223 52 1MB
English Pages 202 Year 2022
FOOD CONNECTIONS
THE ANTHROPOLOGY OF FOOD AND NUTRITION General Editor: Helen Macbeth
Eating is something all humans must do to survive, but it is more than a biological necessity. Producing food, foraging, distributing, shopping, cooking and, of course, eating itself are all deeply inscribed as cultural acts. This series brings together the broad range of perspectives on human food, encompassing social, cultural and nutritional aspects of food habits, beliefs, choices and technologies in different regions and societies, past and present. Each volume features cross-disciplinary and international perspectives on the topic of its title. This multidisciplinary approach is particularly relevant to the study of food-related issues in the contemporary world. Volume 10
Volume 5
Food Connections: Production, Exchange and Consumption in West African Migration Maria Abranches
Researching Food Habits: Methods and Problems Edited by Helen Macbeth and Jeremy MacClancy
Volume 9
Volume 4
Food and Sustainability in the Twenty-First Century: Cross-Disciplinary Perspectives Edited by Paul Collinson, Iain Young, Lucy Antal and Helen Macbeth
Drinking: Anthropological Approaches Edited by Igor de Garine and Valerie de Garine
Volume 8
Food in Zones of Conflict: Cross-Disciplinary Perspectives Edited by Paul Collinson and Helen Macbeth Volume 7
Liquid Bread: Beer and Brewing in Cross-Cultural Perspective Edited by Wulf Schiefenhövel and Helen Macbeth Volume 6
Consuming the Inedible: Neglected Dimensions of Food Choice Edited by Jeremy MacClancy, C. Jeya Henry and Helen Macbeth
Volume 3
Food for Health, Food for Wealth: Ethnic and Gender Identities in British Iranian Communities Lynn Harbottle Volume 2
Food Preferences and Taste: Continuity and Change Edited by Helen Macbeth Volume 1
Food and the Status Quest: An Interdisciplinary Perspective Edited by Polly Wiessner and Wulf Schiefenhövel
FOOD CONNECTIONS Production, Exchange and Consumption in West African Migration
Maria Abranches
berghahn NEW YORK • OXFORD www.berghahnbooks.com
First published in 2022 by Berghahn Books www.berghahnbooks.com © 2022 Maria Abranches All rights reserved. Except for the quotation of short passages for the purposes of criticism and review, no part of this book may be reproduced in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system now known or to be invented, without written permission of the publisher. Library of Congress Cataloging-in-Publication Data Names: Abranches, Maria, author. Title: Food connections : production, exchange and consumption in West African migration / Maria Abranches. Other titles: Anthropology of food and nutrition ; 10. Description: New York : Berghahn Books, 2022. | Series: Anthropology of food and nutrition; 10 | Includes bibliographical references and index. Identifiers: LCCN 2021040511 (print) | LCCN 2021040512 (ebook) | ISBN 9781800733725 (hardback) | ISBN 9781800733732 (ebook) Subjects: LCSH: Bissau-Guineans--Food--Social aspects--Portugal. | Bissau-Guineans--Portugal--Social life and customs. | Food--Social aspects--Guinea-Bissau. | Immigrants--Portugal--Social life and customs. Classification: LCC GN407 .A47 2022 (print) | LCC GN407 (ebook) | DDC 394.1/2096657--dc23 LC record available at https://lccn.loc.gov/2021040511 LC ebook record available at https://lccn.loc.gov/2021040512 British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN 978-1-80073-372-5 hardback ISBN 978-1-80073-373-2 ebook https://doi.org/10.3167/9781800733725
CONTENTS
CONTENTS
List of Figures Acknowledgements Introduction
vi viii 1
1. Spaces of Production
23
2. Migration, Body and Adaptation: Preparing and Consuming Food away from the Land
51
3. Temporal Connections: The Making of Memories and Aspirations through Food
77
4. Transnational Exchange of Food: Gifts, Reciprocities and Trade
103
5. Food Livelihoods and New Economic Spaces: A Critique of ‘Informality’
129
Conclusion
155
References
165
Index
185
FIGURES
0.1. Map of Guinea-Bissau.
11
1.1. Agricultural swamp area divided in plots. Zona Sete, Bissau, 2010.
29
1.2. Urban smallholding. Cuntum, Guinea-Bissau, 2010.
35
1.3. Bideiras selling fresh vegetables and fruits. Caracol market, Bissau, 2010.
37
1.4. Women catching mangos at the state mango orchard. Cumura, Guinea-Bissau, 2010.
41
1.5. Dried and liquid tree parts (medicines) and animal products for amulet-making. Bissau, 2010.
45
2.1. Grilling corn in Damaia, Portugal, 2010.
58
2.2. Shipment of kola nuts. At a storehouse in Bissau, 2010.
68
2.3. Materials and tools used for amulet-making in Lisbon, 2010.
70
3.1. Tree bark and roots, palm oil, mangos, baobab fruit and velvet tamarind (dry pulp) for sale in Damaia, with a radio on top playing RDP Africa. Portugal, 2010.
