Introduction

Today’s globalizing world is regularly presented as the world of spaces of flows (Castells 2000) and transnational scapes (Appadurai 2001) enabling images, goods and money to move freely within dispersed communities. At the same time, it is widely acknowledged that global spaces are not always frictionless spaces (Mitchell 1997).

The geographer Doreen Massey (1994, 2005) has convincingly argued against the homogenizing effect of time-space compressions as she put forward that people simply have different accesses to ‘the global.’ In line with this, Zygmunt Bauman (1998) stresses that people’s potential to be physically mobile in favored directions is one of the most important stratifying factors of our times. This has become particularly striking in the framework of contemporary African migration towards the European Union (EU). The increased connectedness of many African societies – by means of, among others, mobile telephony (de Bruijn, Nyamnjoh and Brinkman 2009) and the Internet (Tandian 2010; Burrell 2009) – contributes to people’s ‘imagined worlds’ (Appadurai 2001) and triggers ‘adventurers’ (van Dijk, Foeken and van Til 2001) to reach faraway places. These migration aspirations emerge, however, in times that the popular destination ‘Europe’ is doing its best effort to keep out would-be immigrants originating from the global South. This leads to a widespread frustration among sub-Saharan Africans over their relative immobility (see for a Cape-Verdean case Carling (2002), for Malian cases Barten (2009) and Jónsson (2008), and for Senegalese cases Poeze (2010) and Nyamnjoh (2010)). Partly because of this frustration, many migrants take the route of irregular migration in order to reach the EU (de Haas 2007).

This article focuses on the often risky and fragmented migratory journeys of sub-Saharan African migrants towards the EU. Because of their fragmented character, these journeys undermine our understanding of migration as a one-off movement from a place of origin to a certain destination (Collyer 2007) – a single move from here to there. They are rather stepwise, multi-local, and process-like phenomena evolving ‘en route’ (Schapendonk 2011, 2012). I stress that we can better understand the dynamics of these journeys when we focus on their spatial dynamics. This implies that we need to follow migrants’ mobilities through different places, their periods of immobility in these places, as well as their translocal connections reaching faraway places.

By taking this as the central point of departure, this article has three main objectives. Firstly, I aim to analyze migrants’ physical mobility in relation to other mobilities. I therefore present an analytical lens that is mainly inspired by the so-called mobilities turn (e.g. Sheller and Urry 2006; Urry 2007; Cresswell 2010; Adey 2010). This lens takes the five mobilities that are distinguished by John Urry (2007) as a starting point to investigate migration journeys. These mobilities are: the physical mobility of people, the physical mobility of objects, imaginative travel, travel through virtual spaces, and communications. By analyzing three migration trajectories, I will show how the different mobilities interact with each other, and how these interactions help us to explain individual migration processes.

Mobilities research, however, always runs the risk of forgetting about significant immobilities (Cresswell 2010). Scholars who engage in the mobilities debate therefore particularly pay attention to the facilitating role of relative fixities in mobility processes (e.g. Sheller and Urry 2006; Bærenholdt and Granas 2010). In line with this, my second objective is to explore the role of ‘place’ in the story of migrants’ movements. I thereby distinguish two sorts of places. The first place is the geographical locality providing the physical infrastructures facilitating migrants’ journeys. This place is more or less locatable. The other place is a social and more mobile kind of place. Massey (1994, 2005) inspires us to analyze places as specific nodes in which numerous social networks meet. Inspired by this conceptualization, I argue that there is a ‘place’ carried by migrants as the node of their social networks in their migration processes. With regard to this networked place, several questions arise from a mobilities perspective. What if the node of the network moves itself? What if the connections move, or do not move, with the node? I explore these issues by discussing the relation between places as geographical localities, and migrants’ networked places that are mediated during their journeys.

Finally, my third objective is to reflect on the reciprocal relationship between mobilities and places (Bærenholdt and Granas 2010). Where places facilitate mobilities, mobilities also change and give meaning to places (Massey 2005). With this in mind, I explore how migration journeys change the places they pass through. I therefore introduce the metaphor of ‘sediments’ being the spatial transformations that are the effects of migratory movements (see also Adey 2010).

By investigating these spatial dynamics of migration trajectories I will reveal the complex causalities of migration journeys. This will help us to realize that individualistic explanations (in particular the neo-classical push-pull model (Lee 1966)¹ in which localities are believed to attract migrants without much interference) as well as structuralistic explanations of migration are not very helpful to explain the dynamics of migration journeys.

On the Road: African Stepwise Migration Towards the EU

The age of migration (Castles and Miller 2009) implicates the paradox of distances that become smaller while the barriers erected by governments to overcome these distances become larger for the majority of the world’s population (Massey et al. 1998). The EU can be used as a telling example to explain this pattern of exclusion. On the one hand there is the fundamental right of free mobility of EU citizens across EU’s internal borders. At the same time, for most of the people living outside the political entity, the EU means social closure and blocked access (Verstraete 2001). Most illustrative in this context is that the European Commission holds a clear-cut visa regime consisting of two visa-lists: one list of states whose nationals can enter the EU without visa applications, and another list, the black list, of states whose citizens require a visa to enter the EU. Remarkably, the latter contains 135 (including all African states) out of the total 195 states in the world (van Houtum 2010). This suggests that the ‘global village’ is for many people indeed more like an enclave society in which immobility regimes rule (Shamir 2005; Turner 2007).

