1The issue of migration has spawned abundant research and prompted wide-ranging theoretical debate. A selection of papers, articles and book chapters spanning several decades, many of which were first written in English, have been translated into French and brought together in a book published by INED as part of a new series devoted to the founding texts of demographic theory. These texts were chosen for their original contributions to the discipline. They represent landmarks of demographic thought, and provide new insights for analysing and understanding demographic processes. The inaugural volume of this series was edited by Victor Piché. In an introductory chapter, of which this article is an abridged and slightly revised version, he places these founding texts in their historical perspective. He highlights the diversity of approaches applied to understanding migration: countries of origin or destination; micro, meso and macro levels; individual behaviours, networks, migration policies. He also shows how these theories have evolved to take account of changing local and international migration dynamics.
2This article explores the development of contemporary migration theories as reflected in some twenty founding texts that have marked the field over the last fifty years.  Before proceeding further, two points must be made clear. First, by founding texts, we refer to the landmark articles or book chapters that have shaped the progress of migration studies. They are constantly recurring references, both in the theoretical frameworks used by scholars in their empirical studies, and in the numerous literature reviews offering critical overviews of these theories. The second point concerns the period covered, namely the decades from the 1960s up to the year 2000. Of course, by choosing this period, we exclude many great figures of sociology and economics – Comte, Durkheim, Weber, Marx and Smith, to name but a few. Yet while these major classics have marked the social sciences, their contributions to the field of migration are few. It would be of utmost interest to explore the conception of mobility in these great works, but our purpose is elsewhere.
3Our first question concerns the very definition of migration. The first selected text is by Alan Simmons (1987*). It is a pioneering study which aims to clarify the various definitions and typologies and, above all, to place migration theories in their historical context. He suggests using three major dimensions to define migration: a change in residence, a shift in employment and a shift in social relations. In general, the first dimension – a change of residence – is the main criterion used. Simmons suggested that this definition be broadened, and his innovative idea rapidly gained ground, notably in research focusing on macro-structural dimensions.
4Simmons also observes that the field is highly fragmented because migration theories cover specific types of migration grounded in particular social and historical contexts. This fragmentation is especially discernible in an area that has dominated migration research, namely the distinction between the causes and effects of migration. But it also affects the levels of analysis, be it micro, macro or meso. In the presentation of the founding texts, we address this fragmentation by distinguishing between causes and effects on the one hand, and between micro-individual and macro-structural approaches on the other.
I – Origins and causes of migration
5One of the very first explanatory approaches to both internal and international migration focused on individual decision-making. Before deciding to leave their place of residence, individuals examine the costs and benefits of migrating. This approach is often associated with the paper by Larry Sjaastad published in 1962*, in which he sought to identify the costs and returns and to determine the “rate of return on resources allocated to migration”. He sees migration as an “investment increasing the productivity of human resources, an investment which has costs and which also renders returns”. Costs can be broken down into money and non-money costs.
6Without doubt, Sjaastad’s greatest contribution was to introduce the notion of human capital into migration theory to get around the problem of estimating returns. For Sjaastad, “it is particularly useful to employ the human capital concept and to view migration, training, and experience as investments in the human agent”. The basic postulate of his approach is explicit, namely that the analysis of private costs and returns is valid only in the case of voluntary migration which, in a competitive economy, satisfies the requirement of “optimum” allocation of resources.
7Sjaastad’s paper laid the groundwork for the general schema for migration presented by Everett Lee in 1966*. Under this schema, migrant characteristics provide a means to explain volume of migration, migration streams and counter-streams. He begins by postulating that migration is the result of an individual calculation based on positive factors at destination and negative factors at origin. Lee’s model introduces the original concept of intervening opportunities between the places of origin and destination. He points out that it is not so much the actual factors at origin and destination as the perception of these factors which results in migration. Among the factors influencing the migration decision, Lee mentions personal contacts and sources of information about the situation at destination. This prefigures the notion of migration networks that was to become central to migration theory from the 1980s.
8Alongside the notions of negative and positive factors, and of intervening obstacles and opportunities, Lee alludes, when discussing Ravenstein’s laws of migration (1885, 1889), to additional factors that would later be developed by other scholars. They include, for example, the notion of specialization in particular skills and occupations that prefigures the hypothesis of labour market segmentation advanced by Portes (1981) among others, and which will be discussed below. He also talks about discrimination and the inauguration of “other kinds of diversity among people”, laying the foundations for wide-ranging subsequent research on multiracial and multicultural societies. He also affirms that migration increases with time, and that this increase is driven by growing economic disparities between developed and developing countries, by education and training, and by technological progress, notably in communications and transportation. These are precursors of the research questions that were set to dominate the scientific literature on international migration in the context of globalization. Last, he notes that the increase in migration may also be due to migration itself, given that the first waves of migrants overcome the intervening obstacles, making it easier for subsequent waves to follow in their path. It is tempting to see in this intuition the notion of cumulative causation developed by Douglas Massey (1990*), which will be discussed further below.