94
4.1. Neusa’s agency in Bissau, 2010.
104
4.2. Foodstuffs for sale in Caracol market. Bissau, 2010.
120
Figuresvii
4.3. Small pre-arranged packages of foodstuffs for sale in Damaia, Portugal, 2010.
120
4.4. Filling and weighing containers with kaldu di mankara. Damaia, Portugal, 2010.
121
4.5. Labelled 1kg containers of kaldu di mankara for sale. Damaia, Portugal, 2010.
122
4.6. Crate of mango and foli (Guinea gum vine). Damaia, Portugal, 2010.
123
5.1. Preparing food boxes to be sent to Lisbon. Tony’s office, Bissau, 2010.
137
5.2. Food boxes ready to be sent to the airport. Tony’s office, Bissau, 2010.
138
5.3. Rossio, Lisbon, with the Palace of Independence at the back, 2021.
143
5.4. Inside Wilson’s van, near Lisbon, 2010.
150
ACKNOWLEDGEMENTS
This
book is the result of a long multi-sited journey that benefited from invaluable help received along the way. My first thanks go to the Guinean food producers, traders, consumers and others who participate in the transnational travel of food and shared so much of their lives and their foods with me, in Guinea-Bissau and in Portugal. Without their generosity this book would not have been possible. Although I do not name them in order to preserve their anonymity, if this ever reaches them, they will know who they are. A number of other Guinean friends and colleagues inspired and assisted in my research. In Portugal, Heidi Pinto generously offered me Guinean Creole lessons and helped me to prepare my fieldwork trip to Guinea-Bissau in ways that words cannot describe. I am also indebted to Zé Augusto and his family for the weekends and evenings of shared stories, to Nazaré Baticam for introducing me to the medicinal effects of food and plants, and to the Landim family for offering me my first Guinean meal – a delicious siga dish – and arranging my first accommodation in Bissau. Carlos José Mendonça ‘Alante’ changed not just the course of the rainy and gloomy day we first met with his good humour and enthusiasm, but of my whole fieldwork in Lisbon by facilitating contacts with some of the women and men who were to become key participants in my research. Scholars of Guinea-Bissau also offered invaluable assistance and insights: José Nanafé, Marina Temudo, Ramon Sarró, Carolina Höfs, Eduardo Costa Dias, Clara Saraiva, Lorenzo Bordonaro, Chiara Pussetti, Celeste Quintino and, finally, Fernando Luís Machado, who has mentored and encouraged me over many years. I would also like to thank Ana Estevens for having remained a regular sender of messages and news from our common Guinean friends in Lisbon, after my return to the UK. In Guinea-Bissau, Miriam Makeba Nicolay and her family were my first hosts, to whom I am grateful for guiding me during my first days in a new city and country, as I am to Binta Baldé, Nhalim Biai and Raimundo Có for the same reasons. I am indebted to Heidi’s family, who welcomed me like one of them, especially to João and Emília Cruz Pinto for the endless evenings of storytelling on their veranda, where I learned so much about the history of Guinea-Bissau and life in general. Isabel Garcia de Almeida (Belita) and
Acknowledgementsix
David Francisco Vera Cruz (Xikinho) were my second and longest hosts in Bissau and became a second family to me. I am grateful to them in many ways, but I particularly remember the evenings when Xikinho, now sadly passed away, gracefully asked questions about my research, offering me thoughtful suggestions and often challenging my thinking. At INEP (Instituto Nacional de Estudos e Pesquisa) I am grateful to the Director at the time, Mamadú Jao, for having institutionally hosted me in the first place, and to Miguel Barros for his friendship and support throughout my stay in Bissau and beyond. Many other friends in Bissau have made life away from home much easier: special thanks must go to Alexandre Abreu, Joana Sousa, Manuel Abrantes, Sara Guerreiro, Catarina Laranjeiro, Susana Costa, Mamae Nanafé and Zé, the taxi driver who became a friend after so many late excursions together to the airport. Equally, my research would not have been possible without the support of the Fundação para Ciência e a Tecnologia (FCT) of the Portuguese Ministry for Education and Science, which funded the project (SFRH/BD/47 395/2008). In the UK, I am thankful to James Fairhead, Katie Walsh and Russell King for the stimulating discussions and advice, as well as to many others who contribute to the wonderful intellectual environment at the University of Sussex, where this journey started. Further encouragement and motivation were received from Harry West and the other staff and students at the SOAS Food Studies Centre where I happened to knock once, and whose door was so kindly and enthusiastically opened. Both Sussex and SOAS, as well as ICS (Instituto de Ciências Sociais) in Lisbon, invited me to present my work and the feedback I received was invaluable. Finally, I am especially grateful to the students and colleagues at the University of East Anglia, which has been my home for the past seven years and has generously given me the time and space to write this book. I want to finish by extending my deepest gratitude to my family. In Portugal, my parents and sister have encouraged me along the way, even if that has meant living away from home for the past twelve years. My father, who left us too soon to see the result of this research, taught me to find the poetics of academic work, and I returned to this lesson often, when things looked more difficult. In the UK, Susanna and Richard Spall have been the most wonderful grandparents to my children, and the time they have dedicated to them has allowed me to write this book. Finally, my most heartfelt thanks go to Eva, Ikey and Noemi, for growing up patiently and wonderfully alongside this project, and to John, for always being there with his love, encouragement and support and making this such a joyful and fulfilling journey.