Due to this geo-political closing, a considerable number of Africans attempts to enter the EU by unauthorized means. For the migrants, these journeys require, among others, a considerable sum of money, courage, the right contacts, and some organisational capacity (Schapendonk 2011). Many individuals do not have the right mix of these ingredients resulting in a growing number of sub-Saharan African migrants waiting in EU border regions for their chance to enter their ‘El Dorado’ (Zoomers 2006). There are considerable African ‘transit communities’ in the Maghreb countries (e.g. Bensaad 2007; de Haas 2007; Hamood 2006; Collyer 2007), Turkey (e.g. Brewer and Yükseker 2006; Içduygu and Yükseker 2008; Suter 2009), and Libya (Hamood 2006).

Transit migration is a highly debated topic in policy arenas and academia (Düvell 2006; Collyer, Düvell and de Haas 2012; Papadopoulou-Kourkoula 2008). In general, this debate has a simplistic and linear understanding of migration as if migrants depart, go through a transit phase, and settle afterwards (see for similar critical comments Collyer, Düvell and de Haas 2012). In line with this, the empirical focus in transit migration studies is mostly migrants’ intentions. I have argued elsewhere (Schapendonk 2010) that this intentional approach is not very helpful to understand these fragmented journeys. Such a focus runs the risk of positioning migrants as rational actors following a straightforward and pre-known plan. Migrants in transit situations may in fact have multiple scenarios in their heads. They may decide under certain conditions to stay in a perceived transit country, while they do not at the same totally forget about moving onwards. I have also stressed that migrants’ plans and aspirations during their journeys are subject to change (Schapendonk 2010, 2011). As aspirations change, so too do migrants’ transit statuses. In other words, transit situations may turn into home-like situations and, equally so, destinations may turn into transit places (see also Du Toit 1990). Moreover, several authors underline that migrants in transit often lack a clear destination where they are heading to (Collyer 2006, 2007; Brewer and Yükseker 2006). I have found that many sub-Saharan Africans who are living in Morocco and Turkey have rather abstract destinations in their heads. Instead of ‘Paris’ or ‘my family in Madrid’ they are moving to ‘Schengen,’ ‘Europe as a place of freedom,’ and a ‘good place to live in’ (Schapendonk 2011). Partly because of migrants’ changing aspirations and abstract destinations, fragmented migration projects evolve ‘en route’ and are therefore dynamic, open, and complex phenomena (Schapendonk 2009; Collyer 2007, 2011).

It is a misunderstanding that fragmented migrations end when migrants succeed in reaching the European Union. Many African migrants actually continue journeying across national borders in Europe. There are examples of Somalis moving from the European mainland to the UK (Bang-Nielsen 2004; van Liempt 2010), Senegalese moving between Italy and France (Schuster 2005), and Ghanaians moving from the Netherlands to Spain (Schapendonk 2011). Non-African examples are: Tamils moving from continental Europe to the UK (Lindley and van Hear 2007) and Kurds moving stepwise to Germany (Schuster 2005).

I will argue that the physical localities migrants move to are not the most important to understand the dynamics of migration journeys. Barcelona, Paris, or Tilburg are not pulling migrants from afar; it is rather the interconnectedness of mobilities that direct them to these localities. To come to this argument, I present the so-called interconnected mobilities lens helping us to analyze the dynamics of migration journeys.

The Interconnected Mobilities Lens

It is striking that migration has predominantly been analyzed from a sedentarist perspective; it is merely seen as a movement from one settled stage at the place of departure (dis-placement) to another settled stage at the destination (re-placement) (Cresswell 2006; Schapendonk 2012). By focusing on the act of moving, this study follows the general critique on social scientists that they mostly focus on the roots instead of the routes of social interactions (Clifford 1997; Urry 2000).

This critique is underlined by scholars who are involved in the ‘mobilities turn’ that positions mobility at the center of social sciences, both methodologically and conceptually (e.g. Sheller and Urry 2006; Cresswell 2006, 2010; Urry 2007; Adey 2010). Mobilities scholars do not argue that social science has never dealt with movement, but emphasize that social science has traditionally approached mobility as ‘residual death time’ (Sheller and Urry 2006; Cresswell 2006). The fact that migration research has mainly overlooked the journey as an empirical object underlines this statement. As Brian du Toit (1990: 308) already argued 20 years ago:

[M]ore attention should be given to the “journey” than to the “origin” or “destination” because people who move may not know exactly where they are moving to, nor do they necessarily remain there once they reached this destination. The migrant may explore better opportunities, may move on to a new situation, or may return to the point of departure. Migration is not an act but a process.

I present here a so-called interconnected mobilities lens that helps to understand the dynamics of contemporary migration journeys. The starting point of this lens is that we should not perceive migrants’ trajectories as isolated and straight paths that occur according to well-weighted plans of individual actors. Instead, trajectories are approached as open processes in which migrants’ corporeal mobility interconnects with five other forms of mobility. These other mobilities are (Urry 2007):

• The corporeal mobility of third persons

• The physical mobility of objects: including goods and money

• Imaginative travel of migrants: daydreaming, imaginations

• Virtual travel of migrants: traveling with the help of the internet, television, or other media

• The mobility of information through communications

To understand how these five mobilities interact, we first need a specific notion of geographical space as being constantly made and re-made by socio-spatial relations. Hence, the mobilities lens is based on relational geography (Amin 2002; Massey 2005). At the same time we need to avoid a simple post-modern argument that everything is on the move. I therefore pay attention to factors in migration journeys that are not so mobile.