9While Lee’s model is mentioned in literature reviews on migration theories, it has been widely criticized. Two aspects deserve attention. First, the critics agree that it is not a theory as such, but rather a conceptual framework for classifying the various factors that explain migration. Second, critics note the predominance of micro-individual factors, primarily those linked to human capital, almost to the exclusion of macro-structural factors. Indeed, Lee’s model is inseparable from the microeconomic postulate of voluntary migration within a competitive economy, a postulate central to the microeconomic theory of migration that focuses on individual choice. This theoretical model was expanded by Michael Todaro (Todaro, 1969; Harris and Todaro, 1970) and Borjas (1989). The considerable merit of these approaches is to have added the notion of expected gains or, in the words of Sjaastad, the “expected” net return on investment.
10Remaining within the micro-individual approach, another question concerns the evolution of migration patterns. Wilbur Zelinski (1971*) was among the first to outline a theory of mobility based on the notion of transition so dear to demographers. He seeks to include migration in the theory of demographic transition which traditionally focuses solely on changes in fertility and mortality. Like the classic theory of demographic transition, the theory of the mobility transition (Zelinski speaks more modestly of a hypothesis) forms part of the theory of modernization that dominated in the 1970s.
11Despite the old and outdated vocabulary of Zelinski’s article, his contribution is significant on two levels. First, he proposes to incorporate migration into the general theory of demographic transition. This provides a means to sidestep the fragmentary nature of the demographic field by calling upon the notion of demographic regime. Throughout history, each society has developed demographic reproduction strategies by combining the reproductive mechanisms of fertility, mortality and migration. Migration is not, therefore, an isolated strategy, but ties in with other demographic behaviours (Gregory and Piche?, 1985; Mertens, 1995). Second, even though Zelinski’s presentation of the demographic (or “vital”) transition is now largely outdated, that of the mobility transition remains pertinent, especially in its advanced phase. Unfortunately, few other scholars have continued along this path. The work of Alan Simmons (1995, 2002) is an exception, establishing an explicit link between the historical development of migration patterns, their social and economic significance, and the various phases in the development of capitalism and globalization.
12Zelinski’s approach is most commonly criticized for its evolutionist outlook, based on the theory of modernization. Until today, this “western-centric” approach has remained central to the conception of social change and development in demography. Indeed, the first formulations of the theory of demographic transition, published in the height of the colonial period, are strongly influenced by evolutionism and present “traditional and non-industrialized” societies as a reverse reflection of modern, industrialized societies. In an evolutionist perspective, these societies will develop if they adopt more modern structures and the attitudes that underpin them. Today, the evolutionist perspective has been almost entirely abandoned, in sociology and anthropology at least, in part due to the influence of the postmodern approach which contests the universality of theories in the social sciences. While postmodernism has never held sway in demography,  two research currents have exerted a certain influence in the field of migration theory. A first current challenges the claim to universality of statistical categories, suggesting that categories are historically determined social and political constructions (Szreter et al., 2004; Cordell, 2010). This critical approach has been applied to question the official categories produced by censuses, notably racial and ethnic categories (Nobles, 2000; Simon and Piche?, 2012). A second current, linked to postcolonial studies, examines the consequences of transferring the human and symbolic colonial legacy to the heart of the metropolis. Indeed, unlike immigration from Europe, migration from former colonies, designated as postcolonial, possesses specific features “resulting as much from the shared experience of the colony as from the continued experience of the post-colony at destination, marked by ethnic and racial prejudice and by discrimination” (Simon, 2010, p. 362).
13For several scholars, decisions to migrate can only be understood in a more global context. A first approach sought to explain migration patterns in terms of a system of multiple flows between origin and destination places: flows of persons, but also of goods, services and ideas. It was Akin Mabogunje (1970*) who first proposed a systems approach. His analytical framework sought to identify all the elements liable to influence migration, ranging from the economic environment to technology, the social environment and political factors. He also mentions two other factors that were later to receive much attention from scholars, namely the key role of “information” and continued “feedback” with the place of origin, opening the door to numerous studies on the importance of social and family networks and of monetary transfers in the migration process. Mabogunje’s approach sees migration not as a linear, one-way movement, but as a circular phenomenon embedded in a system of interdependent variables.
