New Migration Dynamics on the South-Western Periphery of Europe: Theoretical Reflections on the Portuguese Case

I reflect on the overall contents of the book and highlight the key theoretical insights contributed therein. I draw on my own theoretical writings on the nature of migration trends in Southern Europe – the so-called ‘Southern European model’ of migration, with its various historical and geographical phases – and enrich this analysis through specific reference to the Portuguese case. Notions of cores and (semi-)peripheries are crucial to understanding both the political economy and evolving typologies of Portuguese migration trends over the long term and in recent years. Key here is the coexistence of Portugal’s status as the metropolitan core of a (post)colonial periphery and its (semi-)peripheral situation within Europe. This dual perspective helps to explain the complex geography and history of Portugal’s migration flows. Finally, I point to four areas which could have been given more emphasis: the within-Portugal regional patterning of emigration trends, the relative lack of attention given to return migration, the question of gender in ongoing Portuguese migration patterns and the overall impact of migration on Portuguese economic and social development.


Introduction
Grateful to the editors for the invitation to write this concluding chapter, I first want to make it clear that my perspective in doing so is that of a partial outsider to the rich and vibrant field of Portuguese migration studies. Whether this yields any advantage in my reading of the dynamics of Portuguese migration, it is for the reader to decide. Optimistically, my semi-outsider status, bringing in a range of comparative and theoretical perspectives, offers a useful counterpoint in a book pretty much entirely produced by Portuguese authors working within Portuguese universities and research institutes. I also position myself as a partial outsider to the majority authorship of this book, most of whom are sociologists or have received a predominantly sociological training. As a geographer, I bring to the table a broader comparative view of the migratory phenomenon: comparison not just regionally and between countries but also in terms of diverse methodologies and scales of analysis. 1 This is not to denigrate the importance of sociology or any other discipline, absolutely not, and in fact sociology has been at the forefront of the development of the interdisciplinary field of migration studies; but I do believe that the perspective of the geographer has been under-valued within the social sciences generally, even if there are signs that this is now changing (see, for instance, Favell 2008, 262).
Having set out my stall, the chapter continues as follows. The next section delves into the core-periphery model, a concept which is at once beguilingly simple and also complex when interrogated in diverse and changing historical and geographical contexts. Core-periphery structures are used to elaborate Portugal's somewhat ambiguous position in what I have previously and perhaps contentiously labelled as the 'Southern European migration model' (King 2000). My conclusion, briefly, is that Portugal's position within this model is sustained, albeit with more exceptions than the other Southern EU countries. The key difference is that Portugal is also the hub of its own 'Lusophone' migration system, which gives it a 'semi-peripheral' position within global migration patterns. The subsequent section shadows the structure of the book. Here I attempt to draw out key theoretical insights and findings from the chapters in the various sections of the book -on 'old' and 'new' Portuguese emigrants, high-skilled and labour migrants, and questions of migratory identity. The final part of the chapter nominates four areas which, in an ideal world (and a much longer book!) could have been accorded more emphasis -the spatial patterning of Portuguese migration origins, the issue of return migration, the gender perspective and the relationship between migration and development.

On Cores and (Semi-)Peripheries: A Portuguese Perspective on the Southern European Migration Model
Like the rest of Southern Europe (King 1982), Portugal illustrates the tension between dependency on the one hand and development on the other, such that its style of economic and social progress has been labelled 'dependent development' (Holland 1979). Dependency theory has its roots in the Latin American dependencia school of, amongst others, André Gunder Frank (1967) and its re-application via the core-periphery model to the global scale by Wallerstein (1974) in his historical world-systems model of capitalist (under)development, in turn re-applied to the European scenario by Seers et al. (1979). Seers et al. introduced an explicit geographic analysis, mapping the core-periphery system in a series of maps and diagrams (see, especially, Seers 1979, 4-5, 17, 21) which distinguished flows of migrants, trade and tourists between European countries classified as 'core' (France, West Germany, Belgium, the Netherlands, Luxembourg, Switzerland, Austria, Sweden, Norway, Denmark), 'periphery' (Ireland, Portugal, Spain, Greece, Yugoslavia, Turkey) and 'semi-periphery' (Finland, Italy and -controversially -the UK). This classification is far from water-tight, and the danger of 'fixing' it in cartographic form is that the static map obscures its dynamic nature. Particularly fluid and ambiguous is the category of semi-periphery -a point I shall return to presently with reference to Portugal and this country's complex migration dynamics. At the root of the core-periphery model is the process of uneven spatial development which, many Marxist-inspired scholars argue, is intrinsic to the capitalist system. Spatial polarisation in and around the core is driven by economies of scale and agglomeration; this process of cumulative causation generally outweighs overspill effects to the periphery. Selwyn (1979, 37) stresses that 'core-periphery systems, whilst not unchanging, are very persistent', due to the fact that the resultant spatial inequalities are underpinned by 'very deep rooted economic, social and political structures'.
