Narrating an ideal migration world? An analysis of the Global Compact for Safe, Orderly and Regular Migration

ABSTRACT The Global Compact for Safe, Orderly and Regular Migration (GCM) was adopted in 2018 under the auspices of the United Nations. In a context in which the global governance of international migration remains weak, and in which states regularly confront migration/refugee crises, the GCM provides an internationally agreed-upon assessment of the political issues raised by international migration and identifies non-binding policy recommendations regarding how governments should address them. This article analyses the content of the GCM in light of the migration-related discussions at the international level since the 1990s. It critically examines its core arguments and argues that, because of the diverging worldviews and interests among governments and other stakeholders, the GCM is marked by major internal contradictions. This results in a depoliticised document and in a political language that hides the dilemmas raised by migration politics. There is therefore little consensus among states regarding the nature of migration and the political responses to govern the transnational mobility of people.


Introduction
In 2018, 164 states adopted the Global Compact for Safe, Orderly and Regular Migration (GCM) under the auspices of the United Nations (UN). In a context in which the architecture of global migration governance is contested (Van Riemsdijk, Marchand, and Heins 2020), the objective of the GCM is to identify what an ideal, uncontested, migration world should look like.
The GCM presents itself as 'a milestone in the history of the global dialogue and international cooperation on migration' (para 6). 1 Yet, as it also recognises, 'discussions about international migration at the global level are not new' (para 3), and the GCM builds upon 20 years of international debates that have taken place inside international organisations (IOs) such as the International Organization for Migration (IOM), the International Labor Organization (ILO), the World Bank or the United Nations Development Programme (UNDP), as well as in intergovernmental fora like the Global Forum on Migration and Development (GFMD) or the UN High-Level Dialogues on Migration (HLD). Since the late 1990s, these discussions have resulted in a corpus of narratives about how migration should be governed (Pécoud 2015).
This article analyses the GCM's content and political orientations. Reaching a global consensus on migration is difficult, because migration is a divisive issue between states (especially along the sending/receiving divide) and within societies (between governments, civil society and the private sector). To overcome these disagreements, the GCM must develop a narrative that is compatible with the worldviews and interests of these different stakeholders in different parts of the world. This search for consensus leads to a depoliticisation of migration: the GCM cannot eliminate the controversies and disputes over migration, but can reach -on paper at least -a certain level of discursive coherence in which they are neutralised.
In light of the different meanings of the concept (Beveridge 2017), depoliticisation is here understood in the following way. The GCM (1) voluntarily stays away from the confrontations that characterise debates over migration in other arenas; this serves its purpose of (2) actively silencing political disagreements and dilemmas, through (3) the reliance on a technocratic and neutral language, and in order to (4) privilege a consensus that is presented as the only way forward for migration policy (thereby excluding alternative worldviews). The GCM's depoliticised language is therefore the opposite of discourses that favour antagonisms or conflicts. Such depoliticisation is frequently associated with IOs, because these organisations cannot work without a certain degree of intergovernmental consensus and themselves contribute to sideline contention (Petiteville 2018). Yet, as the debates around the GCM illustrate, there are limits to this strategy, and even the most sanitised approaches may find themselves in a politicised turmoil.
This article is constructed in the following way. It first provides a framework to understand the role of narratives in global (migration) governance. It then examines the genesis of the GCM in the framework of the broader history of global migration governance. The third section analyses the GCM's core arguments, distinguishing between long-standing and new topics. The fourth section sheds light on the contradictions that pervade the GCM.