From Geography of Scale to Geography of Connections

The mobilities turn is part of a broader geographical-theoretical project that aims to go beyond fixed container-like analyses of territories (Sheller and Urry 2006). Its main theoretical foundations are partly based on complexity thinking (Urry 2003; Law and Urry 2004) and Actor Network Theory (Latour 2005). These relational starting points have two major implications in the framework of spatial dynamics of migration journeys. Firstly, mobilities thinking moves away from the sociality of proximity (Thrift 2004; Larsen and Urry 2010). In the context of migration, this implies that migrants’ movements can be directed by the very distant and (apparently) absent while the geographically near may not have any influence at all. Secondly, mobilities thinking does not follow geometrically hierarchical frameworks assuming that the larger scale (the global) is determining processes on the smaller scale (the local). Instead, it maintains a glocal understanding of the world that is based on “a topology marked by overlapping near-far relations and organisational connections that are not reducible to scalar spaces” (Amin 2002: 386). Scales, like the local, the national, the regional, and the global, are considered social constructs that lack any kind of territorially fixed and pre-set statuses. Following this relational line of thought, geographical space is a matter of connections and networks of different stretches rather than a matter of one scale above the other.

Network thinking has also gained a prominent place in migration theories in recent years (e.g. Palloni et al. 2001). The general assumption is that migrants’ social connections in faraway places provide pathways for migration and thereby these connections lower the costs of the migration process. The notion of bonding and bridging social capital is helpful to understand the functioning of social networks (Woolcock and Narayan 2000). In the migration process, migrants often need contacts other than their inner circle of friends and direct family members (bonding social capital) in order to ‘get ahead.’ These bridging connections are the channels that enable the transfer of new ideas and fresh information (Granovetter 1973). While network approaches are very welcome alternatives to the individualistic migration model, the spatial conceptualization of migration networks is, however, rather static. Networks are all too often approached as consisting of everlasting social ties – they are fixed grids of people’s social connections. Little attention is paid to how these social connections are mediated through time and space (Collyer 2005). This mobilities lens is sensitive to migrants’ continuous processes of connecting, dis-connecting, and re-connecting (see also Schapendonk 2009, 2011).

All That Is Not So Mobile

Mobilities research always runs the risk of overlooking the importance of relative fixities in socio-spatial interactions (Cresswell 2010). Hence, mobilities scholars emphasize that we need to understand mobilities always in relation to relative fixities. There are three reasons for this. Firstly, various mobilities are facilitated by relative fixed infrastructures. The airplane needs an airport, the mobile phone needs its mast, and the internet needs cable infrastructures (Urry 2007; Adey 2006, 2010; Cresswell 2010; Baerenholdt and Granas 2010).

A second reason to argue against the ‘everything is mobile principle’ is the fact that mobilities may bring their spatial ‘sediments.’ Like fluvial sedimentation in the context of physical geography, migratory movements may bring their changes in spatial organizations with a more or less structural character. The sediment of mobility may for example be a new pathway after several people have repeatedly taken the same shortcut in order to cross a field in a park (Ingold 2007). This new pathway makes the continuation of movement easier than it was before. In this context it is interesting to note that the identified trans-Saharan routes taken by sub-Saharan African migrants are quite similar to the trans-Saharan trading routes of the early 20th century (de Haas 2006) indicating that there is some patterning element involved in this ‘new’ migration. However, this does not mean that these structured routes determine individual trajectories. Sub-Saharan African migrants seldom follow exactly the preset pathways from West to North Africa. Because of their flexibilities and due to the involvement of exploitative transporters and corrupt state authorities, migrants regularly move backwards and forwards, take irregular side-paths, and make unexpected detours (Brachet 2005; Schapendonk 2009, 2011). As a result, individual voyages of sub-Saharan African migrants cannot entirely be explained by the age-old structured pathways that are often drawn on migration maps.

Finally, and in line with the former point, I argue against the ‘everything is mobile principle’ as it overlooks the existence of borders that are playing a prominent role in the migration journeys of my respondents. Because of border practices the costs/risks of migration become higher. Borders cause delays in individual trajectories and leave migrants stranded in perceived transit places (Collyer 2010). Consequently, the migration processes of sub-Saharan Africans are not only characterized by physical mobility but also by involuntary immobility (Carling 2002). In other words, what is often typified as transit migration is, for the migrant, not so transitory at all since they remain, possibly against their will, in certain perceived transit areas for considerable periods of time (Içduygu and Yükseker 2008). Thus, while the mobilities lens that I present here emphasizes the importance of mobilities and connections, it does not completely forget about immobility.

Interconnected Migrations

In this section, I will analyze three migration trajectories with the help of the interconnected mobilities lens. I have gathered in-depth information on migrants’ trajectories by means of in-depth interviews with sub-Saharan African migrants (see also Schapendonk 2011). I interviewed migrants who appeared to be ‘on the road’ to the EU’ (so-called transit migrants in Morocco and Turkey) as well as migrants who appeared to be settled in the EU. In addition, I gained some longitudinal information on migrants’ trajectories by maintaining contact with some of my respondents through the internet and mobile telephone communication. In total, I conducted 108 interviews and followed 16 migrants for a longer period of time.