As Seers (1979) and others have more recently emphasised (eg. King 2015 and, in this volume, Pereira and Azevedo, Chap. 1), migration is perhaps the most important of the flows linking peripheral to core countries in an asymmetric relation of power which ranges across political economy, economic geography and social inequality. Moreover, migrant labour has a cyclical function, flowing into the core countries at a time of economic expansion, and blocked and 'returned' during a recession. Thereby, economic fluctuations in the core are transmitted to the periphery, which acts as a disadvantaged shock absorber protecting the core (Selwyn 1979, 38). However, this is only one of a range of migration mechanisms which can be interpreted through the core-periphery lens, as can be seen by the case of Portugal.
In its simplest form, and generalisable (with partial exceptions) across all four main Southern European EU countries (Spain, Portugal, Italy and Greece), this migration-centred model has moved through several distinct phases over the past century and more. 2 Stage 1 is the transatlantic cycle of emigration from Southern Europe which started in the mid-to-late nineteenth century, in the footsteps of, and overlapping with, major emigration flows from Northern European countries. Whereas the Northern Europeans settled mainly in North America, Southern Europeans' destinations were in both North and South America (except for Greeks who mainly went to the United States). For the Portuguese, the key transatlantic destinations were Brazil, the US and Canada. Across the four Southern European countries, the transatlantic cycle reached its apogée in the decades spanning the turn of the century. It then faded into pulses interrupted, in chronological order, by World War I, the Great Recession and World War II, with a final phase during the 1950s and 1960s. Its partial reappearance in very recent years, as part of the so-called 'fourth wave' of emigration, will be commented on below.
Although these nineteenth-and early-twentieth-century migrations were largely born out of poverty and unemployment -so a typically global-scale periphery-tocore migration -for the Portuguese there was also colonial settlement migration, especially in Brazil, with professional involvement in farming, business development and administration. Sapelli (1995, 31-32) wrote that these early Southern European migrations were historically significant in that they created a 'culture of emigration' which endures to the present day.
Stage 2 is the postwar phase of European labour migration, concentrated in a much shorter time-frame of circa 1950-1973. As portrayed by Seers (1979, 4, 6, 23-24), this was the classic periphery-core spatial model functioning across a South-North divide within Europe. These were 'Fordist' migrations of an era of marked industrial expansion in the core economies of Europe, whose factories and construction sites became reliant on labour sourced not only from the 'near periphery' of Southern Europe but also, in the cases of the UK, France and the Netherlands, from an 'outer periphery' of colonial and former colonial territories. France and West Germany were the main magnets drawing in supplies of South European migrant labour: in the case of France mainly from Portugal, Spain and Italy (as well as Algeria, Morocco and Tunisia); in the case of West Germany from Italy, Greece, Yugoslavia and Turkey. From a sending-country perspective, Portugal had a highly diversified set of destinations, especially at the start of this period (USA, Canada, Brazil, Venezuela, Angola, Mozambique) whilst, within Europe, France was the main receiver of Portuguese emigrants, followed by West Germany and Luxembourg.