Narratives, interstate cooperation and global migration governance
Global migration governance is an elusive concept that conveys at least two distinct meanings. It may designate the complex ways in which migration is governed at the international level. In such a meaning, global migration governance is usually described as weak, because states are keen on their sovereignty and display reluctance towards international cooperation. It is also fragmented, because different norms and mechanisms address different aspects of migration dynamics under the auspices of different organisations (Betts 2011). 2 This contrasts with the governance of refugees, which is the object of a regime 3 based upon the 1951 Convention on the Status of Refugees and the Office of the UN High Commissioner for Refugees.
According to another meaning, global migration governance is associated with the normative quest for 'better' governance. The assumption is that migration lacks appropriate governance mechanisms, a situation that contrasts with the multilateral cooperation over other types of transnational flows (goods, capital) and other global issues (development, climate change, human rights). From a functionalist perspective, global realities cannot be governed at the national level and lack of cooperation thus yields suboptimal results, in both socio-economic terms (concerning the benefits of migration for growth and development) and legal/moral terms (with the abuses and tragedies stemming from irregular migration).
In this context, three observations can be made about the GCM. First, it seeks to overcome the fragmentation of migration governance by addressing all the different issues raised by migration. It builds upon international standards in human rights, development, climate change, labour protection, or smuggling/trafficking (paras 2 and 6) and addresses the multiple ways in which migration relates to these. This makes for a comprehensive approach that contrasts with the previous assemblage of unrelated principles and norms. This being said, the GCM coexists with another compact: the Global Compact on Refugees (GCR) was drafted and adopted concomitantly, but remains a distinct document, concerned only with refugees. The distinction between migrants and refugees is thus both challenged and reaffirmed: the two categories are discussed in similar and parallel ways, but eventually they are the object of two separate texts.
Second, the GCM is about cooperation, both between states and between states and non-state actors. Because 'no State can address migration alone' , the GCM proposes a 'cooperative framework' and calls for 'international cooperation among all relevant actors on migration' (para 7). The GCM calls this a 'whole-of-society approach' and recommends 'multi-stakeholder partnerships' that bring together 'migrants, diasporas, local communities, civil society, academia, the private sector, parliamentarians, trade unions, National Human Rights Institutions, the media and other relevant stakeholders in migration governance' (para 15). This makes for what the GCM calls 'shared responsibilities' , thereby challenging influential views according to which migration is a matter of national sovereignty that should remain outside the realm of cooperation and of non-state actors' influence (Oelgemöller and Allinson 2020). This approach echoes broadly defined patterns of global governance: this notion is indeed regularly used to refer to decision-making processes characterised by their multi-stakeholder and multi-level nature, in which states cooperate with each other and with civil society, the private sector and IOs (Panizzon and van Riemsdijk 2019).
Strictly speaking, however, cooperation is about how to govern migration and says nothing of what should be done. This leads to the third feature of the GCM, namely its normative aspiration to propose a programmatic vision and to identify what an ideal migration world should look like. Going beyond mere calls for cooperation, the GCM lays down the desirable political horizon towards which all actors should progress (Jubilut and Casagrande 2019). This is also in line with global governance, which implies hybrid spaces of discussion in which no single actor can impose its views and take decisions alone -hence the elaboration of shared visions that pull stakeholders together (Rosenau and Czempiel 1992). The distinction between calls for cooperation, on the one hand, and the construction of an ideal migration world on the other is necessary because cooperation over migration may serve very different purposes. Western states, for instance, work with sending and transit countries to externalise border control and implement strategies of remote control (Zaiotti 2016); this entails cooperation, but towards an objective -the surveillance of human mobility -that is not exactly what the GCM advocates: as discussed below, the GCM calls rather for 'facilitating' migration.
By promoting multi-actor and multi-level cooperation, and by outlining the objectives of such cooperation, the GCM is an instrument of global migration governance in the second (normative) sense mentioned above. Yet, the word itself is absent from the text (there are only three occurrences of the expression 'migration governance' without 'global'). This may have to do with states' reluctance towards the concept: according to the International Catholic Migration Commission (2009), the notion of migration 'governance' too explicitly conveys the idea that governments should cede part of their sovereignty and is therefore met with caution. But the so-called 'Sutherland Report' 4 is less shy in this respect: it calls for 'improving the governance of migration' (United Nations 2017, 1) and explicitly envisages the 'good governance of migration' (United Nations 2017, 14).
Another way in which the GCM corresponds to the idea (if not the expression) of global migration governance is through its soft-law -or 'non-legally binding' -nature. Although the GCM is clear on this point (paras 7 and 15), this has spurred much debate. Part of the confusion stems from the uncertain nature of a 'compact': before the GCM and the GCR, the UN only used the expression once (with the 2000 'Global Compact' about the social responsibilities of private enterprises), and it was unclear whether states abiding by a 'compact' were engaging in binding commitments. In addition, while non-binding, the GCM anchors its recommendations in (binding) international human rights law (paras 1 and 2) and can therefore be interpreted as a political commitment to the hard-law instruments that compose international human rights law (Guild, Basaran, and Allinson 2019). It remains, however, that the GCM follows the dominant pattern in global migration governance, according to which states agree on principles -but without translating them into legal obligations: this contrasts with earlier efforts to adopt (binding) conventions, 5 which have nevertheless not been widely ratified (Pécoud 2009).
If one understands global governance as a process in which different stakeholders engage in cooperation over complex transnational issues, but without a clear hierarchical decision-making procedure, and without taking formal/legal commitments, then this confers an important role to narratives such as the GCM: if convincing enough, and if shared by a wide range of stakeholders, narratives may indeed pull actors together and support cooperation. From a constructivist perspective, they shape the cognitive structures through which an issue is made legible, and hence governed. They thus exert a strong political influenceand may ultimately turn into what Ruggie calls an 'episteme' , that is to say 'a dominant way of looking at reality' (Ruggie 1998, 55).
The cognitive/political function of narratives comes along with a social function. Shared narratives indeed support epistemic communities, or at least networks of stakeholders that, despite their differences, are united and connected by their adherence to similar worldviews: 'It is, in part, through adopting shared vocabularies, theories and explanations, that loose and flexible associations may be established between agents across time and space' (Miller and Rose 1990, 10). Migration manifests itself differently across regions, involves different stakeholders and is the object of different perceptions and understandings: in this context, the GCM aims to provide an overarching narrative that transcends these differences and facilitates cooperation.
The GCM presents itself as a catalogue of 'actions' , described as 'relevant policy instruments and best practices' (para 16). This echoes what Stone and Maxwell (2005) call a 'romantic' idea, according to which narratives provide evidence-based support to inspire policymaking. By contrast, a realist criticism posits that 'authority comes to language from outside' (Bourdieu 1991, 109) and that policymaking is therefore based not on ideas, but on states' interests: even the most convincing narratives will, then, fail to influence politics if not supported by powerful actors. Yet this disconnection between narratives and practice may stem not only from the powerlessness of ideas in a world governed by interests, but also from the very nature of political language. Rather than a guide to practice, narratives such as the GCM are then apprehended as a kind of myth: whatever their practical value, myths are central to societies because they support their cohesiveness and provide answers to questions that lack clear answers; in this anthropological vein, one of IOs' missions would be to produce ritual texts that cannot solve global problems, but outline a mythical horizon. As Cornwall and Brock write about development narratives, good argument has its place here, but is secondary to something that is of quite a different order: a feeling of rightness, backed by the creation of normative instruments, … which serve an almost ceremonial function in bolstering a feeling of togetherness and purposefulness, of a visionary goal towards which to strive. (Cornwall andBrock 2005, 1055)