Below I outline three trajectories in order to illustrate how we can apply the mobilities lens as outlined above. I have followed all three migrants for a period of two years, and at the moment of writing I am still in contact with them. This enables me to understand the (un)foreseen twists and turns of their migration projects. It also means that I have become part of their social networks. Although I was never asked by my respondents to assist them during their migration journeys, I can be considered a resourceful person who directly or indirectly has an impact on the courses of my respondents’ trajectories (see for more information Schapendonk 2011).

Trajectory I: The Love Connection

Dawda² is a 30 year old Gambian man. I met him for the first time at the train station of my home town Nijmegen (The Netherlands) in September 2008. From then on we met each other on a regular basis. Dawda told me how he, as a teenager, moved from the village he was born in to the capital of The Gambia, Banjul. In Banjul he pursued his secondary education. After he finished school he found a job as a guard at a luxury hotel in the same city. There he came into contact with many tourists from Europe. Since the 1970s, The Gambia became an increasingly popular tourist destination for Europeans. And as Dawda explained, the first contacts with Western tourists led him to consider emigration from The Gambia for the very first time. He was not only inspired by images of luxury the tourists brought into the country, but also the stories of Gambian men falling in love with European women, giving them a chance to visit Europe, made him think of going elsewhere. Some years ago, Dawda met a Dutch woman who stayed in the Hotel where he was working. After a while the love connection was made between him and the Dutch tourist. They remained in contact by email, chat conversations, and telephone calls. The woman re-visited him in The Gambia and Dawda was able to arrange twice a two-month visa for the Netherlands (in 2007). After the second visa expired he went back to his home country. There he started the procedure to be a candidate for a permanent stay in the Netherlands. Therefore, he took the official Dutch integration test in Dakar (Senegal).³ In Dakar he stayed with a Gambian friend who previously had moved to this city. He passed the exam, yet he had to wait for some months to obtain his visa. He finally arrived in The Netherlands in the summer of 2008. From then on he has lived with his girlfriend in the city of Nijmegen. He nowadays holds a residence permit that is valid for one year. However, this permit may be extended to a renewable permit in the near future. Any confrontation with the Dutch authorities may harm this process of getting a renewable permit. For this reason, his informal work as a gardener in the region of Nijmegen is risky. In the summer of 2009, Dawda was so happy to tell me that his girlfriend gave birth to a baby, which really made him “settled” in The Netherlands. Yet, he is still waiting for his renewable permit.

In this first example we notice that Dawda’s migration to Nijmegen is interconnected with other forms of mobility. Firstly, his imaginative travel (the day-dreaming of European luxury) strengthened his motivation to go out of The Gambia. This imaginative mobility was triggered by the travel of European tourists to his country of origin. Then, of course, the mobility of a particular person, Dawda’s future girlfriend, is important to understand his migration trajectory. Without her going to Banjul, Dawda would never have lived in Nijmegen. At the same time, the visit of this Dutch woman was, in itself, not necessarily decisive in Dawda’s migration. The love relationship had to be maintained. This involved all kinds of mobilities: the re-visit of Dawda’s girlfriend to Banjul, Dawda’s repeated visits to the Netherlands, the virtual travel on the internet, and the numerous long-distance communications between these two people.

To explain Dawda’s movement to Nijmegen, we may argue that his movement as a teenager to the Gambian capital (a relatively ‘small’ event) matters as much as the transnational tourism flows directed to Banjul (a ‘big’ structure), or the fact that a Dutch women was in the process of making a reservation at the hotel where Dawda was working (another ‘small’ event). Moreover, we should not forget that Dawda was able to pass the integration test in Dakar. Although this story contains some structural and place-specific elements (Banjul as a popular tourist destination, the embassy in Dakar), these are just some of the many factors influencing the migration process. Altogether, there is no one single cause that can be identified as the decisive factor in explaining why Dawda moved to the Netherlands.

This certainly does not mean that Dawda is a passive person lacking any say about his life and migratory trajectories. He has made vital decisions and he has put a lot of effort into his migration project (e.g. maintaining contact with his girlfriend once she left The Gambia). At the same time Dawda was already aspiring to move to Europe at the time he became inspired by the images of European luxury. Despite this ‘decision’/motivation he had not been able to move in this preferred direction. For this reason I stress that Dawda’s agency is not a personal asset of the human being that is called Dawda, but is in fact a relational effect (Urry 2000). In other words, there is not much left of Dawda’s agency when we disconnect him from his social network and from all the material objects (cars, airplanes, visa, computers, etc.) that help him to get ahead. His agency is in fact networked. Instead of focusing on either ‘big things’ (like structures) or ‘small things’ (like individual agency), I argue that we can understand Dawda’s movement better by focusing on how big and small things come together in single migratory trajectories.