According to figures for this period culled from the OECD's migration observatory, which I have analysed in detail elsewhere (King 1984(King , 1993, Portugal's peak emigration phase was 1969-1971, when over 150,000 migrated per year to other European countries. In 1973, when these labour migrations were brought to a close by the oil crisis, the 'stock' of Portuguese workers living abroad in Europe was 469,000 (compared to 858,000 Italians, 582,000 Turks, 535,000 Yugoslavs, 527,000 Spaniards and 332,000 Greeks). However, unlike emigration from most of these other countries, which was shaped by bilateral recruitment agreements, most Portuguese emigration to Europe, especially to France, was clandestine, due to the restrictions on 'free' emigration imposed by the fascist regime then in power. The demise of that regime in 1974 coincided with the end of the Fordist era of intra-European mass labour migration.
For the remainder of the 1970s, throughout the 1980s and into the 1990s, we see a much more diversified set of migration processes in stage 3 of the 'Southern European migration model'. Return migration was initially the dominant flow, driven by a variety of factors: recession and unemployment in the North European host countries, the prospect of a better future in the now-developing economies of Southern Europe, and the more cultural, emotional and life-stage factors which are also an intrinsic part of return decision-making. In the case of Portugal, returns from recession-bound Europe were massively outnumbered by the estimated half-million retornados who repatriated from the newly independent colonies in the mid-1970s. The Portuguese migration profile also differed from that of the other three Southern European countries at this time in one other important respect -emigration continued at quite a high level. This fact reflected, on the one hand, the ongoing strength of the Portuguese culture of emigration and its diversified array of destination countries and, on the other, the country's status as a weak member of the European economic periphery, lagging behind the progress achieved by Spain, Italy and even Greece at this juncture. Switzerland became an important destination for Portuguese emigration, in addition to the former key European targets of France, Germany and Luxembourg.
Finally, in this complex interplay of return-emigration-immigration which characterises the Portuguese 'third wave', we see the relatively new phenomenon of immigrants arriving in Portugal -from Brazil, the former African colonies and, after 1990, some East European countries where new emigration possibilities suddenly opened up after the collapse of the communist regimes. The scale of this immigration, however, has been lower than in Spain, Italy and Greece. Apart from the economic factor mentioned above, geography also plays an explanatory role in this differentiation. Located at the south-western periphery of Europe, bordering only one other country and distant from the 'new' migration access routes across the Mediterranean (to southern Spain and Sicily) and from neighbouring Eastern European countries (to Italy and Greece), Portugal was geographically 'out on a limb' in this new era of immigration to Southern Europe.
The suddenness, scale and epoch-making nature of the Southern European 'migration turnaround' to net immigration in the 1980s and 1990s (see King et al. 1997) gives support to reclassifying the Southern European countries as 'semiperiphery' since they are no longer confined purely to the status of suppliers of migrant labour to the 'core' countries of Europe. This is not the only rationale for viewing Southern Europe, including Portugal, as semi-periphery, as we shall see shortly. Before that, let us record the occurrence of stage 4 of the Southern European migration model, which refers essentially to the last decade or so.
Although the Eurozone financial crisis of the late 2000s was the defining moment of the 'fourth wave', Pereira and Azevedo argue in Chap. 1 that the fourth cycle of Portuguese emigration started around 2001, with the stagnation of the economy and rising unemployment, as well as the currency switch from the escudo to the euro in Portugal. I would concur with this, also on the basis of evidence from the other Southern EU countries where, well before the financial crisis, one could observe a more deep-seated structural crisis of youth unemployment and lack of hope for the future, especially amongst unemployed graduates (see, for example, Conti and King 2015). Meanwhile, in Portugal, according to figures presented by Pereira and Azevedo, annual emigration rose from around 50,000-60,000 in the early 2000s to reach 110,000-120,000 in the mid-2010s -rates comparable to those of the 'emigration boom' in the late 1960s.

Portugal as a Migration Semi-periphery
Although the basis of my argument in the foregoing section has been that Portuguese migration dynamics can be best understood within the combined framework of the core-periphery model and the four-stage Southern European migration regime, there are certain unique features which characterise the Portuguese case. These specifically Portuguese features lead to the classification of Portugal as semiperiphery on the following grounds.