Genesis of the GCM
International debates on migration are at least one century old. In 1919, the ILO mentioned in its constitution 'the protection of the interests of workers when employed in countries other than their own' . However, and in contrast with the refugee regime that developed in the second half of the twentieth century, migration remained a neglected issue inside the UN. There was no specific agency on this issue and the only IO specialised in this field -the IOM -went through a chaotic history and remained outside the UN until 2016 (Pécoud 2018). This started to change in the 1990s: in 1994, migration was discussed at the Cairo Conference on Population and Development and became the object of growing attention from the international community. Key steps in this process include the 1997 New International One of the outcomes of these ongoing discussions is the elaboration of 'international migration narratives' (IMNs), a corpus of arguments and worldviews that characterise the perception of migration at the UN level. The core arguments include (1) the construction of migration as a global issue that should be addressed globally; (2) a positive/optimistic appreciation of migration as a normal process that can benefit both sending and receiving societies, and migrants themselves ('triple win'); (3) an emphasis on the linkages between migration and other areas of international cooperation (such as development); and (4) an adherence to universal principles, including human rights, but also free-market through the recognition of the need to facilitate the mobility of labour (Pécoud 2015).
As discussed below, these arguments pervade the GCM. One of the reasons for which they were reassembled in yet another document was the euro-Mediterranean migration crisis: as of the early 2010s, this crisis illustrated the major governance challenges faced by states and the severe abuses stemming from badly governed migration (including in particular thousands of border deaths). In reaction, the UN organised the Global Summit on Refugees and Migrants in 2016, at which states adopted the New York Declaration for Refugees and Migrants and launched the elaboration of the GCM and the GCR. This, importantly, explains the GCM's Western-or euro-centric orientation (Martin Diaz and Aris escarcena 2019). After several months of consultations in 2017, a 'zero draft' was elaborated in February 2018; a few intermediary versions followed, before the GCM's final version was adopted in July of the same year. 6 This fairly inconspicuous process was first shaken in December 2017, when the United States decided to withdraw: the hostility of the Trump administration towards the UN -and towards multilateralism in general -did not come as a surprise, but weakened the GCM's ambitions. A few weeks before it was to be approved at the Intergovernmental Conference to Adopt the Global Compact for Safe, Orderly and Regular Migration in Marrakech in December 2018, the GCM found itself in the midst of unexpected turmoil, as more and more governments expressed concerns. The GCM became perceived as a symbol for liberal and migrant-friendly policies, and states such as Israel, Australia, Austria, Poland and Hungary also withdrew. even governments with a tradition of support for migration-related multilateral efforts -like Belgium and Switzerland -displayed reluctance (Ferris and Donato 2020). As Guild, Basaran and Allinson write, 'so much controversy when ratification will not alter the legal obligations of states seems unjustified ' (2019, 44). This being said, one of the GCM's objectives is to 'enhance availability and flexibility of pathways for regular migration' (para 21) -that is to say, to increase legal migration: in a world in which governments and populations regularly express scepticism towards migration, this remains an audacious claim.
The GCM is a document of 34 pages that relies upon 10 'guiding principles' and establishes 23 'objectives' . These objectives encompass a very wide range of issues, such as root causes (obj. 2), decent work (obj. 6), human smuggling and trafficking (obj. 9 and 10), border management (obj. 11), social cohesion and the fight against discrimination (obj. 16 and 17) or the relationship between migration and development (obj. 19 and 20). In addition, the GCM addresses other policy fields that are arguably less complex, such as the provision of information to migrants (obj. 3). This makes for a somewhat confusing document: at times, the GCM reads like a political manifesto urging high-level political leaders to design migration-friendly political strategies; but other parts of the text consist in fairly arid and technical policy recommendations to be implemented at much lower levels of responsibility.