INTRODUCTION
Food and migration are two central domains in West African economic and social life. Both receive not only academic interest, but also media, public and political attention. However, unlike in other regional contexts (cf. Ben-Ze’ev 2004; Brown and Mussell 1984; Hage 1997; Kershen 2002; Law 2001; Petridou 2001; Ray 2004; Wilk 1999),1 they are rarely examined in combination in studies not only of West Africa specifically, but also of Africa or African migration more generally. Most often, food in Africa is considered from the perspective of nutrition, famine and food security, with only a few studies addressing food from a cultural and social position in African contexts (Devisch, de Boeck and Jonckers 1995; Flynn 2005; Froment et al. 1996; Holtzman 2006a, 2007, 2009; Osseo-Asa 2005). In anthropology, despite the growing attention that has been given to consumption and food from the 1980s onwards and the resulting advancement of the anthropology of food as a field of studies (Mintz and Du Bois 2002; Watson and Klein 2016), Africa is still largely left outside its scope. In a similar vein, existent social science scholarship on African migration (including West Africa), too vast to outline here, has, surprisingly, mostly overlooked the intimate relationship between migration and food. When both domains intersect, it is usually from the perspective of consumption in destination countries (e.g. Parveen 2017; Tuomainen 2009; Williams-Forson 2014), or of the impact migration has on food and nutrition security (e.g. Crush 2013). Instead, multi-sited research that pays attention to the role of food in connecting African migrants’ and their home-based kin’s social and economic lives is still limited (with the exception of Abranches 2013a, 2013b, 2014a, 2014b; Oliveira 2018; Renne 2007). Yet from the start of the ethnographic research that led to this book, this relationship revealed its central importance in the lives of people in GuineaBissau – a small country on the West African coast – and of those who, from there, had migrated to Portugal, in Europe. When I arrived in Portugal in May 2010, coming back from Guinea-Bissau to complete the last stage of my one-year, multi-sited fieldwork, I was introduced to Aliu, a Guinean Fula food trader in Portugal, originally from the region of Bafatá (Figure 0.1).2
2
Food Connections
He had been pointed out to me as one of the first protagonists of the story I had been following for nearly one year, and which is documented in this book – that of the regular travel of Guinean food and plants from GuineaBissau to Portugal, and of the people involved in all stages of this movement. In Bissau, his brother Samba had explained their creation of the first familybased organized supply of homeland food to the growing migrant community in Portugal, which had resulted from an absence of that food, experienced by the first arrived Guinean migrants as physically painful: My brother had this idea in Lisbon because there were other Guineans working in the construction sector, like him. Since they couldn’t eat food with the ingredients they were used to, they asked each other, ‘my brother, see if you know someone coming from Bissau and ask them to bring some badjiki [roselle leaves], even if only 5 kilos…’ If one of them managed to have some sent, he would have to share it with the others. That is how my brother decided to have a box of Guinean produce sent every week, distribute it between his colleagues and charge them for it. I met Aliu in Largo São Domingos – a square in the Lisbon downtown area of Rossio, which Guinean migrants use for socializing and exchanging homeland food, and one of the most important field sites in my ethnography (figure 5.3.).3 There, Aliu confirmed what his brother had told me in Bissau: from both countries, the two siblings had contributed to the early stages of the making of a Guinean transnational lifeworld through food.4 Their original initiative of sending and receiving food and plants across borders was intimately related not only to Aliu’s personal history of migration and an identified need and demand, but also to the larger colonial and postcolonial history of Portugal and Guinea-Bissau. This story therefore also illustrates the way in which spaces are shaped by history, memory and the materiality of the environment (Connerton 1989; Nora 1984; Stewart and Strathern 2003; Sutton 2001) – all of which are processes that can be elucidated through the study of food. Pointing to a tree nearby, Aliu explained: In the old times, the Disabled War Veterans Association was right here, behind that tree, and that’s where we started selling. I was a colonial soldier between 1973 and 1974 in Guinea-Bissau, and I came here to receive a pension. When I arrived in 1989, however, I was told that only the injured were entitled to it… But we were always hanging around here. We even had lunch at the association, people slept there… that’s how we started selling here… We saw that there was demand. People who had come to Portugal, like us, they liked those things. So, we asked our relatives there to send more food over here for us to sell… After us, other people started doing the same, and now there are many sellers, as you can see.
Introduction3
This book is primarily concerned with the effects that the activity the two brothers initiated in the early 1990s has on the economic and social lives of Guineans at home and abroad, and on the connections that it generates between both lifeworlds. As I mentioned above, the importance of food in their lived experience became evident to me after beginning fieldwork. Previously, my interest had been on Guinean transnational trade more broadly, which I wanted to explore from a material culture perspective, looking at what objects were at stake and at what they could tell us about the lives of those involved in their trade from both ends. This interest was built on the recognition that migration from Guinea-Bissau had been significantly overlooked in studies of African migration, and that the effects of group-organized transnational trade on migrants and, even more so, on their families at home, were largely ignored in the literature on migration more generally.5 Yet not long after I set off to explore the materiality of trade in the Guinean migration experience, food clearly emerged as one of the most present materials in Guineans’ lives, not just in relation to trade but to a much more comprehensive reality. It proved to be the first travelling material from a historical perspective, accompanying the first migration movements to Portugal, and the one that remains the most significant in terms of quantity and the regularity with which it travels. Moreover, as was to become clear throughout the course of my ethnography, the materiality of Guinean food and plants is embedded in a particularly intimate relationship with the land that connects all living things and the spirits of the ancestors, therefore linking the experiences of production, exchange and consumption across borders in unique ways. In light of these observations, I narrowed the focus of my research and endeavoured to understand what made food such an important material for Guineans. In order to do that, I first looked at spaces of production in Guinea-Bissau and at the relationship between people, food and the land where the crops are grown. I then followed the food’s journey from its farming sites to the local food markets of Bissau and from there to the final destination in Lisbon, where it is received and exchanged by Guinean migrants in spaces such as that described by Aliu. This provided evidence of how everyday practices and meanings related to food are as much part of the material and social world of migrants as they are of those who stay back home. As a result, it also helped to understand the role of food in bridging physical distances between people in both locations. The main premise of this book is that the interplay of migration and materials like food and plants, as well as related practices of production, distribution and consumption across borders, affect people’s lifeworlds in ways that indicate a particular investment in connections. Spaces of production in Guinea-Bissau, of food preparation and consumption in a new environment and, in and between both countries, experiences related to memories and aspirations triggered by food, acts of sharing, giving and reciprocating it, and
4
Food Connections
livelihoods based on its exchange, are all connected at a transnational level and analysed, in turn, in the different chapters of this book. While carrying out this examination, the book will follow the close-knit lives of farmers, food sellers, those who buy, pack and transport the material to the airport and distribute it upon arrival, carriers and airport staff, brokers and a number of other people involved in the process of sending and receiving Guinean food. It will provide an understanding of how these different roles are created or changed through an increased demand of homeland food with migration, how the people who perform them experience those changes, and how they remain connected across borders. As the importance of such materials and their capacity to connect derive equally from their ‘symbolic’ form (meaning) and their bodily qualities (substance), as well as from the relationships in which they are integrated, I adopt here a phenomenological-oriented approach that is focused on experience and meaning of things as equally constitutive of people’s realities.