Trajectory II: The Flexible Plan

Eric is a 26 year old Nigerian man. He originates from the Edo state in southern Nigeria. I met him for the first time in Rabat (Morocco), in January 2008. There he told me how he, back in Nigeria, was inspired to migrate to ‘Europe’ by Nigerian football players in European leagues. He kept the European dream alive for several years and, by working as a butcher, he was able to save some money to realize this aspiration. He did not inform his elders about his plan to move to Europe; he only asked his older brother for advice. Eric’s brother also gave him some money for the first steps of his journey. In 2006 Eric departed to Kano (northern Nigeria). There he searched for Nigerien traders who could bring him to Zinder (a city in southern Niger). According to Eric, this border crossing was very easy to arrange. In Zinder, he became good friends with three other Nigerians who were also on their way to Europe. The four Nigerian “brothers” created a “Nigerian crew.” The four friends arranged transport to Agadez (Niger), a city where the trans-Saharan journey can be further arranged. Eric told me that this is the place where the real struggle began. Due to the involvement of “big businessmen” migrants are easily cheated and exploited. Luckily, Eric and his friends managed to arrange transport to, respectively, Tamanrasset, Ghandana, and Maghnia (Algeria). In Maghia, Eric and his friends were victims of robberies by co-migrants (according to Eric the robbers were Nigerians). They took his money and his clothes. For this reason, Eric was forced to stay longer in Algeria than expected. After some weeks he crossed the Algerian-Moroccan border and ended up in Oujda (Morocco), from where he later traveled to Rabat. At the moment I met Eric, he was in Morocco for a period of one year and three months. The reason for this immobility was the fact that he was not able to finance the last jump to Europe. However, after some more months of waiting, Eric came into contact with a Moroccan re-migrant who had stayed in Greece for a couple of years. This man supported Eric financially and gave him shelter. After a while, Eric told this “new friend” about his plan to go to Europe. Since passing to Spain had become extremely difficult, due to the increased surveillances of the Spanish border areas, he changed his original plan of going to Spain by reaching Europe via Libya. He arrived in Libya by crossing Algeria overland. This journey was financed by his new friend (the Moroccan re-migrant). In Libya he worked in the construction sector. Instead of money, he was paid for the hard work with a “ticket to Europe.” Together with approximately 40 other migrants he boarded a boat that was departing for Lampedusa (the Italian Island in the Mediterranean). Eric reached this island after an exhausting boat journey. He stayed at Lampedusa only for a few days as he was transferred to a detention center in Crotone (southern Italy). From this place, Eric called me to tell me that he had successfully reached Europe. At the same time, he told me that he was scared of the rumors in the detention center. It was said that the authorities would deport all migrants in detention. For this reason, he “escaped” the camp. He took a train to Firenze. During his stay in Tripoli (Libya) and in the detention camp (in Crotone), he learned that many Africans were living in Firenze, and that the local authorities were relatively migrant-friendly. But the first days in this promising city were extremely difficult for Eric as he lived for six days as a wanderer at the Firenze train station. On the sixth day, he came into contact with a compatriot living in Prato (a city nearby Firenze). This man brought Eric to Prato and, for the moment, he is living there with two Nigerian friends in a rented house. He earns his money as a street vendor; he sells imported housekeeping products that he buys from Chinese bulk supermarkets.

Investigating this migration trajectory with a mobilities lens shows that Eric’s migratory mobility is interconnected with other forms of mobility. Firstly, there is the mobility of third persons (Nigerian friends, the Moroccan re-migrant, the African migrants on the boat, and the migrants in the detention center). Secondly, Eric traveled imaginatively since he had dreamt about being a football player in Europe. Thirdly, there is the mobility of information that is shared among co-migrants. Finally, we may argue that the mobility of Chinese imported goods is important in this story as it enables him to sustain a living in Italy.

This case also clearly illustrates that migration journeys are not only about being mobile; Eric has been physically immobile in different places (Tamanrasset, Maghnia, Rabat) for different periods of time. Eric’s immobility is strongly related to the specific costs/risks involved in contemporary migration to the EU. The increased controls of the Spanish-Moroccan border are important in this respect, and he was also literally immobilized by the Italian authorities by keeping him in detention in Crotone.

It would be simplistic too state that Eric was pulled by the city of Prato. It would also be an overestimation of the power of Eric’s will if we say that his decision to move to Italy is the most vital aspect of his trajectory. To the same extent, it would be an over appraisal of the functioning of his existing social networks if we say that his binding and bonding connections helped him to reach Italy. In fact, Eric needed new contacts outside his initial social network (the Moroccan re-migrant, the migration industry in Libya, the man at the Firenze train station) in order to successfully arrive in the place he is living now. It follows that we can hardly understand Eric’s trajectory by perceiving his movement as a result of individual decision-making at a certain time, in a certain place. Not instant go/no-go situations, but process and contingency matter here. I suggest putting the process/trajectory itself central in order to understand the outcomes of migratory trajectories.