Content of the GCM
This section analyses the content of the GCM. It distinguishes between the arguments that were already present in earlier IMNs and those that are new and specific to the GCM. Methodologically, it relies on qualitative content analysis; following Krippendorff, this entails the following steps: [1] sample text, in the sense of selecting what is relevant; [2] unitize text, in the sense of distinguishing words or propositions and using quotes or examples; [3] contextualize what they are reading in light of what they know about the circumstances surrounding the text; and [4] have specific research questions in mind. (Krippendorff 2004, 87) As far as points (1) and (2) are concerned, I have identified the GCM's key arguments and examined their content and the relationship between them, including through the selection of quotes and excerpts. This analysis is synchronic, as I examine only the final version of the GCM and do not provide a diachronic study of the successive versions of the text. Concerning point (3), I have relied upon available knowledge about the GCM but, given its recent adoption, there is limited research available on the political sociology of the drafting process. 7 As discussed in the introduction, the key research question (point 4) regards the ways in which the GCM overcomes the core disagreements and divergences between actors. This calls for unearthing the contradictions and dilemmas that are silenced in the text: to do so, I start by analysing the content of the GCM, before turning to a study of its internal contradictions.

Root causes
Objective 2 of the GCM is to 'minimize the adverse drivers and structural factors that compel people to leave their country of origin' . It mentions well-established development objectives, such as poverty eradication, food security, education, economic growth, infrastructure, decent work, gender equality, rule of law, good governance and human rights (para 18). Although development differentials play a central role in migration dynamics, the GCM devotes only one objective among 23 to this topic, on equal footing with arguably less decisive topics. On the other hand, the GCM is about migration, not development in general: if the GCM were to conclude that development is the solution to all migration problems, it would lose its raison d'être as there would be no point in designing migration-specific policy recommendations.
While Objective 2 is about the effect of underdevelopment on migration, the GCM also looks at the effect of migration on development, through remittances, for example (see below). Migration is therefore portrayed as both an undesirable outcome of underdevelopment and as a strategy to remedy such root causes. The relationship between migration and development is thus ambivalent: some governments (principally in receiving countries) see development as a strategy to reduce migration, while others (mainly in the Global South) see (more) migration as a way to achieve development (Geiger and Pécoud 2013).

Labour migration
Labour migration is one of the GCM's main concerns. This is reflected in Objective 5 (entitled 'enhance availability and flexibility of pathways for regular migration') and in Objective 6 (about 'decent work'). The GCM believes that labour mobility should be 'facilitated' 8 because of 'demographic and labour market realities' (para 21) and that migrant workers should be protected 'against all forms of exploitation and abuse' (para 22). Objective 18 further examines 'skills development' and the 'mutual recognition of … qualifications and competences' , with the ambition of making it easier for migrants to be employable in both sending and receiving labour markets. Objective 22 addresses 'the portability of social security entitlements and earned benefits' , with the aim of enabling labour migrants to enjoy social protection even when they move from one country to another. Objectives 18 and 22 thus envisage a kind of global labour market in which workers would circulate across borders with minimal administrative hassle. This nevertheless leaves out the role of irregular migration and undeclared work. The GCM focuses on 'formal labour markets' (para 34) and ignores the existence of informal labour markets in which migrants often concentrate (Crépeau 2019). This reflects the GCM's tendency towards a kind of 'wishful thinking': it discusses problems and outlines solutions, but also silences other aspects of social reality -as if unnamed realities would simply cease to exist. The GCM further displays a number of contradictions. This is especially the case with temporary labour migration, which is often advocated as an ideal policy scenario: receiving economies would benefit from foreign labour and avoid the costs and social tensions associated with permanent migration, while sending countries would limit the brain drain and benefit from the skills acquired by their workers abroad. The GCM thus calls for 'flexible … labour mobility schemes … at all skills levels, including temporary, seasonal, circular, and fast-track programmes' (para 21).
At the same time, however, the GCM recommends 'fair and ethical recruitment processes that allow migrants to change employers and modify the conditions or length of their stay' (para 22); it also wants to 'facilitate access to … family reunification for migrants at all skills levels' (para 21). But temporary labour migration tends to prevent migrants from changing employers and exclude their families. As Martin (2018) argues, it is therefore difficult to understand what the GCM has in mind as it features policy recommendations that seem largely incompatible with each other.

Human rights and protection
Human rights constitute one of the nine 'guiding principles' of the GCM, which wants to 'ensure effective respect, protection and fulfilment of the human rights of all migrants, regardless of their migration status, across all stages of the migration cycle' (para 15). This is reiterated throughout the document, with 58 occurrences of the words 'human rights' . Yet there is no objective about human rights, which are therefore addressed here and there in other objectives. Objective 7 is about the 'vulnerabilities in migration' that arise 'from the circumstances in which [migrants] travel or the conditions they face in countries of origin, transit and destination' , and which should be addressed by reviewing legislation, policies and practices as well as by specific gender-and child-sensitive strategies (para 23).
Relatedly, Objective 14 calls for sending states to improve 'consular protection' and 'assistance' to their emigrants, while Objective 15 discusses 'access to basic services for migrants' (mainly as far as health and education are concerned). Finally, and as already mentioned above, Objective 6 is about labour protection and 'decent work' . The absence of a specific objective on human rights is all the more visible because other objectives address aspects of immigration policy that have human rights implications, like immigration detention (obj. 13) and return/readmission (obj. 21).