Phenomenology and the Body: Some Brief Definitions Just as philosophical traditions of phenomenology are remarkably diverse, so are its anthropological uses. Studies of health and illness, sense of place and religion comprise some of the main areas where phenomenological-oriented approaches have been more widely used in anthropology. As I will come back to later in the conclusion, by investigating migrants’ and their home-based kin’s food-related connections with a phenomenological orientation, in this book I suggest the development of what we can call a phenomenology of food and migration, as a theoretical framework with which to explore this still emergent area of research. Ram and Houston (2015: 1) suggest a definition of anthropological phenomenology as ‘an investigation of how humans perceive, experience, and comprehend the sociable, materially assembled world that they inherit at infancy and in which they dwell’. While its uses can be of a methodological or theoretical nature, this book adopts it as a theory of perception and experience. An ‘anthropology of experience’ has been increasingly followed since the mid-1980s, when an undue focus on meaning, discourse, structural relations and political economy started to be seen as oblivious of the everyday experiences, contingencies and dilemmas that weigh on people’s lives (Desjarlais and Throop 2011: 92–93). Here I follow Ram and Houston’s (2015) approach to experience as a form of sensing and comprehending the world through our perceptions, senses, endeavours and intentions. One of the most influential contributions to phenomenological approaches in anthropology is probably the idea of body subject and the direct relationship between the human body and its world. In anthropology, the body as a site of analysis owes much to Heidegger’s (1962) conceptualization of
Introduction5
being-in-the-world and Merleau-Ponty’s (1945) emphasis on the embodied person as the subject of experience. Bourdieu’s (1977, 1990) practice theory, although situated outside the phenomenological perspective, has also been a central reference to most analytical and theoretical work on the embodied experience. Focusing on the notion of ‘habitus’ as a system of routinized dispositions that emerge out of the relation to wider objective structures, Bourdieu presents a structurally mediated mode of subjectively perceiving and appreciating the lived world. In his theory of the body, perception and thought are inculcated through activities performed in symbolically structured space and time. Bourdieu tries to merge phenomenological subjectivism with structuralist objectivism, arguing that neither alone is enough to explain social action. ‘Habitus’ therefore explains individual experience while retaining the role played by objective structures, and the concept’s use in studies of migrant transnationalism is especially helpful, since migrants’ everyday experiences are necessarily linked to the structural context in which they occur, both in socioeconomic and politico-institutional terms.6 Bourdieu’s critique of phenomenology draws on the argument that it ‘sets out to make explicit the truth of primary experience of the social world’ (1977: 3), hence unquestioning what is familiar and taking the apprehension of the world as self-evident. However, Bourdieu’s practice theory also received a number of criticisms.7 Piot (1999), for example, called attention to its Eurocentric conception of persons and the social, which betrays a self-evident economism and seems to be unaffected by history. In fact, Piot contends, how can we use conceptual terms like ‘strategy’, ‘interest’ or ‘accumulation’ of symbolic capital outside a certain cultural and political terrain based on the late twentieth century’s capitalism and the language of the individual, propriety and finance? (ibid.: 16). Despite this concern with the inapplicability of practice theory in non-western contexts, Bourdieu’s discussion of the multidimensional sensuous and corporeal qualities of human practices and things is useful for the discussion presented in this book, as it helps to situate Guinean food and plants in the interplay between objects, personal stories and broader narratives. He describes the world of objects as a book ‘read with the body, in and through the movements and displacements which define the space of objects, as much as they are defined by it’ (Bourdieu 1990: 76). Echoing Bourdieu’s theory, Connerton’s (1989) focus on bodily practice and performance influenced the theoretical focus of my examination of food-related practices, as has Weiss’s (1996) use of the concept of ‘engagement’ to capture the reciprocal interchange between people, their world and its objects. When exploring the relationship between gender and the material world, Moore (1994) also drew attention to the insufficiency of representational theories in explaining it, in the sense that meaning is interpreted by acting social beings, rather than merely inhering in symbols. The body is thus seen as the set of activities constructed in space, and embodied practices as what gives meaning to that construction (ibid.: 71). Moore’s more recent