Trajectory III: Pushed in Transit

Jean-Louis is a 36 year old man from the Democratic Republic of Congo. He left the Congolese capital Kinshasa in 1997 because he and his family were accused of being too closely connected to the Mobutu regime. Together with his sister and his mother, he went to Congo Brazzaville first. This border crossing was quite easy as the family was able to cross the Congo river. After a relatively short stay in Congo Brazzaville, Jean-Louis took an airplane to Nigeria because he felt unsafe in Brazzaville. He stayed in Nigeria for six years. He lived a happy life there as he was well respected by the Congolese community in Nigeria and he had a good income by working as an electrical engineer. In 2003, however, rumors circulated within the Congolese community in Nigeria about secret government agents coming to Nigeria in order to prevent an eventual political revolution. After two suspicious lethal accidents (Jean-Louis was sure that within 48 hours two of his friends were poisoned by Congolese government agents), he decided to consult his sister who had moved in the meantime from Congo Brazzaville to France. His sister advised him to leave Nigeria as soon as possible. At that point, Jean-Louis had no exact destination in mind. He was thinking of Benin, but he thought this country was too poor. He thought about Ivory Coast, but he was not sure about the safety of this country. At the end, he decided Mali was the best option. He went to Bamako by taking several buses. In the Malian capital, however, he did not find a proper job and therefore he was not willing to stay there. Other Congolese migrants in Bamako informed him about Rabat, and particularly about a Congolese pastor assisting the Congolese community there. This was the main reason why Morocco was his next destination. Via some African “brothers” in Bamako he bought a false Malian passport. This is important since Malians are allowed to move freely between Algeria and Morocco. After his stepwise journey through Algeria (via Tammanrasset), he reached the border town Maghnia. At the time he entered this Algerian-Moroccan borderland, Jean-Louis was forced to wait for a specific period of time since the ‘famous’ African invasion of Ceuta and Melilla, the Spanish enclaves on Moroccan territory, had occurred some days earlier (this was September 2005). After this incident Moroccan authorities strengthened their authority performance towards (irregular) migrants and therefore Morocco was not very easy to enter. After some weeks Jean-Louis finally reached Rabat and got in contact with the Pastor he had heard of in Bamako. After a while, he managed to get a small job at a hotel with a French owner. During the interview in Morocco in January 2008 he seemed to be content with his life in Morocco. However, Jean-Louis sent me a surprising email just after the summer of 2008. He stated that he had arrived in France. He explained to me how he ‘borrowed’ a UK passport from an African man who was legally living in the UK. With this passport he was able to cross the border to Spain by just boarding a conventional boat. This strategy of borrowing someone’s passport to be able to cross borders is known as the look-a-like strategy. After a while, Jean-Louis told me the reason behind his onward movement. He was involved in a violent conflict with the son of his employer (the French hotel-owner). Jean-Louis was attacked by this man and became seriously injured. Consequently, he lost his job, and more importantly, he lost confidence in life in Morocco. He was staying at his sister’s place at the moment of the email. Later he moved to Grenoble, where he applied for asylum.

By analyzing this third trajectory from a mobilities perspective we notice again that the migratory mobility of the migrant in question is dependent on other forms of mobility. The movement of his sister to France is, evidently, important in this framework. But labeling Jean-Louis’ movement as a classic example of chain migration is like making a zig-zag line straight. Such simplification of migration processes would only reinforce the sedentarist notion of migration as a single move from one settled stage to another. To understand Jean-Louis’ stay in Grenoble today, we have to include his movements to respectively Congo Brazzaville and Nigeria. Without these movements he would not have stayed in Mali and Morocco, and he probably would not have ended up in France. We understand these intermediate steps better by taking into account the mobility of information in different places (e.g. rumors about government agents in Nigeria, information about the Congolese pastor and organization in Morocco), the mobility of people (his sister in Paris, his compatriots in Nigeria, Mali, and Morocco), and the mobility of goods (the Malian passport that enabled him to cross to Algeria, the European passport that enabled him to cross from Morocco to Spain).

With this focus on mobilities, I indeed plead for a non-reductionist framework to analyze migration trajectories. Migrants’ movements are directed by a specific coming together of different mobilities and different social relations. In other words, trajectories can be understood as a ‘thrown-togetherness’ of mobilities and connections. This notion of a thrown-togetherness of factors is very much in line with Massey’s relational understanding of places (Massey 2005). A place is, according to Massey, made out of a simultaneity of manifold relations with different temporal and spatial characteristics. In other words, a place is always related to other places. The specificity of a place, then, does not derive from a local history or internal roots, but from the historical and spatial relations with other places. Similarly, as I argue here, we can think of a migration trajectory as always related to other mobility trajectories. The trajectory of a migrant is directed by the trajectories of other people, information and objects. Thus, it is not so much the locality ‘France’ that pulled Jean-Louis to Grenoble, nor is it only his social network (the connection with his sister) that directed him to this place – it is rather the contingent connectedness of different mobilities that brought him there.

What Is Left for Place?

The three examples illustrate that it is not the destination that determines migration processes. With these illustrations I contrast migration models that emphasize the role of localities – as if localities push or pull migrants. Instead I suggest approaching migration trajectories as a thrown-togetherness of mobilities. This, however, does not mean that I claim that everything is on the move and that we can simply forget about place, territory, borders, locality, etc., when we analyze migration. So what then is left for ‘place’ in the analysis of migration?

To understand migration trajectories I distinguish two sorts of places. There is the geographical locality as ‘place’ and migrants’ ‘place’ as the node in social networks. I may argue that geographical localities may not be decisive factors in migration trajectories, but that does not mean that they are not important. Firstly localities can be seen as geographical access points. People meet other people in specific localities and connect to other mobilities also ‘in place’. Transport infrastructures are locality-specific and they provide different travel opportunities in certain directions. In terms of mobility opportunities it matters, for instance, if you are nearby or far away from a train station. It matters also what train station is nearby as different stations provide different mobility opportunities. Similarly, a cyber-café is a locality that provides the opportunity to travel. Localities, in short, are the facilitators that make mobilities possible. Hence, it matters, in terms of travel opportunities, in which locality you are.

Next to this facilitating role of localities, we may state that they are also embedded in mobilities. The image of Rabat in the story of Jean-Louis traveled to him by a form of mobility. Firenze was brought to Eric by similar flows of information. Localities are indeed on the move (Casey 1993; Amin 2002); they are embedded in the ‘trans’ of translocal interactions. In this context we may argue that localities do not matter for the migrant unless the locality is embedded in one of the mobilities, be it by his/her own mobility (their arrivals in new locality), personal communicative connections (e.g. talking to a brother or sister living in Paris) or by virtual connections (e.g. seeing the Paris business area La Défense on the internet).