14True, the systems approach is not easy to operationalize, given the wide array of factors he identified. It nonetheless gives rise to a conception of international migration that ties in with globalization, even suggesting the idea of a world labour market in a globalized economy (Petras, 1981; Simmons, 2002). This global perspective became increasingly fashionable from the 2000s, and generated abundant literature on transnational networks (Schiller et al., 1992; Faist, 2000; Vertovec, 2009). We will return to this point later in relation to the founding text by Oberai and Manmohan (1980*).
15As mentioned earlier, a characteristic inherent to the systemic approach is that of circular migration, a notion theorized by Burawoy (1976*). Burawoy’s original contribution is double: he first extended the circulation model to cover all forms of circular migration (international especially), then illustrated his hypothesis via a comparison between Mexico-United States and South Africa. Challenging the postulate of the rational actor maximizing interests under market forces, Burawoy introduced political and structural factors. The central notion of his theory is based on the principle of geographical separation of the processes of labour force renewal (reproduction) from those of maintenance. It is the interplay of these two processes that drives the circular system. The domestic economy must continue to function, not only for subsistence, but also as a social security system for all family members, including those who have emigrated and who enter a foreign labour market with no guarantees, i.e. without social security in the event of injury, illness or unemployment. At the same time, cash needs oblige some family members to emigrate to those places where labour markets are associated with a market economy. 
16According to Burawoy’s thesis, this connection entails a twin dependency and is founded both on an economic basis and on political and legal institutions. “The twin dependency on two modes of production does not reproduce itself without recourse to noneconomic institutions”. This assertion deserves to be qualified. Indeed, the well-documented West African example shows that even after the disappearance of coercive structures (i.e. the abolition of forced labour), the system of temporary migrant labour continues to exist (Cordell, Gregory and Piche?, 1996).
17This model challenges the classic approach linking development and migration, whereby development leads to emigration by destroying pre-industrial society and releasing manpower to work in new urban labour markets (Massey, 1988), and whereby migration, considered as a method for reallocation of resources, will eventually restore the balance between origin and destination areas (Todaro, 1969). The notion of circular migration suggests that “preindustrial society, characterized by a domestic mode of production, is not destroyed, since it continues to ensure the subsistence of those who stay behind, and the ‘social’ security of those who have emigrated” (Gregory and Piché, 1983). Burawoy’s model remains pertinent today for another reason. The temporary work programmes now re-emerging in industrialized countries also involve a twin economic and institutional dependency. In fact, even if these workers receive market wages (which is not necessarily the case), the maintenance costs linked to socioeconomic integration are kept to a minimum by depriving them of citizenship rights. 
18Burawoy’s approach introduces macro-structural factors into the circular migration process. But in a more general perspective, certain authors see migration as a response to demand for labour. Saskia Sassen (1988*) formulates most explicitly the factors influencing immigrant labour demand. In her view, immigration is a primarily urban phenomenon, concerning the major urban centres of the developed world. It is Sassen who can take credit for developing the concept of the global city from where the world economy is managed and serviced. She examines the reorganization of industrial production, as reflected in the proliferation of sweatshops exploiting undocumented workers, and of industrial homework. This new economy has also produced a surge in the supply of low-wage jobs. As she states “even the most dynamic and technologically developed sectors of the economy generate jobs that can conceivably be held by unskilled foreign-language workers”. Hence, the massive arrival of immigrants from low-wage countries over the last fifteen years must be interpreted in the light of these transformations.
The gender approach
19Up to this point, the migration literature focused essentially on men. The special issue of the International Migration Review on female migration, edited by Mirjana Morokvasic (1984*) points out that migration also concerns women. Morokvasic’s paper examines the diverse trajectories of female migrants across the world and illustrates the many cases of female labour exploitation. For this author, female migration can be positive (emancipation, financial independence), but can also reinforce gender inequalities.
20Despite Morokvasic’s powerful plea, the feminist approach to migration has never fully penetrated the dominant spheres of migration research. Recent literature reviews pay scant attention to female migration (see, for example, Massey et al., 1998; Zlotnik, 2003). However, following on from Morokvasic, the role of gender relations in migration decisions has been addressed as part of a specific theory of women’s place in society. Under this theory, the mere consideration of gender as a variable among others is not enough; it must be developed as a central concept (Boyd, 1989*; Pessar, 1999; Lutz, 2010). This theoretical construct stresses the gender division of labour, which forces women to assume the majority of household tasks and places them in a subordinate position, restricting their geographical mobility in places of origin, or confining them to insecure jobs in places of destination. In short, for several authors, women’s marginal position on the labour market is a reflection of family choices which perpetuate gender inequalities (Tienda and Booth, 1991).