Migration and development
As discussed in relation to root causes, the relationship between migration and development is a central but ambivalent issue in the GCM. Along with human rights, another 'guiding principle' is sustainable development: the GCM refers to the Sustainable Development Goals (SDGs) and believes that 'migration contributes to positive development outcomes …, especially when it is properly managed' (para 15). This fits into an 'optimistic' interpretation of the migration and development nexus, which characterises IOs' agenda (De Haas 2010).
Objectives 19 and 20 address the two major arguments that support such optimism, namely remittances and the role of diasporas in fostering the development of their homeland.
By contrast, there is only one occurrence of the word 'brain drain' , a phenomenon that is associated with the negative impact of migration on sending regions (Levatino and Pécoud 2012). Here again, the GCM displays 'wishful thinking' and silences a reality that challenges its optimism. Objective 6 addresses 'fair and ethical recruitment': yet while IOs sometimes use this expression to refer to the recruitment of skilled migrants, 9 the GCM associates it rather with the role of labour brokers and intermediaries in fostering exploitation. Highly skilled migration (and the trade-offs it raises) is absent from the GCM.

Circular, temporary and return migration and reintegration
Circular and temporary migration refer to policy strategies that facilitate labour mobility. As discussed above, this is the focus of Objectives 5, 6, 18 and 22. Return migration and reintegration designate another aspect of such circulation, namely migrants' return to their country, and often constitute euphemisms for coerced return or expulsion. In the GCM, this is addressed in Objective 21, entitled 'Cooperate in facilitating safe and dignified return and readmission, as well as sustainable reintegration' . 'Facilitation' has a different meaning here: the goal is to not to make migration easier, but to return migrants more easily to their country. This return should be 'safe' and 'dignified' , which means not returning migrants to countries where their lives would be in danger (non-refoulement principle). This also entails carrying out 'individual assessments' (para 37), which excludes collective expulsions and is in line with another of the GCM's 'guiding principles' about 'Rule of law and due process' . 'Sustainable reintegration' means that sending states should readmit their citizens and that socio-economic conditions should make it possible for returned migrants to resettle there.

Irregular migration
While not the focus of a specific objective, irregular migration is a topic that runs throughout the GCM. As noted above, the GCM is a reaction to the 'migration crisis' and to states' perceived inability to keep unauthorised migration under control. Objective 4, about 'proof of legal identity and adequate documentation' , aspires at 'allowing national and local authorities to ascertain a migrant's legal identity upon entry, during stay, and for return' (para 20). Objectives 9 and 10, which address the issues of human smuggling and trafficking, refer to well-established policy efforts to dismantle the networks that help migrants cross borders unlawfully and exploit them in receiving countries. Objectives 13 (about migration detention) and 21 (about return) are also connected to the issue of irregular migration.
In addition, Objective 11 addresses border management with the aim of increasing coordination between states. This objective also deals with 'the particular situation of countries of transit' , along with pre-border control measures and cooperation with private carriers (para 27). Objective 12 develops the need for 'the adequate and timely screening and individual assessment of all migrants' (para 28). Irregular migration is also indirectly present in other objectives, such as in the above-discussed labour migration objectives (as legal labour migration is understood as a strategy to combat irregular migration and undeclared work). Root causes and development objectives also aim at reducing emigration pressures in sending regions.
In general, the GCM refers to 'migrants' , an all-encompassing category that seems to include both regular and irregular migrants. On certain occasions, however, it refers to 'all migrants, regardless of their migration status' (see para 31 for example). This implies a distinct treatment of documented and undocumented migrants, but nowhere does the GCM make explicit the distinct treatment that should apply to these two categories. Crépeau (2019) observes that the GCM does not use the notion of a 'firewall': this notion refers to the separation between immigration enforcement activities and the provision of essential services, which is necessary to secure irregular migrants' access to welfare and protection mechanism. Objective 15 discusses the 'basic services' that should be made available to all migrants, but does not envisage a situation in which undocumented migrants would refrain from asking for such services for fears of being identified. Finally, the GCM makes no reference to the regularisation of irregular migrants, a sensitive issue that was present in the first draft but subsequently disappeared from the text (Guild, Basaran, and Allinson 2019).

Social cohesion and integration
These two issues are addressed in Objectives 16 and 17. The first is about the 'full inclusion' of migrants in societies, with the idea that well-integrated migrants are better empowered 'to contribute to prosperity' (para 32). The second is about the fight against discrimination, racism, xenophobia and intolerance. These objectives are in line with the general emphasis on 'inclusive societies' since the 1995 World Summit for Social Development. As already noted, there is a contradiction between the call for temporary labour migration and the 'full inclusion' of migrants.
The GCM ties migrants' inclusion to the issue of information. Public opinion would misperceive migration as a burden and should therefore be instructed about 'the positive contributions of safe, orderly and regular migration' (para 33). Objective 3 calls for the diffusion of better information both to migrants and to the inhabitants of receiving regions. This raises a dilemma with respect to freedom of expression: the GCM insists that its recommendations should be implemented 'while upholding international human rights law, in particular the right to freedom of expression' and 'in full respect for the freedom of the media' (para 33); but at the same time it indicates that the general message conveyed by the media on migration should be about their 'positive contributions' .