6
Food Connections
concern with ‘hopes, desires and satisfactions’ (2011) has also been a source of inspiration in the conceptual framework that I apply in this book, due to its engagement with temporal processes and the potentialities of human agency and human subjectivity in meaning-making and, consequently, in new ways of being. In human geography, where important research on migration and transnationalism has taken place, the body has also gained a central role (Longhurst 1997; Nash 2000; Rose 1999; Thrift 1996). In the 1970s, Buttimer (1976) was already trying to stimulate a debate between phenomenological approaches and human geography, in order to enhance the connections between space and the human experience. Since then, many have argued for a turn away from language and texts towards non-representational theories, focusing on expressive embodied practices, performativity, showings and manifestations of everyday life (Thrift 1996). In studies of migration and diaspora, the notion of home has been reconceptualized in line with what I call a phenomenological orientation. Home is nowadays seen as part of a physical and cognitive cosmos made through practices and memories, rather than a static place. Disciplines like geography, anthropology, sociology, feminist, cultural and postcolonial studies explore home as a process that involves material and symbolic elements not just left behind, but also newly encountered and negotiated in a new country (Ahmed 1999; Al-Ali and Koser 2002; Blunt 2005; Brah 1996; Chapman 2001; Fortier 2000; Gardner 1993; Levitt and Waters 2002; Rapport and Dawson 1998; Robertson et al. 1994; Salih 2003). Some have addressed the concepts of home and homemaking through the study of objects and people’s embodied experiences of them, and studies that tackle the link between migration and material culture have in the meantime proliferated (Basu and Coleman 2008; Burrell 2008; Dalakoglou 2010; D’Alisera 2001; Miller 2001, 2008; Parkin 1999; Tilley 2006; Tolia-Kelly 2004). However, a focus on the body also raises other questions, such as ‘[w]hat happens to the project of “giving voice” to the marginalized, if the concern is with what cannot be expressed rather than what can?’ (Nash 2000: 662). In the study of food and migration, the corporeal and sensorial dimensions that are at the centre of both experiences, and expressed as such by migrants themselves, help to minimize this risk. As Gardner (2002: 3) has argued, transnational movement is first and foremost a series of physical events, whose effects are experienced in the body. Likewise, food that is grown, ingested, shared and exchanged connects bodies that are displaced by ‘making’ them with the same substance (Abbots 2016). Moreover, a focus on the interrelation between cognition and bodily experience helps to overcome the ‘visualist bias’ in western anthropology (Sutton 2001) and acknowledges the power of other senses, like taste and smell, demonstrating their capacity to maintain connections between places, events and people. Although some anthropological studies of food and displacement have focused on the body
Introduction7
and the senses as unit of analysis (Ben-Ze’ev 2004; Dudley 2010; Law 2001; Petridou 2001; Sutton 2001), none of these has used an approach that links people’s entwined perceptions and experiences of food to contexts of production, exchange and consumption. The phenomenological orientation I adopt here also takes into consideration the wider context in which people move. Authors like Goody (1982) and Mintz (1985) have placed history at the heart of the anthropological study of food and foodways.8 Mintz’s influential sociohistorical study of sugar, for example, demonstrates the importance of looking into historical processes when socially analysing foodways and meanings of authenticity, which are constantly subject to change, inasmuch as they are entangled in social relations rather than abstractly fixed. As this book will elucidate, the economic and social lives of Guineans who participate in the transnational circulation of food are historically shaped by the colonial past, agricultural and migration policies, unemployment and labour market deregulation and encounters with new foods and lifeworlds. As Mintz has put it, ‘the anthropology of just such homely, everyday substances may help us to clarify both how the world changes from what it was to what it may become, and how it manages at the same time to stay in certain regards very much the same’ (1985: xxvii).