It has been already argued that migrants’ ‘places’ in their social networks are highly important to understand their journeys. This specific ‘place’ is mobile for at least two reasons. Firstly, as the migrant is on the move, so too is his/her social network. By visiting other localities, geographically and virtually, new friendships can be made. By leaving geographical localities, former close contacts may become weakened or even lost. In this context it is interesting to note that some scholars highlight the fact that African migrants create more or less spontaneous collectives during the stepwise overland journeys from West to North Africa (Collyer 2007; Brachet 2005). After a specific period (e.g. an ‘immobile’ period in a migration camp in the bushes of Morocco, or a life-threatening boat-trip to one of the European islands), many collectives break up for the most part. Very close ‘brothers’ may become distant ‘brothers,’ both socially and geographically (Schapendonk 2011). At the same time, where contact remains between individuals, close brothers remain close brothers regardless of the geographical distance between them. Journeying is indeed “a source for both solidarity and detachment, a cement and a solvent of human association” (Leed 1991: 294). With these (dis)connections migrants mediate their place in their social networks (Schapendonk 2011).

A second reason why migrants’ places are mobile during the process of moving is the fact that they intensively make use of portable and mobile information and communication technology. All migrants whom I interviewed in Morocco and Turkey (57 in total) carried one, or more (!), mobile phones. In that sense, migrants’ actual networked place can often be traced back to the tiny sim-card in the mobile telephone. It is then not very surprising that the first asset border guards take from irregular migrants is their mobile phones, as a Moroccan NGO representative stated: “then they take them out of the network” (see also Schapendonk 2009).

In this framework, relevant questions are: how do the geographical locality and the mobile networked place relate to each other? Is there a relation at all? The answer to these questions is already partly provided by stating that localities can be viewed as access points for translocal information. In unfamiliar localities, migrants often need some ‘locally’ embedded contacts to profit from or adapt to a strange environment for a specific period of time. As the examples of Eric and Jean-Louis show, these ‘local’ connections may result in temporary jobs and housing. The same local actors may also provide the bridge for translocal connections. Transport agents, for example, bring migrants from one place to the other. Other actors in localities may provide important migration-related information and/or they might connect migrants with migration facilitators elsewhere. In some places, like Kano (Nigeria), Gao (Mali) Agadez (Niger), Oujda (Morocco), Tamanrasset (Algeria), and Istanbul (Turkey) there is an accumulation of specific migration-related services to help migrants get ahead. Some services are directly linked to the facilitation of migration, such as transport businesses and the renting out of rooms for migrants. Other services, like cyber-cafés and telephone shops are indirectly linked to the facilitation of migration (see also Brachet 2007). Because of the presence of these services, many migrants go to these ‘migration hubs’ in the hope of reaching other places. All this indicates that locality-specific elements still matter in migration processes. But they matter only in relation to the access they provide to various mobilities. As a result, local and translocal connections must not be seen as ‘either/or categories’ since these connections overlap. A local actor is never entirely local (Amin 2002).

Yet, it is safe to state that localities lose some of their importance for migrants in their migration process since migrants are able to connect with their bonding, but geographically distant, social capital from almost any locality. Some migrants even stated that they had some connections by mobile phone during the crossing of the Sahara desert (see also Schaub 2010). These networked places prevent migrants becoming placeless; someone may feel very ‘in place’ in a locality from which he/she is completely isolated because bonding contacts are not necessarily proximate contacts. A direct reference to this networked place came from a Congolese migrant in Rabat who responded as per the following to my question whether I could visit her again the next time I visit Rabat by stating: “I don’t know, you better call me because my mobile is my address.” This single quote illustrates that migrants’ networked places make migrants rather flexible in the process of moving.

To sum up, I go back to Edward Casey’s quote at the beginning of this article. This quote implies that ‘places’ indeed enable people to travel; to be ‘in place’ is to be somewhere, and this somewhere creates certain potential for movement. Thus, journeying is essentially about navigating places. At the same time, I have noted that migrants create a sense of place by mediating local and translocal connections in different localities along the journey. This networked place is fundamental for migrants’ movement since they enable them to connect with different mobilities possibly directing them to distant localities.

Mobilities and Sediments: Spatial Outcomes of Migrants’ Movement

It follows that places and mobilities are related. This relation is, however, not one-sided. It has been shown that places facilitate mobilities. Equally so, mobilities may bring their spatial sediments (Adey 2010); they change places. The presence of migrants and the demand for migration-related services have turned several African towns into so-called migration hubs. Migration hubs are described as crossroad places for migration because of a concentration of migration services. In these hubs, migrants can (re-)organize the next steps of their journeys. The emergence of migration industries in these different localities indicates that inward and outward mobilities transform places. The Moroccan town Oujda, for instance, has seen a recent increase of cyber-cafés and phone houses because of the fact that sub-Saharan Africans use Oujda as a transit place (Schapendonk 2011). Moreover, in Rabat African migrants have founded Pentecostal churches, created their own informal businesses, and installed their self-organizations that are sometimes embedded in transnational networks of border activists. These are all transformations in the social and physical landscapes of this Moroccan city that are directly related to new migration dynamics.