Gender
'Gender-responsiveness' is one of the GCM's 'guiding principles' , and the word 'gender' appears 27 times. Most often (17 times), it is used in the expression 'gender-responsive' and coupled with 'child-sensitive' . Gender is particularly referred to in Objectives 7 (about vulnerable groups) and 10 (about trafficking). But as with human rights, there is no objective devoted to gender, nor are there any references to issues that are central in this respect, such as sexual and reproductive health services (which are not included in the 'basic services' listed in Objective 15). Women remain apprehended as a 'special group' along with other vulnerable groups like children or migrants with disabilities, within a framework centred on 'protection' rather than rights (Hennebry 2018).

Governance and cooperation
As discussed above, the GCM is by nature about governance and cooperation. 'International cooperation' is another 'guiding principle' and is necessary because 'no State can address migration on its own due to the inherently transnational nature of the phenomenon' (para 15). Virtually all objectives mention cooperation, at bilateral, regional or international levels. Cooperation is also the focus of a dedicated objective: Objective 23, the last one, is about 'strengthen[ing] international cooperation and global partnerships for safe, orderly and regular migration' , with an emphasis on helping weak states to implement the GCM.
In addition, the GCM announces a follow-up mechanism to monitor its implementation. The UN HLD is to be replaced by an International Migration Review Forum, scheduled to take place every four years as of 2022. These last pages indicate that when it comes to migration, cooperation remains something to be achieved. In other fields of international cooperation, the key actors and the operational mechanisms are fairly well established and the debate focuses on what should be done. By contrast, the GCM needs to reaffirm the need for cooperation, while at the same time reiterating its respect for state sovereignty (which constitutes another 'guiding principle'), and crafting new implementation mechanisms.

Evidence, research and data
evidence, research and data are constitutive of the evidence-based philosophy that inspires the GCM. They are so important that they are the subject of the first Objective, entitled 'Collect and utilize accurate and disaggregated data as a basis for evidence-based policies' . In addition, other objectives are indirectly based on the existence of reliable knowledge, such as Objective 3 (about the need for information), Objective 4 (about the identity documents that make data collection possible), and Objectives 11 and 12 (about border management, 'certainty' and 'predictability' in migration procedures).

New issues
The previous section shows the continuity between earlier IMNs and the GCM. But the GCM also contains a number of new topics; four of them are identified in this section.
The first is about saving lives. Objective 8 calls for states 'to cooperate internationally to save lives and prevent migrant deaths and injuries through … search and rescue operations' . It further commits 'to identify those who have died or gone missing, and to facilitate communication with affected families' . These recommendations reflect the ongoing migration crisis in the Mediterranean Sea, but this context is not mentioned. By ignoring the context that underlies its existence, the GCM is a 'self-constituting discourse' (Maingueneau 1999), namely a text that detaches itself from its context and thereby claims a universal, 'disconnected' nature that legitimises its content and message. This obscures some of the GCM's biases (such as its Western bias), while depoliticising its recommendations, as it appears to be a general and abstract statement rather than the outcome of specific social and political dilemmas.
The second innovation concerns border management. The GCM is strongly concerned with border control and the monitoring of human mobility. Objective 11 is devoted to border management, Objective 4 to the need for migrants to carry proper identity documents, and Objective 12 to the 'screening' of migrants. The notion of 'screening' is not very detailed, but seems to refer to mechanisms that would identify migrants, provide them with information on what is required from them (to enter a country, work, study, etc.) and channel them into the right policy direction (for example, in case they are entitled to protection mechanisms). This leads to the third new policy area, namely detention. Objective 13 recommends 'us [ing] immigration detention only as a measure of last resort and work[ing] towards alternatives' . This is an issue of contention, especially as far as child detention is concerned (Muntarbhorn 2019;Tennant and Wolff 2018). Civil society was hoping for the GCM to forbid child detention, but the GCM only calls for 'working to end the practice of child detention' (para 29).
Although border control is an obvious priority for states, and despite mounting evidence that governments have recourse to immigrant detention, IMNs had so far stayed away from these sensitive issues. The GCM, by contrast, devotes attention to them. The result is a document that is more control-oriented than earlier IMNs. While this enables a more explicit balance between the different concerns of states and other stakeholders, it exposes the GCM to criticism, especially from non-governmental organisations (NGOs)without, as the heated debates surrounding its adoption indicate, preventing states' criticism.
Finally, the GCM is characterised by a strong emphasis on climate-related issues. This reflects the adoption of the SDGs in 2015, which mix socio-economic development with environmental concerns. The GCM has no objective exclusively devoted to climate change, but Objective 2 has a subsection on 'natural disasters' , 'climate change' and 'environmental degradation' (para 18). Aleinikoff (2019) recalls that there have been debates as to whether people moving for environmental reasons should be categorised as 'refugees' or 'migrants': this issue could arguably have been dealt with in either of the two compacts, or in both. Moreover, the 2016 New York Declaration (which dealt with both migrants and refugees) presented climate change as one of the many factors that propel people to leave their country (para 1). 10 Yet governments decided to keep this topic out of the GCR and to address it only in the GCM. The reason was that, given the binding nature of international refugee law, including climate change in the GCR could have been interpreted as creating new legal commitments for states, whereas no such risk exists in the case of the GCM (as global migration governance is characterised by its non-binding nature). Kälin (2019) nevertheless argues that the inclusion of this issue in the GCM is a breakthrough: at the international level, it had traditionally been discussed in negotiations over climate change, as one of the 'human consequences' of global warming, but discussions over migration had tended to downplay this issue (Gemenne 2011).