Food Connections: On Transnationalism and Multi-Sited Ethnography Most research on migration, even that which focuses on the transnational experience, is ethnographically positioned at the ‘arrivals’ end of the movement only, exploring the points of departure largely from the migrant’s narrated experiences.9 Within existing multi-sited research on migration, only a few relate to food (Alvarez and Collier 1994; Cook and Harrison 2007) or West African migration (Johnson 2009; Stoller 2009), but, apart from a few exceptions (Abranches 2013a, 2013b, 2014a, 2014b; Oliveira 2018), none examines both in combination.10 Yet the fact that migrants’ and their homebased kin’s lives are connected in many ways in what Brah (1996) has named ‘diasporic space’, and that homeland food plays a key role in that connection, means that one side of the migration should not be ignored in the investigation of meanings and experiences of transnational life (Smith 2006) and food. To be at both points of departure and of arrival is, as Hannerz (2003) notices, an ideal methodological strategy in studies of migration. It is also a privileged position from which to analyse the shifting status of the ‘things’ (such as food) that make up people’s lifeworlds, as they circulate though different contexts (Appadurai 1986). In order to gain an in-depth understanding of the way in which Guineans’ transnational lifeworlds are connected across borders by people and their foods, this book is based on multi-sited ethnographic work conducted in
8
Food Connections
the country of origin – Guinea-Bissau – and that of destination – Portugal. Anthropologists like Appadurai (1996), Coleman and von Hellermann (2011), Gupta and Ferguson (1997), Hannerz (2003) and Marcus (1995), who have deeply reflected on meanings of field locations when doing ethnography with progressively less spatially bounded groups – as is the case of migration – have influenced this methodological option. Marcus, who first coined ‘multi-sited ethnography’ as a method in 1995, defined it as ‘designed around chains, paths, threads, conjunctions, or juxtapositions of locations in which the ethnographer establishes some sort of literal, physical presence, with an explicit, posited logic of association or connection among sites that in fact defined the argument of the ethnography’ (1995: 105). Within this ethnographic method, he distinguished six techniques of ‘following’: 1) the people, 2) the thing, 3) the metaphor, 4) the plot, story or allegory, 5) the life or biography and 6) the conflict. This book was methodologically inspired by the technique of tracing things (food), through which people’s movements, memories, aspirations, histories and lived experiences were also followed. More than focusing on mobility in its own right, I use it to reach the food-related materials, spaces, practices and relationships that, from both countries, are part of Guineans’ making of a transnational lifeworld, and that work to bridge physical distances and narrow ensuing ambiguities. Indeed, grounded research is needed to address the materiality of migrants’ pathways, insofar as ‘transnationals’ are not only mobile and ‘travelling through’, but also emplaced and ‘dwelling in’ places (Dunn 2010). Although it may at first appear to be a contradiction, multi-sited ethnographic research on food connections, as is presented in this book, can be done with an approach to transnationalism that focuses on the tangible everyday experiences of migrants and their kin. Recognizing the importance of grounded empirical work in studies of transnational processes is not new in disciplines like anthropology, human geography or sociology (Burawoy et al. 2000; Coleman and Collins 2006; Mitchell 1997). While Glick Schiller, Basch and Blanc-Szanton (1992) initially coined transnationalism as a multiplicity of ties (familial, economic, social, organizational, religious and political) that are maintained by migrants in both home and host societies, a grounded approach to transnationalism has actual individual experiences, rather than abstract ties, at its core. In this book, the experiences under analysis are shaped by meaningful practices centred around the sending and receiving of food and embedded in people’s memories, stories and emotions (Levitt, DeWind and Vertovec 2003). They are also shaped by specific social relations between individuals situated in what Smith and Guarnizo have named unequivocal localities at historically determined times (1998: 11). The combination of grounded transnationalism, phenomenology and multi-sited ethnography will hopefully contribute towards attenuating some of the criticisms that the multi-sited method has unsurprisingly received,
Introduction9
particularly those related to a supposed focus on transnational processes and practices rather than an actual concern with the participants in the research (Wilding 2007).11 Likewise, while avoiding a discourse of victimization, it will consider important inequalities and the influence of historical and political contexts (colonialism and neo-liberal reforms, for example) on migrants and their home-based kin’s transnational food-related practices. The obstacles Guineans encounter in their migration or return projects, livelihoods and practices of exchange, for example, all of which are shaped by the circulation of Guinean food, provide evidence to respond to a critique of transnationalism that associates it with the false idea that people can just escape state control (Friedman 2002; Kearney 1995; Waldinger and Fitzgerald 2004). Indeed, following the debate on globalization and global space as an exaggerated celebration of movement (Gupta and Ferguson 1992) and a product of practices of power and exclusions (Massey 2005), it is important not to let the often elite ideology of transnationalism blind us not just to emotional attachments to place (Jackson, Crang and Dwyer 2004: 6–7), but also to existing inequalities. However, in the context of Africa, an emphasis on local processes, or on how global processes – including those related to international migration – are interpreted and adapted locally, is key to avoid a perspective that sees the continent merely as a victim of limited opportunities and constraints (Chabal, Engel and de Haan 2007; de Bruijn, van Dijk and Gewald 2007). Detailed ethnographic accounts are, as Sanders and West (2003: 11) have put it, ‘required to enhance our appreciation of the myriad ways in which people around the world engage with globalizing processes, ranging from resistance to embrace, but including, most importantly, the vast and complex swath of strategies lying in between’. In this book, the strategies under analysis are those relating to food production, exchange and consumption in two countries of a migratory landscape. Examined with a phenomenological-oriented perspective that looks at both meanings and experiences of food, this multi-sited ethnography will reveal new ways in which lifeworlds separated by migration are reconnected.