21Neoclassical theory was strongly criticized by the “new economics of labour migration” associated primarily with the economist Oded Stark (1991). Stark and Bloom (1985*) depart from micro-economic theories by introducing the notion of family strategy which highlights the mutual interdependence between migrants and their families, and places emphasis on risk handling and risk pooling. Migration is analysed at the household level and is seen as a form of social insurance. In short, alongside the human capital so dear to neoclassical theory, network and kinship capital also exists (social capital). This approach was later developed widely in research on migration in developing countries, particularly with regard to survival strategies and migrants’ capacity to become actors of change (De Haas, 2010). Moving away from an individualistic and atomistic vision, migration can now be conceived as the product of collective and family actions linking migrants and non-migrants in a set of relationships that are captured in new analyses centred on the notion of networks.
22The text by Monica Boyd (1989*) is a key contribution to the literature on networks and gender, introducing the idea of networks as links between places of origin and destination that mediate between individual actors (micro) and larger structural forces (macro). The family plays a central role in this system. She also focuses on the relationship between gender and networks in migration, stressing that the gendered division of labour must be included in any account of the social relations of production in a society. She thus subscribes to the distinction, established by Burawoy (1976) and Meillassoux (1975), between the public and private spheres, and above all, the necessary articulation between the two when analysing female migration strategies.
23The network approach also underpins the model of cumulative causation proposed by Douglas Massey (1990*). The network as an element of social structure is a notion used by Massey to establish a link between networks and the effect of feedback on migration. After a certain time, via a process of circular and cumulative causation, migration becomes self-perpetuating. Another important notion in Massey’s text is the distinction between the processes that initiate migration and those that maintain it. At the outset, the creation of markets in developing regions progressively disrupts traditional patterns of social and economic organization and generates conditions conducive to migration. It is only after migration has begun that “a variety of self-reinforcing mechanisms come into play that perpetuate and expand the migration flows over time, feeding back on community structures to promote its cumulative causation”. The existence of networks leads to the creation of a social capital, a notion that explains why and how belonging to a network increases the probability of migrating: thanks to network resources, the costs and risks of migration diminish while the benefits increase (Palloni et al., 2001).
24This approach centred on networks and social capital as positive factors has been criticized, by Krissman (2005) among others. In particular, he considers that the notion of migration network as developed by Massey is overly restrictive as it focuses on social and family networks primarily based in the same regions of origin as the migrants themselves. For Krissman, networks include many other stakeholders who serve as intermediaries, either at national borders, or in the regions of destination. They may be employers looking for migrant labour, but also traffickers. Hence, the actors involved in migration networks are not always facilitators; some may also be exploiters. The numerous reports on human trafficking testify to the existence of vast smuggling networks, often linked to criminal organizations (Skeldon, 2002; Bélanger, forthcoming). 
II – The effects of migration
25As mentioned earlier, the field of migration remains fragmented, with some theories explaining the reasons for migration and others explaining its effects. Research on the effects of migration is in turn fragmented, between macro and micro approaches, and between the contexts of developed and developing countries.
The economic effects of immigration: structural, macro approach in developed countries
26For Stephen Castles and Godula Kosack (1972*), immigration has become a structural necessity in response to the needs of western capitalism. Their pioneering text on the economic contribution of immigrants advanced a series of hypotheses that served as a guiding thread for subsequent research. Their approach represents a strong current in Europe, still valid today for several types of immigration, notably unskilled labour migration, irregular migration and temporary migration in sectors such as agriculture, construction, hotels and catering. But Castles and Kosack’s key contribution was to highlight the hierarchical structure of employment, with immigrants often at the bottom of the socioeconomic ladder, leading to a division of the working class.