Further south along the trans-Saharan routes, the spatial effects of African migration are also visible, as the work of Brachet (2005) shows. He describes how the remote desert town of Dirkou (northern Niger) has changed from a small oasis town into a regional center as an effect of the recent increase of trans-Saharan migration to Libya (see Minvielle (2006) for a similar argument in the framework of the Algerian town Tammanrasset). The transformation of Dirkou includes the emergence of new public services, police stations, and profound changes in the regional labor market. As an effect, the town has grown considerably in size. Brachet’s case of Dirkou is a perfect example of how place-making occurs along paths of movement (Mazullo and Ingold 2010). This latter notion of spatial development along paths of mobility becomes clearer when we realize that many desert towns have a rich history because of pre-colonial trans-Saharan trade industries (Baier 1977), and industry that was based on mobility. In this respect, the economic opportunities related to migration are welcome development boosts for places that flourished a long time ago along different trans-Saharan mobility paths. In fact, it was the lack of mobility that affected the region in negative ways in times of colonial rule.⁴

However, this spatial development is only one side of the story. As with airports and train stations (Urry 2007; Adey 2010), migration hubs have also become hubs of surveillance and mobility control. This means that trans-Saharan migration is not only discovered by local engineers as an economic opportunity, but also by authority agents attempting to gain extra income by means of bribes and dubious ‘transit taxes’ (Brachet 2005; Hamood 2006; Bensaad 2007). As some authors argue, the Sahara must not be analyzed as a smooth passage space for migrant travel, but as a holding zone where sub-Saharan migrants are discouraged from continuing to move northwards (Bensaad 2007). When migrants lose their financial resources they easily become stuck in certain places which they initially aimed to pass. This indicates that the same places that facilitate migrants’ movements may also become the places that obstruct their movements.

The latter is most clearly illustrated by the case of the Spanish enclaves of Ceuta and Melilla on Moroccan territory. From the beginning of 2000, African migrants have entered these enclaves in increasing numbers. Partly because migrants have used Ceuta and Melilla as their access points to Europe, these places have transformed into strictly bordered and fortified localities that are nowadays extremely difficult to enter (Ferrer-Gallardo 2011). Thus, the spatial sediments that are related to migration do certainly not simply favor migrants on their way to their preferred destinations.

Conclusions

This article has put process geography (Appadurai 2001) into practice. I have explored a complex and interconnected understanding of migration trajectories and encouraged researchers to think more in terms of mobilities when they analyze migration. At the same time, I have argued that we should not entirely forget about the role of places in migration trajectories.

Although the examples of Dawda, Eric, and Jean-Louis could have been explained with a more fundamental complexity framework (for instance, by elaborating on the role of all kinds of material and fixed infrastructures), these examples succeed in transcending the logic of the migrant as the informed and rational actor moving according to well-weighted plans as in neo-classical explanations of migration. The three examples show that migration is often much more complicated than the question, ‘should I stay or should I go’? At the same time, it is illustrated that small events and small decisions are not necessarily dominated by macro-economic and political processes as it is put forward in many structuralist analyses. Instead, structural forces and small events all come together in places and migrants have to navigate this ‘thrown togetherness’ of places (Massey 2005) in order to move in preferred directions.

For many readers, however, the three examples above are no more than exemplifying cases in which contingency is overemphasized. They might even argue that the stepwise migration processes of sub-Saharan Africans are rather exceptional in this age of migration as most migrants do reach their destinations with rather direct journeys. This might be true. In fact, the trajectory of Dawda, the Gambian man, is such a migration process; he simply changed his country of origin (The Gambia) for a new country (The Netherlands). Traditional migration research would categorize this as an ordinary, uncomplicated, and linear movement from A to B. The interconnected mobilities lens, however, shows that this single movement consists of a sequence of movements starting with Dawda’s rural-urban migration in The Gambia. Furthermore, it includes the temporal visits of his girlfriend to The Gambia and his own temporal stay in The Netherlands and Senegal to understand his trajectory. When we omit these dynamics we again tend to overlook mobilities in general, and migrants’ journeys in particular, and approach them as residual phenomena and empty spaces. Then we would not only move in the direction of sedentarist thought; we also fail to do justice to the efforts of migrants, their aspirations that have emerged long before their departures, the en route assistance they get, and the luck they sometimes need in order to cross borders. The mobilities lens is just a tool to make sense of these complex migration processes and their (un)expected twists and turns, without claiming that the decisive element can be found in individual agents or abstract structures.

Finally, I have argued that incoming and outgoing mobilities contribute to place-making. With this relational understanding of places and mobilities I contrast the conventional geographical notion of mobilities being a threat to places – as if too much mobility leads to a certain loss of identity and function of places, as if too much mobility leads to placelessness (Cresswell 2004) and non-places (Augé 1995). By acknowledging that mobilities create and give meaning to places, we avoid a static, sedentarist, and exclusive version of places (Massey 2005) that perceive places as spatial entities with locatable and uncontested borders. Thus, a relational understanding of mobilities and places does not only imply a certain way of thinking about migration, but also about places and the borders of places. The (re-)emergence of ‘migration industries’ and ‘migration hubs’ along migratory routes across Africa is, in the framework of this study, perhaps the most convincing illustration for the argument that mobilities give meaning to places. The continuous passing by of migrants has renegotiated businesses, neighborhoods, and authority controls in North and West African countries. These places have not developed as an effect of local and internal processes, but have developed because of the constant coming and going of temporary visitors. Interestingly, these spatial sediments change places and therefore they also influence the migratory journeys of future migrants who are heading to aspired-to destinations. Some places become more attractive transit places, while other places are strictly controlled and become therefore less popular among migrants. The relation between places and mobilities forms indeed a moving landscape.