First and foremost, Portugal has historically had a Janus-faced position in the global hierarchy of metropolitan cores and dependent peripheries. Less than half a century ago, so within the living memory of many older Portuguese, the country was the metropolitan core with its own colonial periphery. First Brazil and then Goa and the African colonies were lost and, with the revolution of April 1974, the axis of Portuguese international relations was inverted. In Stuart Holland's words, 'from being the nominal centre to a major African periphery, Portugal became a minor nation state on the periphery of Europe to which, by geography and culture it had always belonged ' (1979, 139). Holland's somewhat glib dismissal of the Portuguese colonial era as an aberration from the country's 'true place' on the periphery of Europe plays down the impact of Portugal's long period of colonialism on the country's kaleidoscope of migration flows, both 'out' and 'in', over the four waves portrayed above and highlighted in several chapters in this book.
Still today, Portugal looks both to Europe, especially since its membership of the EU from 1986, and outwards to Latin America and Africa; it is not only European but also Atlantic and, arguably, Mediterranean in its vision and identity. This reinforces its multi-faceted and intermediate positionality within the core-periphery system, and is reflected in the way it continues to send various types of migrant to places as differentiated as the UK, Switzerland, Brazil and Angola, and receives migrants from the Portuguese-speaking world as students, professionals and labourers.
In their landmark paper arguing for Portugal to be classified as a semi-peripheral country in the global migration system, Góis and Marques (2009) draw attention to the country's integration, as both a sender and receiver of migrants, in several subsystems of regional and global significance. 3 However, key to their argument is the identification of a 'Lusophone migration system'. Especially over the past two to three decades, new flows, moving along pathways of an older colonial system based on a common language, have bonded the Portuguese-speaking countries together, with Portugal at the core of this system.
Following Arrighi (1985), Góis and Marques (2009, 27) maintain -and I agreethat the semi-periphery is neither a new nor a transitional or residual category within the core-periphery model but, rather, a distinct and permanent feature of the world system. Portugal, with its unique location on the south-western periphery of Europe, its history as a colonial metropole and its heterogenous, shifting and mixed migration flows, rather well exemplifies this theoretical point about the validity of the notion of semi-periphery.
Four other migratory features confirm Portugal's status as a semi-peripheral country. The first is its experience of immigration flows from other (semi-) peripheral countries. The largest of these inflows are from the Eastern European peripherythat which lies within the EU (Poland, Romania etc.) -and from a non-EU 'outer' periphery (Ukraine, Moldova, Georgia etc.). Additionally, in Chap. 8, by Queirós, we learn of the migration of Portuguese construction workers to neighbouring north-west Spain, which can be seen as an interesting case of migration to a bigger, contiguous neighbour within the semi-periphery.
The second 'semi-peripheral' feature is the phenomenon of replacement migration. Góis and Marques (2009, 41, 43) document the close link between Portuguese migration to Switzerland and Cape Verdean migration to Portugal, often involving workers engaged in the same economic sector. The interdependence or 'structural coupling' of these two migrations occurs because of the membership of Portugal and Switzerland in a common labour circulation area, whilst Cape Verdean migration to Portugal continues the colonial relationship.
A third type of migration, which Portugal has pioneered in Europe, is that of 'posted workers'. Quite different from the well-researched history of Portuguese labour migration to France or North America, this arises due to the way that Portuguese companies act as labour subcontractors for big European construction companies, using the advantages of lower Portuguese wages and willingly mobile workers to 'post' contingents of labourers to construction sites within the European free movement area (Marques and Góis 2017, 74-75). Over the years 2007-2011 Portuguese posted workers were the largest nationality within the EU, averaging over 60,000 per year, mainly to France, Germany, the Netherlands and Spain.
Finally, we can observe how Portugal, like a number of other Southern European countries, increasingly functions as a transit country for migrants who move there from countries of the less-developed world, and use it as a short-or long-term stopover before moving northwards to more prosperous Northern European countries. In the Portuguese case, migrants of this kind typically originate from Brazil, China, South Asia and the Portuguese African ex-colonies. Whilst Góis and Marques (2009) describe this 'redistributive' function of Portugal in general terms, a recent paper by McGarrigle and Ascensão (2018) documents the specific case of Punjabi Sikhs who use Lisbon as a 'translocality' to stay for a while during their long-term migratory journeys to other European countries such as Germany and the UK.