27Following on from Castles and Kosack, Marxist-inspired studies tended to focus on the negative effects of immigration. This negative viewpoint was reiterated in research on migration and development, a point we will discuss in the next section. Recent research on the more global economic effects of immigration (at macro level) has produced results that are contradictory and uncertain to say the least (He?ran, 2002). The most problematic aspect of these studies is of a methodological nature. Most researchers acknowledge that final conclusions cannot be drawn using existing techniques of analysis. This also explains the substantial divergence between results – sometimes positive, sometimes negative, and sometimes indeterminate. And in all cases, the measured effects are very small, or even non-significant. This methodological shortcoming is linked firstly to the huge number of parameters to be included in the models which, as yet, have defeated all attempts at empirical implementation. A second problem is more fundamental: evaluation studies consider the short term, yet the full benefits become visible only in the medium and long term (Goldin et al., 2011). Unfortunately studies of the long term are rare. According to Carter and Sutch (1999), whose study covers the entire twentieth century (1820-2000), the beneficial aspects of migration are clearly discernible. Immigration can have a major impact on the entire economic structure, including labour-force participation rates, population skill levels, quantity and quality of capital, and organization of production (Carter and Sutch, 1999) Another more recent review of the effects of immigration in developed countries concludes that they are generally positive, in terms not only of growth, but also of innovation and tax revenue (Goldin et al., 2011, chapter 6).
Migration and development: the case of developing countries
28In developing countries, debate on the economic effects of migration has taken a radically different turn. Researchers no longer focus on the situation in immigration regions, but rather on the links between emigration and development in the emigration regions. This has not always been the case, as the text by Oberai and Manmohan (1980*) shows. These authors turn the problem around by considering the links between emigrants and sending regions via the notion of remittances, a key vector, in their view, of the impact of emigration in developing countries. The precise effects of these cash transfers on the rural economy is difficult to determine ex-ante. They may contribute to productive investments aiming to develop and diversify agriculture or non-farming activities in rural areas; they may be spent on housing or education or, quite simply, serve to alleviate the poverty of those left behind in the villages. Remittances may thus be used unproductively, and making them more productive is a leitmotif of future research.
29The authors mention that seasonal migrants may start sending remittances quite soon after arrival, and this idea has been taken up by several authors, including Portes (2009) who, in his research overview, concludes that temporary migration is the form of migration with the greatest positive effects. Oberai and Manmohan also mention that the relative effect of remittances is strongest for the poorest households.
30Their work has fostered a whole series of research projects that have developed in two directions. They began with a realization of the enormous volume of migrant remittances, and several researchers have sought to estimate these cash flows at global level. In 2011, for example, remittances to developing countries totalled USD 372 billion, an increase of 12.1% with respect to 2010, and with an annual growth rate of 7-8%, the total could reach USD 467 billion by 2014 (Ratha and Silwal, 2012). So all the international organizations involved in development are now convinced that migrants can become agents of development (Faist, 2008). The second direction of research, linked to the first, has spawned numerous studies on transnationalism (Vertovec, 2009). Here, migration is no longer seen in terms of permanent rupture, and attention focuses rather on the links that are maintained between the home society and the host society, since migrants’ lives cut across national boundaries and bring two societies into a single social field.
31Transnationalism often conveys a positive image of migration, and this image has been embraced and disseminated by international organizations including the World Bank, the International Organization for Migration, and the various instances of the United Nations. However, many critics have sought to temper this enthusiasm for the developmentalist capacities of remittances and transnationalism. In particular, literature reviews on the impact of remittances suggest that situations are very heterogeneous and that cash transfers alone are not sufficient to boost the economic development of a region (or country) if there are no genuine investment opportunities in the localities where the beneficiary households live (Skeldon, 2008 and De Haas, 2010). In short, if households cannot surmount the structural obstacles to development (access to credit, trust in institutions, migration policies favourable to investment, etc.), remittances will not significantly accelerate local or national development.
The micro-economic effects of immigration
32There are two dimensions to the effects of migration at micro-individual level. The first asks the question: is the migration experience positive for migrant men and women? The second concerns the impact of migration on non-migrant populations and on the native-born inhabitants of destination countries. Curiously, the first dimension has rarely been studied in developed countries, as if the positive effects of migration at individual level could be taken for granted. It is as if the microeconomic hypothesis of the rational individual maximizing his interests has become a postulate that needs no verification. The second dimension, on the other hand, has become a central issue, focusing on three questions: How does immigration affect the earnings and labour market opportunities of native residents? Do immigrants really have a negative effect on these opportunities? Last, are all groups of native residents affected in the same way by immigrants’ arrival on the labour market?
33George Borjas (1990*) was a key figure of research in this area. For Borjas, there are two opposing views about how immigration affects the native labour market. Some observers claim that immigrants take the natives’ jobs, while others argue the opposite, asserting that immigrants have no impact on labour market opportunities for natives. Borjas tends to favour the second view. His most important conclusion is that the methodological arsenal of modern econometrics is incapable of providing a single shred of evidence to prove that immigrants have a substantial and adverse impact on the earnings and labour market opportunities of US natives. Why is this so? Because native workers and immigrants are not interchangeable in the production process. Borjas also shows that if the arrival of new immigrants on the labour market does have an adverse effect, it is on the pre-existing immigrant population, and on the least skilled native workers. But in all cases, these effects are small, or even negligible (Card, 2009).
