set of citizens, and its role in the constitutional order – what the citizens do, which rights and duties they have – determines the state’s constitutional identity. Precisely because migration policy impacts on the constitutional bedrock of the state, there is a salient reason why the populist conception of the People ought to be of interest to both migration and constitutional lawyers. It is important to understand the causal triangulation between democratic decay, authoritarian populism and restriction of rights for migrants. One expression of this triangulation is the resurgence of the synecdoche that characterises populism: People-as-a-part is taken to embody the People-as-a-whole where a section of society thinks that it may speak for the whole to the effect that the irreducible pluralism of individuals composing the collective is muted. This denial is often a key step in the path to othering.