Highlights of the Book
In this section of the chapter I review the contents of the book and pick out some of the stand-out features of the chapters, especially from the point of view of their theoretical insight and empirical originality. This is an inherently subjective exercise and no doubt my biases will be evident.
At a general level, the book functions as an excellent state-of-the-art on Portuguese emigration. It brings together for an English-language readership much of the best recent Portuguese scholarship on migration, packaged neatly within five themes: new patterns, skilled migration, labour migration, postcolonial continuities, and identities. The various chapters exemplify a range of methodologies, appropriate for a topic -migration -that has many facets and is therefore best studied via 'mixed methods' (Vargas-Silva 2012; Zapata Barrero and Yalaz 2018). Some chapters are desk studies, others present their own quantitative survey data, some rely on in-depth qualitative interviews and yet others personify a more ethnographic, humanistic or even autobiographical approach.
Part I of the book -New Patterns of Portuguese Emigration -sets out the broad theoretical, historical and statistical context. In their opening chapter, Pereira and Azevedo help us to understand the key features of recent Portuguese emigration, its structural factors, the characteristics of its diverse flows and bipolar destinations -'North' to 'old' (France) and 'new' (UK) Europe, and 'South' along routes first mapped out by the history of Portuguese colonialism (Brazil, Angola). This historically multi-layered diaspora results in the Portuguese being the most numerous immigrant (foreign-born) group in Brazil and Luxembourg, the second largest in Switzerland and the third largest in France. In terms of structural factors, the authors unpick the ultimately devastating impact of the 2008 financial crisis when, in the following years, Portugal and other peripheral European countries (Spain, Greece, Ireland, Latvia and Lithuania) entered a period of austerity and economic entrenchment which resulted in very high rates of youth unemployment and hence enhanced emigration potential. But behind this severe conjunctural crisis lay another one, more deep-seated -the blocked career, and life, aspirations of a whole generation of young people, especially those with higher education whose investment in long years of study had not led to employment opportunities in their own countries consonant with their aspirations.
Pereira and Azevedo broadly accept the notion of Portugal's location within the 'Southern European migration model' (King 2000(King , 2015 but, along with the succeeding chapter by Pires, they also challenge it by pointing out some 'deviant' features of the Portuguese case. Three points seem key here. First, the immigration phase of this historical model (1980s-2000s) was less marked in Portugal. Second, emigration from Portugal continued at a high rate, even during the immigration phase. And third, whilst most Portuguese emigration has continued to be directed northwards to Europe, an important channel opened up towards the global South, especially the Portuguese-speaking countries therein. On the basis of migration flow data on immigration, emigration and returns for the years 2013-2015, Pires contends in Chap. 2 that Portugal is more akin to Eastern EU countries such as Lithuania, Bulgaria and Romania. This is an intriguing indication of the 'uniqueness' of Portugal as a 'Western' and/or 'Southern' European country, but this comparison is based on recent and short-term data and may have only temporary validity. 4 Pires' contribution is otherwise notable for its forensic statistical detail on Portuguese emigration and an ability to place these trends in a European and global comparative perspective, something that few authors have done. What is most portentious in his chapter, however, is contained in the final few paragraphs. The combination of continued high emigration (in the absence of significant counterbalancing of immigration and return migration) with a fast-declining birth rate (which has halved since 1980) have pitched the overall Portuguese population into decline -a loss of 215,000 people over the period 2010-2015. With such a constellation of negative economic and demographic variables -high (youth) unemployment, a shrinking working-age and employed population, a stagnant GDP, a total fertility rate at 1.3 (cf. the replacement level of 2.1 children per woman), a rapidly ageing population and the migration trends already mentioned, the overall future for Portugal suddenly appears extremely uncertain. In Chap. 3 by Peixoto and 11 others, attention shifts to the transnational perspective, including return migration. Deriving from the project intriguingly titled 'Back to the Future', the authors' analysis focuses on the extent and characteristics of recent Portuguese migration. Through the now-popular transnational lens, the chapter seeks to understand the relationships that emigrants retain and develop with their home country. In the latter part of the chapter, the authors' particular emphasis on return migration is a much-needed contribution given that, as is pointed out, only 8% of the Portuguese migration literature concerns return migration. The coverage here is based on analysis of two questions administered in the 2011 Portuguese Census about prior residence. Some useful new insights are revealed -to cite just one example, recent returnees are younger than in the past -but, as a desk study, this obviously misses the depth of primary qualitative research.