34One of the conceptual weaknesses of research on migrant economic integration, in both developed and developing countries, is to consider the labour market as unique. Alejandro Portes and his team have shown that migrants may be incorporated into the labour market in several different ways (Wilson and Portes, 1980*). Drawing on the theory of labour market segmentation as developed by Michael Piore (1979), the authors describe three modes of incorporation. The first two concern the primary and secondary labour markets. The primary labour market corresponds to professionals and skilled workers, often unionized and with real opportunities for advancement. This sector is also characterized by stability, promotional ladders, high wages and good working conditions. The secondary sector, on the other hand, is the reverse image of the primary sector, and is characterized by low wages, menial and insecure jobs, and low levels of unionization. It is in this second sector that many immigrants find employment.
35But Wilson and Portes’ most original idea concerns the third mode of incorporation, the “ethnic enclave” which comprises groups of immigrants concentrated in a specific geographical area who set up businesses to serve their own ethnic market and/or the general population (Portes, 1981). The main feature of enclaves is their high proportion of immigrant workers working for businesses belonging to other immigrants (Light, 1972). This third form of labour market incorporation introduces the idea that unqualified immigrants are not all at the bottom of the socioeconomic ladder, and that working in an immigrant enclave may be advantageous as it offers real opportunities for advancement. Studies based on this mode of incorporation, most of which are American, focus primarily on Asian immigrants (Japanese, Koreans), but also on Cubans in Miami.
36More recent research has sought to challenge the ethnic enclave approach. In his critique, Waldinger (1993) concludes that the notion of enclave leads to a conceptual and empirical impasse, and suggests dropping the concept and using only that of the ethnic economy. The debate arising from Portes’ work has focused mainly on the positive and negative effects of enclaves in terms of migrant economic integration. Taking an opposite stand to the proponents of this approach, several scholars have questioned the assumption that ethnic enclaves are advantageous for immigrants (Sanders and Nee, 1992). In a paper published in 2006, Portes and Shafer responded to these criticisms by concluding that the ethnic enclave approach was still valid.
Political effects: the case of refuge migration
37The political effects of immigration are addressed from two angles. First, political factors have been studied mainly in cases of refugee movements; second, the effect of migration on diversity, in terms of the relations between minorities and majorities, is a key political issue at the heart of the identity debate. Most of the texts presented so far focus on legal and voluntary migration. The text by Zolberg, Suhrke and Aguayo (1986*) looks at refugee migration, a form of migration that was very common throughout the twentieth century and remains so today.
38In their discussion of the factors of refugee migration, Zolberg, Suhrke and Aguayo make an important distinction between internal and external effects. While the determinants of persecution, based on the Geneva Convention’s definition of refugees, are internal to the country in question, external effects may also exist in the form of factors that aggravate economic and social conditions, thereby increasing the likelihood of conflicts that generate refugee movements. Indeed, the policies implemented by potential host countries represent the most problematic type of external effects. The decision to grant formal refugee status to citizens of a particular country is generally seen as an implicit condemnation of the government of that country for persecuting its citizens or failing to protect them.
39The authors also discuss refuge migration in the global context, with its enormous asymmetries in terms of power and wealth across different regions of the world. Hampered by structural distortions stemming from their incorporation into the global economic system, developing countries participate in the world economy on disadvantageous terms, and this exacerbates conflict of all kinds, notably ethnic conflicts which, according to the authors, are endemic in Asia and Africa today. In short, they show that the dynamics leading to the inception of social conflicts are not purely internal, but transnational, and that as conflicts develop, they tend to be further internationalized. In conclusion, to the extent that the causes are international, the solutions also call for action at the international level.
40These ideas have not been widely developed since the work of Zolberg, Suhrke and Ahuayo in the 1980s. The refugee question is studied rather with regard to the application of the Geneva Convention. Some recent studies have shown, for example, that European countries are seeking to restrict or even abolish practical access to political asylum in Europe (Legoux, 2006).
The social effects of migration: minorities versus majorities
41The growing diversity of societies is a major consequence of migration and raises key challenges for managing social, racial and ethnic differences. On this question, Castles (1993*) advanced a series of hypotheses on the situation of migrants and minorities in western Europe, notably with respect to migration policies and the issues of citizenship, racism and identity. While we cannot go into all of his hypotheses here, it is important to highlight several aspects which are still relevant today, notably regarding the effects of immigration on ethnic and race relations.