A Historical Note on Guinea-Bissau Guinea-Bissau is a small country situated on the West African coast, whose territory officially covers 36,125 km² (including over 8,000 km² of swampy marshland and 3,200 km² of terrain periodically covered by rain). It is bordered by Senegal to the north and east and Guinea-Conakry to the east and south and has a population of about 1,900 million inhabitants (World Bank 2021). Society in today’s Guinea-Bissau is the result of the long history of migrations across West Africa, as well as wider movements through the continent, which originate from, amongst other routes, the Atlantic slave trade and the
10
Food Connections
trans-Saharan trade paths (Davidson 1966). It is this long history of movements that, alongside hostile campaigns of religious conversion and territorial conquest, makes Guinea-Bissau an extremely complex ethnic mosaic from early times up to the present day, despite its small dimensions in territory and population (Pélissier 2001a). Although numbers are not precise, around thirty main ethnic groups (excluding subdivisions) are estimated to live in the country, of which the Balanta, Fula, Mandinga, Manjaco and Pepel are the most significant numerically (Machado 2002). Far from being clear-cut distinctions, however, most ethnic categorizations are made of complex boundaries, stemming from the fluidity of past movements, transpositions and reciprocal influences of various cultural traditions, ethnic merging and subdivisions, close relations to other people of similar language groups that transcend national borders, Islamization, and colonial campaigns (Lepri 1986; Lopes 1987). The conceptual and historical subjectivities and shortcomings regarding the notion of ethnicity in Guinea-Bissau will be discussed in the next chapter, and the ways in which Guineans use their ethnic belonging in the context of food production, exchange, preparation and consumption will be shown throughout the book. For simplicity, while acknowledging the limitations of these categorizations, I will henceforth use the term ‘ethnic groups’ to refer to this diversity.12 The Fula and the Manjaco are the most represented ethnic groups amongst Guinean migrants in Portugal. They are also, alongside the Mancanha and the Mandinga, those who performed the largest migratory movements to neighbouring countries during colonial times (Carreira 1960; Carreira and Meireles 1959). The Manjaco and the Mancanha are coastal people, organized in political units known as reguladu (kingdoms), who share, with each other and with the Pepel, an affinity in terms of language, agricultural production, religious system and hierarchical sociopolitical organization. In the early twentieth century, many moved towards Senegal and the Gambia in search of better work opportunities and as an attempt to escape colonial taxation policies implemented by the Portuguese (Gable 2003; Galli and Jones 1987). The Mancanha are nowadays, alongside the Pepel – who, however, have had limited participation in international migration – the ones who farm the soils of Bissau, from where fresh vegetables and fruits are sent to the migrant community in Lisbon. The Fula and the Mandinga are Muslim people from the interior regions of the country. During the colonial period, they performed seasonal migration movements from the eastern regions of Gabu and Bafatá to Senegal and the Gambia to work in groundnut plantations, wooden craftwork, weaving and shoemaking, as well as trade. Nowadays, Fula migrants in Portugal prevail in the food trade business, selling the products that the Mancanha and Pepel women harvest in Bissau.13 Given the Fula’s predominance amongst migrant food traders, it is worth noting their history here. Dates and events that describe the first Fula’s
Introduction11
Figure 0.1. Map of Guinea-Bissau. Map No. 4063 Rev. 5, February 2018, United Nations.
12
Food Connections
arrival in what is now Guinea-Bissau are imprecise and differ in historical sources and in people’s oral narratives. In colonial accounts, the Fula, like the Mandinga, were described as foreigners and invaders (Carreira and Meireles 1959), who marginalized other Senegambian groups, pushing them towards the coast (Vigh 2006). This view of the Fula as foreigners persists, to a certain extent, in contemporary Guinea-Bissau, also because many Futa-Fula – people of a Fula subdivision who migrated from the FutaDjallon region (which nowadays includes Guinea-Conakry, Guinea-Bissau, Mali, Senegal and Sierra Leone) in the eighteenth century – still move from present Guinea-Conakry to Guinea-Bissau and identity as Fula. Finally, as Aliu’s testimony has shown, the Fula cooperated in higher numbers with the colonial power. This was due to a common interest in fighting against coastal people and to the fact that, given their clothing and writing practices, as well as their politically centralized structures and the abidance by a ‘religion of the book’, the Portuguese saw them as less ‘backward’ than the coastal people and strategically negotiated with their chiefs (Lopes 1982; Teixeira da Mota 1951). Many Fula chiefs were thus intermediaries in the relationship between the local population and colonial authorities, who aimed at extending Portuguese ruling power to the interior. Trade played a key role in the history of the region and is intimately linked to the movements and encounters described above. For centuries, before the Portuguese presence on the coast, a network of inter-societal long-distance trade had developed with strategic crossroads in Guinean territory and beyond, mainly by Mandinga rulers of the inland kingdoms who secured the trans-Saharan trade in ivory, gold and slaves (Mendy 2003). The Fula’s involvement in trade is, unlike the Mandinga’s, described in colonial accounts as a later tradition that replaced older occupations of goldsmith, blacksmith and tanner, as well as previous roles as herdsmen and cattle traders who travelled long distances in order to find pasture and water for their cattle (Carreira 1966). Despite the country’s small dimension and the cooperation of some Fula with the colonial regime, Guinea-Bissau was one of the Portuguese colonial territories to present the strongest resistance, and Portugal had not only tenuous military control but also little success in colonialist cultural impositions in the territory (Mendy 2003; Pélissier 2001b). Only from 1914, with the project of ‘integrating’, ‘civilizing’ and ‘Christianizing’ the population, did Portugal introduce a special legal system that lasted until 1961 and divided the population into the ‘indigenous’, the ‘assimilated’ and the ‘civilized’. This division was made according to criteria of linguistic competence and ‘manners and customs’, where only the ‘civilized’ were to be granted full rights (Mendy 2003; Teixeira da Mota 1948). Notwithstanding examples such as this imposed social structure, however, people in Guinea-Bissau have shown their own strategies of resistance and negotiation throughout history. Gable’s (1998) research on African
Introduction13
and Portuguese representations of each other during colonial times, and Carvalho’s (2002) analysis of Manjaco chiefs’ iconographic representations, provide several examples of the continuous negotiations that took place between colonizers and colonized in Guinea-Bissau. As Carvalho put it, ‘this relationship should be understood in its double meaning of the establishment of relations of power and dominance on the one hand, and of the creation of new symbols and significants on the other’ (2002: 94). In fact, in the long process that combined peaceful migratory waves with movements of military conquest, there was neither a clear-cut imposition of structures of domination, nor an exclusion of ancient structures of the defeated people. Instead, different elements were borrowed and permeable boundaries created. An interesting example is the creation of a distinct sociocultural group whose descendants do not consider themselves part of an ‘ethnic group’ like the others. These are known as ‘Geba Christians’ and result from the conversion of autochthonous people who congregated near the Portuguese trading outpost of the Geba river in order to benefit from available economic opportunities. Despite the risk that these historical – and contested – distinctions may reinforce difference and boundaries, they are important to acknowledge here, due to the role they play in Guineans’ connections with their food and land, as this book will reveal.