Part II of the book contains a clutch of chapters which are at the heart of the volume's main message -the recent switch to a more highly skilled emigrant profile. Setting aside for the moment the inherently problematic task of defining 'skill' in a migratory (or any other) context, and acknowledging once again the subjectivity of my own perspective, two chapters stood out for me by virtue of their originality. Pereira (Chap. 5) presents much-needed primary research on the emigration of Portuguese nurses, mainly to the UK. This is a gender-specific skilled-migration niche which has its own particular structuring factors, namely the high quality of Portuguese nursing training, and the switch in the UK's nurse recruitment from traditional overseas sources such as India and the Philippines to 'free-movement' EU countries. But, as the author convincingly shows, this specific type of skilled migration is simultaneously reproducing existing socio-economic inequalities and developing new professional and geographical disparities. As with the global migration of care and health workers, this is a periphery to core movement in which, to put it in the simplest of terms, 'Portugal trains nurses, and Britain takes advantage'.
Delicado, in Chap. 7, uses the theoretical standpoint of 'Portugal as semiperiphery' to frame her analysis of the emigration of Portuguese scientists. National policies to encourage the training of scientists abroad, combined with intra-EU scientific mobility schemes, led to a high rate of outward movement. But, at the same time, as Delicado stresses, the Portuguese scientific system, perhaps more so than its Southern European neighbours, worked hard to attract foreign scientists to Portugal. In more recent years, the economic crisis and resulting austerity measures led to cuts in funding for higher education and research, so that a net brain drain has re-emerged as the dominant trend. An interesting question concerns the origin of scientists who are migrating to Portugal. The answer is, mainly from other (semi-)peripheries: Brazil in Latin America; Spain and Italy within Southern Europe; several Eastern European countries (eg. Romania, Bulgaria, Russia, Ukraine); and Asia, mainly India and China. Despite certain attraction factors (climate, culture, low cost of living by European standards), few researchers move to Portugal from the Northern 'core' countries. One can surmise that low stipends are the main reason, since this is a highly cited reason for the non-return of Portuguese scientists working/studying abroad (Gomes 2015).
Part III is on lower-skilled labour migration: two interesting chapters, both of which problematise, in different ways, conventional categorisations of forms of migration and their longer-term outcomes. Chapter 8, by Queirós, was mentioned briefly above. Here I want to spell out more clearly what I found fascinating about this multi-sited ethnography of Portuguese construction workers who have moved sequentially across borders (Portuguese-Spanish, then Spanish-French) as they navigated the years of the Portuguese and Spanish economic crises.
Queirós follows the inspiring statements of Marcus (1995) on multi-sited research and Fitzgerald (2006) about ethnography with migrants. Based on the 'transregional' cross-border space of north-west Portugal and the adjacent region of Galicia in Spain, he traces the evolution of an occupationally driven migration system of Portuguese builders and construction workers up to and through the crisis years of 2008-2010. A diversity of cross-border mobilities -weekly, fortnightly, longer-term -is observed but, as the economic crisis (which impacted the building industry more seriously than any other sector) developed, many of the migrants moved on to France, where Queirós set up another site of ethnographic research. In this way he exemplifies Marcus' entreaty to 'follow the people' -in this case migrants -but also following the 'connections, associations and relationships' of the research participants across space and boundaries. I was particularly taken by Queirós' notion of his participants' 'bulimic' pattern of experiencing time and space through circular migration: on the one hand working hard and accumulating money on construction sites abroad, denying themselves pleasure or non-essential spending; on the other hand, on weekends or holidays back in Portugal, restoring self-esteem by relaxing, sleeping, drinking, socialising and spending money ostentatiously. At the same time, the volatile nature of construction work, with its unpredictability, authoritarian work culture and lack of human dignity, results in a kind of 'identity de-structuring' and, when the building boom collapses, forces the workers to 'move on', rendering them, in Queirós' words, 'true contemporary nomads'.