Discussion
By explaining how migration should take place and how states should govern it, the GCM portrays an ideal migration world. This is one of IOs' traditional missions: as 'missionaries of our time' (Barnett and Finnemore 1999, 713), they outline the broad direction that societiesand the world at large -should follow. But this ideal migration world is crisscrossed by tensions, dilemmas and contradictions. For example, the GCM calls for integration and social cohesion while at the same time advocating for circulation, temporary labour migration, and the persistence of strong ties between migrants and their homelands (which support remittances and diaspora involvement for development). This reflects doubts about the status of migrants, who remain 'birds of passage' (Piore 1979) while also being expected to become full members of receiving societies.
The GCM also oscillates between global and national worldviews. It envisages a world society in which people smoothly circulate in a 'facilitated' way: this is evident in the idea of increasing labour migration and in the recommendations pertaining to the recognition of skills or the portability of social entitlements. At the same time, it is committed to state sovereignty and a kind of national world order: this pervades the emphasis on border management and the concern with return and readmission, both of which imply that states remain key actors in the governance of migration.
The GCM is further torn between a laissez-faire and a planning approach. According to one of its 'guiding principles' , the GCM is 'people-centered': it 'places individuals at its core' (para 15) and wants to 'empower migrants to become full members of our societies, highlight their positive contributions, and promote inclusion and social cohesion' (para 13). Migrants are portrayed as actors who display agency across borders in a globalising world. Yet the GCM is also obsessed with planning: states would need 'greater predictability and certainty' , and 'migration works for all' only if it takes place 'in a well-informed, planned and consensual manner' (para 13). The international community should therefore 'better map, understand, predict and address migration movements' (para 18). In other words, migrants are encouraged to become active members of sending and receiving societies, but their mobility should also be closely monitored by governments.
This planning philosophy contrasts with calls for 'flexible' migration programmes to adapt to labour market needs (para 21). The GCM believes that labour mobility and openness to foreign labour are necessary to achieve growth and prosperity; it combines this market-friendly belief with an emphasis on planning, as well as on 'social' factors (like human rights and social inclusion). This is in line with the 'post-Washington consensus' , namely the conciliation of market-based dynamics and state regulation over protection and cohesion. But just as it ignores the geopolitical factors underlying its existence, the GCM never reflects on the current socio-economic organisation of the world and the particular historical setting in which migration now takes place: for instance, and as argued above, it makes no reference to the deregulation of labour markets in advanced economies or to the role of cheap undeclared labour in certain segments of the economy. This overall economic and political context is naturalised, and therefore uncontestable and uncontested.
The obsession with planning also unfolds into an overarching concern with the transparency and legibility of migration. The GCM repeatedly affirms the need for migration (and migrants) to be better documented (as in Objective 1 on data, Objective 3 on information and Objective 4 on identity documents). Border management and biometrics are also about the systematic documentation of people and flows. The GCM calls for systematic exchange of information and for improved databases, in a way inspired by 'big data' technologies. This is connected to the fight against irregular migration and to the GCM's rejection of the hidden (or underground) nature of unauthorised flows. The emphasis on knowledge and information indirectly reveals the belief according to which current migration dynamics are obscure and insufficiently documented. As Guild (2019) observes, this raises concerns about privacy and protection of personal data (see also Glouftsios and Scheel 2020). The GCM is aware of the dilemma: the expression 'while upholding the right to privacy and protecting personal data' (or equivalent) is used 10 times in the text.
It is interesting to observe that some of the unresolved issues raised by the GCM are addressed as 'guiding principles' rather than as 'objectives' . This is especially the case with the tryptic human rights-national sovereignty-international cooperation, which make for an arguably extremely complex equation; this is also the case with the cooperation between state and non-state actors ('whole-of-society approach'), agency ('people-centered') and gender. As 'guiding principles' , these topics are not associated with in-depth policy recommendations (as are the GCM's 'objectives'): they seem to be taken for granted and to go without saying. It follows that the GCM devotes much more practical attention to issues that could be deemed of secondary importance, such as information (obj. 3), remittances (obj. 20) or the portability of social entitlements (obj. 22), but hardly says anything about how to cooperate over migration policy while at the same time upholding state sovereignty and migrants' rights. In an ambiguous move, the GCM sanctifies these core principles, but thereby places them outside the realm of discussion and avoids venturing into the inevitably disputed modalities through which they could be implemented.
In sum, the GCM pursues a double objective: facilitating migration and empowering migrants, on the one hand, and monitoring their mobility through cooperation, technology and data on the other. Ultimately, the key unresolved question concerns the normality of migration. The GCM recalls that 'migration has been part of the human experience throughout history' and that 'the majority of migrants around the world today travel, live and work in a safe, orderly and regular manner' (para 8). This would indicate that migration is a long-standing, normal and unproblematic reality. But at the same time the GCM is concerned with root causes and with eradicating the structural factors that 'compel [people] to seek a livelihood elsewhere through irregular migration' (para 18). By further arguing that 'migration should never be an act of desperation' (para 13), the GCM seems to imply that 'normal' migration is free and chosen by migrants, whereas unfree migration is 'abnormal' and should be combatted. There is evidence, however, that such a policy and conceptual distinction does not connect easily to reality (Crawley and Skleparis 2018).

