2. Migration and its determinants
2.6 Systems and network approaches
The last set of theoretical approaches to migration to be briefly presented in this chapter consists of the literature on migration systems and networks. More than full-fledged theories on the determinants of migration, these consist of analytical perspectives introduced based on the consideration that, given the self-evident systemic and network features of migration processes, the study of the latter stands to gain from the heuristic application to the field of migration of the concepts and insights derived from general systems theory and network analysis.
The idea of applying a systems approach to migration can be traced back to the work of Akin Mabogunje (1970:1-2), who suggested a conceptualisation of (rural-to-urban) migration as an open system made up of both matter (migrants, institutions and organisations) and energy (both kinetic – the actual flows – and potential – the stimuli
45 behind the flows). Like most other open systems, migration systems are additionally characterised by the existence of feedback mechanisms (cumulative causation), interactions with the surrounding environment, adjustment mechanisms and component subsystems (the various formal and informal institutions that regulate migration processes).
In this article, however, Mabogunje did not proceed much further than contributing this seminal conceptualisation. Theoretically, the main proposition is that “one of the concomitants of the continued interaction between the system and its environment will be the phenomenon of growth in the system” (id ibid:1), but it is difficult to see why this should be so either in the framework of general systems theory (given that many systems are ‘exothermic’) or specifically in the case of migration (given that actually-existing migration systems eventually wane).
Despite the fact that its theoretical contributions failed to match its conceptual breakthroughs, the migration systems approach was met with considerable success. Not only did the expression itself become commonplace in the literature, it also became increasingly acknowledged that the study of actually-existing migration flows could be significantly improved by drawing on some of the insights suggested by Mabogunje. The Committee on International Migration of the International Union for the Scientific Study of Population, for example, explicitly endorsed a migration systems approach in much of its work in the 1980s (Kritz and Zlotnik 1992). However, some of the defining features of this approach were in fact lost, or changed, as of its dissemination process. Specifically, the disseminated version of the migration systems approach (Kritz and Zlotnik 1992; see also Massey 1993) largely discarded the idea of feeding abstract theoretical (as opposed to conceptual) insights from general systems analysis into the study of migration. What was retained above all was the idea of the heuristic worth of looking at sets of ‘sending’ and
‘receiving’ areas in a dynamic way and as open but relatively self-contained systems in interaction with other types of systems.
This stood in contrast to most of the previous literature, especially as concerned empirical studies of actually-existing migration streams. First, the call for consideration of both ends of migration flows in the assessment and analysis of those flows stood in opposition to those approaches that focused solely on one of the two ends, and in that sense paved the way for later calls for methodological transnationalism in migration studies (e.g. Wimmer and Glick-Schiller 2002). Second, concern with the dynamics of the various migration systems followed directly from the realisation of the existence of feedback mechanisms and from the acknowledgement of migration systems as having a momentum
46 and life-cycle of their own, whereby they are born, grow, mature and wane (hence synchronic analyses only capture a biased fraction of the whole process). Third, the idea that certain sets of ‘sending’ and ‘receiving’ areas (i.e. migration systems) make up more or less coherent ‘wholes’ that can be analysed in a relatively self-contained manner served to highlight the fundamental structures in an otherwise seemingly chaotic matrix of migration origins and destinations. This made even more sense in the second half of the 20th Century, when several new international migration systems, with their own specific characteristics and momentums, arose in previous metropolitan-colonial spaces. Finally, the inter-systemic element of the analysis built on the realisation that the flows of people are far from autonomous from a variety of other flows and structures and that, therefore, the study of the former should not overlook their interactions with the latter.
This last feature of the migration systems approach – arguably its most central one – is of course not exclusive to it: for example, Petras’ and Sassen’s historical-structural accounts (see above) also stressed the close geographical association between the global flows of labour and those of commodities and capital. Petras even referred to some of the other ‘systems’ (though not phrased as such) that co-determine the emergence, direction and intensity of migration flows, such as “political and economic networks” or “cultural affinity”. This, along with the emphasis on historical dynamics, shows that the migration systems and historical-structural perspectives are not only logically compatible but also complementary, even concurrent, in many respects (despite the apparent failure of the migration literature to acknowledge this). The main difference between the two is in fact one of emphasis: most historical-structural authors have focused primarily on the functions served by labour migration in the context of capitalism as a world system, and regarded the issue of how it is that these functional raison d’êtres translate into actual processes as a relatively secondary concern; whereas authors working in the migration systems tradition have mostly focused on the meso-level structures and processes that “stimulate, direct and maintain” (Fawcett 1989:671) specific systems in specific historical and geographical circumstances. These meso-level structures and processes provide the essential mediation between structural trends and constraints and the individual agency of migrants, households and organisations, and essentially consist of a variety of linkages that serve to facilitate or constrain the flow of people between two or more specific areas given the existence of “potential energy” in the system as a whole. Arguably, one of the most useful and comprehensive taxonomies of those linkages is that suggested by Fawcett (1989:674-8),
47 who subsumed them under four main categories: “state-to-state relations”, “mass culture connections”, “family and personal networks” and “migrant agency activities”.
One specific type of linkage that would fall under the fourth category indicated above and which plays a particularly important role in channelling and sustaining migration flows is that of migration networks (Boyd 1989, Gurak and Caces 1992, Vertovec 2002).
These consist of social networks, either based on pre-existent kinship, friendship or community ties or constituted for the specific purpose of sustaining migration (e.g. human smuggling networks), which in the context of migration become crucial channels for the flow of people, resources and information. These networks “reduce the costs and risks of migration” (Massey et al 1993:449) by “buffering migrants from the costs and disruptions of migration; insulating migrants from the destination society and maintaining their links to the origin society; determining, to a degree, who migrates from communities and households; influencing the selection of destination and origin sites; conditioning the integration of migrants in the destination society; serving as channels for information, other resources and normative structures; and shaping the size and momentum of migration”
(Gurak and Caces 1992:153).
Whether or not migrant networks are the primary drivers of a given migration system as of its inception, it is virtually inevitable that such networks arise and be consolidated as the migration system matures. In this sense, they “flesh out migration systems” (id ibid:166) and constitute a feedback mechanism that contributes to cumulative causation. Inasmuch as migration systems can themselves be argued to ‘flesh out’ the structural trends and constraints at work in the capitalist world system as far as labour circulation is concerned, the migration systems and networks approaches arguably make it possible to bridge the micro-macro and agency-structure gaps – an idea that is further elaborated in the next section.