In Chap. 9, Marques further problematises the notion of skill through a study of Portuguese entrepreneurial activities in Luxembourg. The story of the Portuguese migration to Luxembourg is a rather hidden element in the overall history of labour migration in Europe. Recruited from the 1960s on, mainly to work in the steel and construction industries, by 2013 Portuguese were the largest immigrant group in the country, accounting for 37 per cent of all foreigners and 16 per cent of the Luxembourg population. Characterised by a generally low level of formal education, the Portuguese have, nevertheless, embedded themselves fairly successfully in the economic and social fabric of the country, including a growing trend to entrepreneurship, mainly in hotels and restaurants. This is exactly where the question of skill arises. If 'higher-skilled occupation' is proxied by education, as it often is (third-level education being the defining criterion), then this excludes an alternative conceptualisation of skill which refers to the 'actual' skills -management, responsibility, scale of financial turnover etc. -involved in running an enterprise. Hence the entrepreneurial route is a way for erstwhile labour migrants to achieve socio-occupational mobility through this 'intermediate' status somewhere between 'high' and 'low' skilled. This is the main theoretical point I wish to draw out from this chapter.
The next two chapters -Part IV of the book -are on 'postcolonial continuities' in migration and focus on the two main 'Lusophone system' channels of post-crisis migration, to Brazil (Chap. 10, by Rosales and Machado) and Angola (Chap. 11,by Candeias et al.). Both illustrate the interconnections between timing and economic trends in shaping new migration flows. In the case of Brazil, it was the coincidence of the ongoing crisis in Portugal with a boom in Brazil, partly due to the Soccer World Cup (2014) and the Olympics (2016). In the case of Angola, it was due to the oil boom, which drew in Portuguese migrants especially across the years 2005-2013. Recent Portuguese migrations to Angola and Brazil also illustrate a point I made earlier about the complex articulation of migratory forms that characterise the semi-periphery, and the way these North-to-South migrations challenge the normative premises of both economic migration theory and postcolonial studies. Hence, Portugal functions as both an old and a new periphery for South-North intra-European migration, it acts as a metropolitan core for immigration flows from ex-colonial and East European peripheries yet it also exhibits enduring colonial-era power relationships in sending its own migrants to those parts of the fast-developing global South where its cultural-linguistic influence is strong, and where there are also jobs and business opportunities.
For Brazil, the migrants, based on Rosales and Machado's survey data, are a predominantly higher-educated group and are differentiated by the two main destinations selected for the survey -Rio de Janeiro and São Paulo. In the former city the migrants tend to be in the academic, educational and scientific sectors, whereas in the latter there is a bigger involvement in the business sector. Alongside the economic motives, Portuguese also move to Brazil to experience a somewhat different lifestyle in an environment which is nevertheless reasonably familiar linguistically and culturally, yet tinged with new images of beautiful landscapes and happy-go-lucky people.
Although the 'old colonial' link is shared by both Brazil and Angola, in the latter the pull factors for this North-South migration are less to do with culture or colonial nostalgia, and more related to new opportunities opened up for Portuguese skilled workers in the oil industry, as Angola is the fifth largest oil producer in the world and the second in Africa after Nigeria. In actual fact, as Candeias et al. relate, the recent Portuguese emigration to Angola operates at two class/skill levels: on the one hand middle-class expats with high levels of technical and professional expertise who live in luxury housing; on the other, working-class labour migrants who can capitalise on their 'Portuguese' origins and credentials to get quite well-paid jobs. These characteristics make the 'new' Portuguese emigrants in Angola older than those in Brazil, and two-thirds are men. And unlike the Portuguese in Brazil, who are on the whole well-integrated, in Angola there is weak integration and the maintenance of strong ties with Portugal.
Finally, in Part V of the book, the focus is on 'Identities' -a complex topic indeed. In Chap. 12, Sardinha, following in the footsteps of the pioneering research of Christou (2006) on Greek-American 'returnees' to Greece, looks at the broadly parallel case of second-generation Portuguese-Canadians who relocate to Portugal in a quest for self-discovery. Once again, the Portuguese economic crisis intervened, turning these optimistic 'homecoming' trips into nightmares of disillusionment, resulting in yet another migratory round, either of 're-return' back to Canada, or 'twice-migration' to another (European) destination. Through longitudinal interviews and ongoing contact with his research participants, Sardinha charts the complex reality of their migratory trajectories, revealing how standard models of migration origin and destination are repeatedly 'reversed' and recycled by these sequences of moves. Sardinha closes his contribution by speculating what future cycles of migration may continue his subjects' migratory peregrinations.