Conclusion
The GCM's ideal migration world maintains a symmetric/opposite relationship to the current state of affairs: for example, when it speaks of 'orderly' migration, this indicates that today's migration is disorderly. The GCM is thus based upon a dissatisfaction with current migration realities. But while this dissatisfaction is shared by many actors (from governments confronted with migration crises to NGOs concerned with abuses, or employers in need of foreign labour), there is less consensus on the nature of a satisfactory migration world.
The GCM develops a consensual narrative that leaves out unresolved questions (such as the respective role of states and of migrants' agency), or provides multiple and contradictory answers. It is in this respect that the GCM depoliticises migration, by silencing or dodging the core political dilemmas raised by its topic. Such depoliticisation is a structural corollary of the quest for better global migration governance, which -as argued -requires a consensual and federating narrative to steer multilateral cooperation.
This does not prevent criticisms. NGOs and critical observers have long challenged IMNs: they regret their lack of emphasis on binding human rights and protection mechanisms, the reluctance to eradicate problematic practices (like detention), the indifference to certain issues (like gender or the exploitation of migrant labour), the overall neoliberal philosophy according to which migrants should provide the flexible and mobile labour necessary to the global economy, or the Northern/Western-centric bias of the discussions (Piper and Rother 2012).
These criticisms now coexist with states' criticisms. What is new about the GCM may lie not inside the text, but outside. IMNs had been characterised by a bureaucratic routine that saw governments and other stakeholders discuss behind closed doors and away from public scrutiny. By contrast, and somewhat unexpectedly, the GCM has spurred a kind of global debate on migration. States' criticisms are largely based on misunderstandings (or bad faith) and will likely further jeopardise the already uncertain implementation of the GCM. But the GCM does address issues that are high on the political agenda in many countries, and which raise major political dilemmas. There is a certain logic, therefore, to seeing the GCM -and, by extension, the architecture of global migration governance -exposed to public scrutiny, with all the polarisation that this entails. states and non-state actors to cooperate over certain issues, thereby lessening the costs stemming from non-cooperation. Typically, a regime combines norms with an institutional actor to implement them. 4. This report, named after the late UN Special Representative for International Migration, Peter Sutherland, was submitted to the UN Secretary-General to 'inform' the GCM's drafting process. 5. These include the ILO Conventions 97 and 143 (Migration for employment Convention (Revised) in 1949 and Migrant Workers (Supplementary Provisions) Convention in 1975), as well as the UN 1990 International Convention on the Protection of the Rights of All Migrant Workers and Members of their Families. 6. All versions can be found at https://refugeesmigrants.un.org/intergovernmental-negotiations 7. Rother and Steinhilper note that 'the final draft […] has been seen as watered down in comparison to the zero draft' (Rother and Steinhilper 2019, 253). Maiyegun (2019) looks at the influence of African states in the drafting process. For insights into the role of europe, see Martin Diaz and Aris escarcena (2019) and Melin (2019). For a chronology of the process and a discussion of the role of the United States, see Ferris and Donato (2020, 100-122). For a diachronic analysis, see the information provided by the Refugee Law Initiative at https://rli.blogs.sas.ac. uk/themed-content/global-compact-for-migration/. For additional insights, see the special issues of the International Journal of Refugee Law (30(4), 2018), Global Social Policy (18 (3), 2018) and International Migration (57(6), 2019). 8. The word 'facilitate' seems to embody the GCM's approach to migration; it is used 62 times in the document (Crépeau 2019). 9. This is the case of the World Health Organization (WHO), which has issued recommendations to lessen the negative impact of the recruitment of foreign health professionals on sending countries (Merçay 2014). 10. The full text of the New York Declaration can be found on the UN website (General Assembly document A/ReS/71/1), at https://www.un.org/en/development/desa/population/migration/ generalassembly/docs/globalcompact/A_ReS_71_1.pdf (accessed 17 April 2020).