42Castles addresses the question of racism in Europe and the danger that “this ‘European consciousness’ will be constructed in exclusionary and discriminatory terms, based upon the perceived threat of being swamped by ‘desperate masses’ from the south”. In his view, “the constitution of new minorities with distinct cultures, identities and institutions, is an irreversible process, which questions existing notions of national identity and citizenship”. For Castles, “multicultural models appear to offer the best solution, but there are substantial obstacles to their realization”. The long-term consequence of immigration will be “the emergence of multicultural societies, leading in turn to new concepts of citizenship and the nation state”.
43Castles’ hypotheses, formulated in the European context, are still very relevant to today’s world. They have spawned two major research currents. First, practically all studies of the factors of economic integration suggest that discrimination is an important factor in the difficulties encountered by certain immigrant groups, notably the so-called “visible minorities”, to use the terminology in vogue in North America (e.g. Piche?, Renaud, Gingras, 2002; Richard, 2004). The second current concerns the effects of immigration on national identities. Here too, there is heated social and political debate between the advocates of pluralism and those who believe that immigration threatens national values. The discourse of the extreme far-right political parties emerging in many parts of the world is based, among other things, upon an anti-immigration stance, sometimes linked to Islamophobia.
III – Migration policies
44Migration theories do not serve solely to understand and explain, but also to justify migration policy choices. All twentieth-century migration policies were founded on a supposedly unshakeable principle: immigration is a privilege and not a right. Under this paradigm – based, moreover, on the principle of national sovereignty in migration policy – these policies are shaped by the economic needs of the countries concerned, and hence focus mainly on the labour market. In this context, the free circulation of persons as advocated by Joseph Carens as early as 1987* may appear to be an unrealistic aspiration.
45Carens starts out from the principle that birthplace and parentage are natural contingencies that are arbitrary from a moral point of view. In his view, the idea that immigration reduces the economic well-being of current citizens is not a valid argument for imposing restrictions. Moreover, “the effect of immigration on the particular culture and history of the society would not be a relevant moral consideration, so long as there was no threat to basic liberal democratic values”. His discussion of the communitarian principle, whereby exclusion is justified by the right of communities to self-determination, brings him to the question: “If freedom of movement within the state is so important that it overrides the claims of local political communities, on what grounds can we restrict freedom of movement across states?” He goes even further, stating a principle also advanced today in the case of temporary migrant workers: “It is right to assert that our society ought to admit guest workers to full citizenship. Anything else is incompatible with our liberal democratic principles”. For Carens, free migration may not be immediately achievable, but it is a goal towards which we should strive. It was not until the late 2000s that this approach resurfaced in a systematic manner, and the recent book by Pécoud and Guchteneire (2009) has shaken many certainties as to the immutability of the utilitarian principle.
46Without going as far as to advocate free movement, which he judges impracticable, Bimal Ghosh (2000*) was among the first to develop a systematic global approach to the migration question. His text highlights some of the inadequacies of existing migration policies and practices, and argues in favour of a “more comprehensive, balanced and transparent multilateral regime to manage migration”. He points up the absence of appropriate policies to address the new challenges of migration, notably the build-up of emigration pressure due to mounting inequalities.
47Rather than addressing the root causes that generate or fuel pressures for disorderly migration in sending countries or that attract migrants to the receiving ones through irregular channels, governments have responded by adopting repressive and restrictive measures. Ghosh’s main thesis is that migration is getting out of control, and that the magnitude of irregular migration across the world reflects the inadequacies of the present migration system.
48Ghosh’s approach has given rise to a new paradigm, that of “migration management”, to achieve more orderly, predictable and humane objectives through global governance of the migration regime. It is founded on the basic premise that well-managed migration can be positive for everyone: the countries of origin and destination, and the migrants themselves (hence the famous slogan “win-win-win”). Several critical analyses see migration management as a new paradigm that seeks to disseminate a global hegemonic approach whereby migration is a normal characteristic of today’s globalized world (Geiger and Pe?coud, 2012). In practice, migration management involves a series of measures for more effective policing of borders, including measures to intercept migrants before they reach their destination.
49It is the fear of huge waves of illegal immigration that shapes some areas of current political discourse and that serves to justify restrictive measures. Unfortunately, scientific research on this type of migration is scarce. The text by Georges Tapinos (2000*) is an exception, and his study of irregular migration provides a useful complement to the analyses of Bimal Ghosh. For Tapinos, examining the economic and political challenges of irregular migration involves identifying what makes this type of migration specific with respect to regular migration. Beyond measurement problems, debate often focuses on the economic impact of irregular migration. For Tapinos, it is employers who benefit the most. The precarious situation of irregular migrants, and their limited bargaining power, foster discriminatory practices.