Sardinha is not coy about acknowledging his own positionality -as a secondgeneration Portuguese-Canadian who has 'returned' to Portugal -as part of his research study. This auto-ethnographic stance is taken further by Cordeiro in her chapter, in which she considers the 'unbearable weight' of being classified as an 'emigrant', and by implication 'less Portuguese' than non-emigrants, in Portugal. This is a subtle exploration of the cognitive dissonance between being seen as an 'immigrant' in the United States, and as an 'emigrant' by the Portuguese in Portugal. Perhaps surprisingly, Cordeiro and her key informants identified much more positively with being an 'immigrant' in the USA -a country founded on immigration, which therefore becomes an inclusive concept -than being labelled an 'emigrant' in Portugal, which is seen as a term of exclusion, of 'othering', of being 'outside' the 'true national family'. This 'imposed hetero-classification', which is often contested or rejected by those Portuguese who live outside of Portugal, is especially painfully felt when visiting their hometowns. In contrast to many 'migrant-identity' studies which tend to make assertions about migrants' feelings of belonging along the lines of (taking the Greek-American case, Christou and King 2014, 210) 'when I'm in America I feel Greek, but when I'm in Greece I feel American', Cordeiro opts for the inverse identificatory construction as somehow 'ideal' and more reflective of a positive dual identification: 'Portuguese-Americans like to be Portuguese in Portugal as well as American in the US'.

Four Topics on Which More Could Have Been Said
Although the book presents a remarkably thorough and well-structured account of Portuguese migration routes, old and especially new, there are certain themes that are under-represented. I nominate four such themes, acknowledging yet again the subjectivity of my perspective, since these topics correspond to some of my main research interests in migration studies.
The first of these is the patchy coverage of return migration. There is a comprehensive treatment of the statistics on returns by Peixoto et al. in Chap. 3, and then Sardinha in Chap. 12 deals in depth with a very specific type of return -of the second generation born in Canada. But we learn little -perhaps because it is still too early to tell -of the return propensities of the recent emigrants who are mainly younger and highly educated fleeing the Portuguese economic crisis. This is also important in the light of a key point made at the end of Pires' Chap. 2, regarding the uncertainty of the country's economic and demographic future.
Second, a more geographical point: I was surprised by the lack of attention given to the spatial origins of the various recent outflows. We know from the extensive Portuguese migration literature that the prewar and postwar emigrations to North America were heavily concentrated from the Azores, whilst the mass migrations of the 1950s, 1960s and 1970s to France and other West European countries were sourced mainly from rural areas of mainland Portugal. But where in Portugal do the more recent outflows come from? Probably much more than in the past from urban areas, given the more educated character of the emigrants, but the evidence is missing.
Gender is my third axe to grind. Like all social processes, migration is fundamentally and unavoidably gendered. Men and women migrate for different -as well as sometimes the same -reasons, and migration is an act that is also -often unappreciatedly so -relational as regards gendered family and other personal relations. One issue with incorporating a gendered analysis into the study of migration is whether to have a separate chapter on 'gender' (or on 'migrating women' or for that matter on 'migrating men') or to ensure that gender is mainstreamed throughout the chapters. I think it fair to say that, with the exception of Chap. 5 on the emigration of Portuguese nurses, gender remains in the background throughout most of the book and perhaps could have been given more attention.
Finally, there is scope for more to have been said about the relationship between migration and Portuguese economic and social development. Once again, there are fleeting references (when discussing 'brain drain', remittances and return migration) but no thorough-going dedicated analysis. What does the latest round of Portuguese emigration -the fourth wave -mean for the long-term evolution of Portugal as a (semi-)peripheral country at the south-western margins of Europe?
Open Access This chapter is licensed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence and indicate if changes were made.
The images or other third party material in this chapter are included in the chapter's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the chapter's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.