50In his view, migrants in illegal employment are one component of the underground economy, but they are not its cause. However, the existence of an underground economy makes it easier to recruit illegal migrants, helped by migrant networks who smooth their entry into the informal labour market. Migration management calls for coordination between governments. For Tapinos, this must be multilateral management; it would be anachronistic to address immigration control exclusively in terms of sovereignty.
IV – Towards a diversified approach
51The period 1960-1980 has often been seen as a time of confrontation between competing migration theories, each claiming supremacy. Micro-individual theories focusing on economic rationality and the notion of equilibrium were pitted against macro-structural theories centred on demand for migrant labour generated by developments in the world capitalist economy (Wood, 1982). From the 1990s, however, these theoretical debates were rendered obsolete by empirical studies which showed that each theory explains a part of the migration process, some being applicable more specifically to a particular region or historical period.
52We have presented the various theoretical approaches by examining the paths along which new ideas unfold. As stated above, rather than competing, each approach brings specific new insights that must be taken into account when attempting to explain migration. They form the pieces of a puzzle, brought together in Figure 1. This figure represents an analytical framework which sees migration as a multifactorial and multidimensional phenomenon and which incorporates its three main ingredients: origin and destination; micro, meso, macro and global analysis levels; and economic, social and political dimensions (Piche?, 2004). In methodological terms, we see that time is placed at the centre of the model. The macro effects of migration only become visible over the long term (historical timescale) while integration in a new society depends on the duration of stay (biographical timescale).
Conceptual framework for studying factors of migration and integration
Conceptual framework for studying factors of migration and integration
53The most recent founding text presented here dates back to the year 2000. Does this signal a recent absence of major theoretical developments in migration studies? Based on the literature reviews published since 2000, one might be tempted to say yes. Looking, for example, at the book edited by Alejandro Portes and Josh DeWind in 2007, or that of Corrado Bonifazi, Marek Okolski, Jeannette Schoorl and Patrick Simon in 2008, we see that the topics covered are not fundamentally different from those discussed in the founding texts. A highly developed theoretical corpus is available to scholars, which must now be developed and applied in specific historical contexts.
54Two dimensions of migration deserve to be more fully theorized in the future. The first concerns the emergence of the new paradigm discussed above, namely the globalization of migration flows, which is transforming the role of international migration in today’s societies (Kabbanji, 2011). Although a few papers have addressed this question, the major challenge of explaining current trends has yet to be taken up. Two questions, in particular, must be studied in greater depth: What are the new manpower needs of the developed economies, and how will they evolve? How will national and supranational states respond to these new needs? For now, governments are tending to restrict permanent migration in favour of circular and temporary migration, resulting in the emergence of new categories of non-citizens. This new paradigm involves “replacing the concept of migration with that of mobility”, the latter being the most advantageous situation for optimizing profit (Pellerin, 2011). Is there a fundamental contradiction here between neo-liberalism, which promotes free circulation of capital, goods and services, and the new protectionist model of migration management based on flexibility and circularity? And above all, for how long will this contradiction persist?
55The second dimension to be more fully integrated in migration theory concerns migrants’ rights. This is a paradigm that extends beyond the purely utilitarian outlook, by proposing that questions of migrants’ rights be included among the political parameters of migration. Research on this question remains overly ideological, and does not focus adequately on practical analysis of the living conditions of migrants and their families. What matters here, is to introduce the question of rights as an integral component of migration policy (Piché, 2009).
The 20 texts brought together in the collective work published by INED are identified in the text and the bibliography by an asterisk (*). Piché V., Les Théories de la migration, Paris, INED, les Manuels / Textes fondamentaux, 536 p., available at www.ined.fr/en
For an exception in the field of fertility and health, see Riley and McCarthy, 2003.
It is interesting to note that at around the same date, in fact a year before Burawoy, Claude Meillassoux (1975), not cited by the author, had proposed the same approach for the African context, articulating domestic and capitalist modes of production, and the separation of the two functions (reproduction and maintenance of the labour force). This approach inspired several later studies on both internal and international African migration (Gregory and Piche?, 1985).
The re-emergence of guest worker programmes is currently receiving strong support, not only from international organizations such as the International Labour Organization and the International Organization for Migration, but also from scholars (Piche?, 2012).
For an overview of the state of human trafficking across the world, see Laczko and Gozdziak, 2005.