Migration from Guinea-Bissau to Portugal Guinea-Bissau declared independence from Portugal unilaterally in 1973, after a decade of liberation struggle led by the influential Amilcar Cabral, who was assassinated eight months before the country’s declaration of independence.14 Although the first flow of Guinean migration to Portugal followed the onset of the country’s independence and Portugal’s return to democracy in 1974, larger inflows occurred during the 1980s and 1990s. These movements resulted both from a search for better work opportunities and from an unstable political situation in Guinea-Bissau, which included a civil war in 1998–1999.15 Notwithstanding the rapid growth of migration to Portugal, Senegal is still thought to host the largest community of Guineans abroad, despite a lack of recent sources to confirm this. The Manjaco, in particular, continued to move to Senegal (and, for some, from there to France) in the post-independence period (Galli and Jones 1987; Machado 2002). As Aliu’s narrative revealed at the start of this introduction, Guineans’ different patterns of participation in the liberation war also influenced the earliest migration records to Portugal. People of Fula ethnic origin who, like him, had sided with the Portuguese, were denied major political roles and influential positions within the newly installed PAIGC,16 and saw some of their local chiefs assassinated. Many, then, left the country towards Portugal.
14
Food Connections
Guinean migration to Portugal had its most significant growth between 1986 and 1996, shifting from the tenth to the fourth most represented foreign nationality (Machado 2002). In 2010, at the time of my fieldwork, the Portuguese Office for National Statistics placed Guineans as the fifth largest group of the total non-EU population in Portugal, with 19,304 individuals (INE 2011), although these statistics overlook a potentially large number of undocumented migrants. From 2011, due to the consequences of the financial crisis in Portugal, net migration was negative for the first time since 1992, yet in 2018 Guineans remained the fifth largest group of non-EU nationals with 16,186 individuals (Oliveira and Gomes 2019). An initial overrepresentation of men amongst Guinean migrants pushed men like Aliu into areas of trade normally run by women, such as fresh fruit and vegetable selling. The subsequent migration of women produced a more balanced gender distribution and altered the gendered composition of spaces and practices of food exchange in Lisbon. Aliu, for example, moved from the trade of fresh produce to that of kola nuts and raw tobacco, and other, longer established men set up ‘transnational agency’ businesses or ownership of small shops. Transnational agencies, known in Creole as ajensia, are small-scale, usually family-run partnerships that include at least one relative in each country and specialize in facilitating the sending and receiving of different products – including food and plants in one direction, and remittances in the other – between migrants in Portugal and their kin in Guinea-Bissau. They will stage some of the stories narrated in this book. The representation of different ethnic groups with regard to migration is, as it is in Guinea-Bissau, difficult to estimate. The only survey conducted with Guinean migrants in Portugal, coordinated by Machado in 1995 with 400 individuals, revealed that half of those who arrived in the first half of the 1990s identified as Fula, Mandinga, Manjaco and Mancanha. It is thus probably not a coincidence that the same ethnic groups represented the majority of my research participants in Lisbon and, partly as a result of kinship ties, also in Guinea-Bissau. Amongst these longer-term migrants of different ethnic origins, the majority speak Portuguese – alongside a variety of ethnic languages – although Guinean Creole is, as in Guinea-Bissau, the most widely spoken language in everyday social and economic relations, and the one I used to communicate in the field.17 In 2010, at the time of my fieldwork, 83 per cent of the Guinean population in Portugal lived in Lisbon and its periphery (INE 2011).18 In 2018 this proportion was estimated at 75.6 per cent (GEE 2020). Rossio, introduced above by Aliu, is an important geographical space for Guinean migrants in Lisbon: it is not only a place of residence for some, but also a space of socialization and exchange of materials – mainly food – and news from home, for many. As Aliu’s narrative revealed, the establishment of the Disabled War Veterans Association in 1974 and the use of that space by the first Guineans who arrived as war veterans played a significant role in its social construction.19
Introduction15
Yet an earlier African presence in Rossio has been documented in Loude’s (2003) vivid historical ethnography, where the area is described as a vibrant place of long-established trade fairs and markets that linked city and countryside. Loude highlights how, in more recent years, Guineans took over the district that had for centuries been occupied by black slaves walking the streets with ladders, lime wash buckets and brushes, offering their lime wash painting services – a sixteenth-century slaves’ task – to the houses of the area. Today, Rossio still connects centre and suburbs, as its major train and ferry stations (where many Africans disembark daily, coming from the periphery where they live) provide evidence for.
Field Locations: People, Their Food and Its Places Rossio, which occupies a central position in this ethnography, alongside other spaces of food exchange, production and consumption in Portugal and GuineaBissau, was the field location that initiated my research. This choice meant that the ethnographic material on which this book is based did not strictly follow the direction of the movement of food that guided the research. The intention was to first become acquainted with the materiality of Guineans’ lifeworld in a setting that I was familiar with. This facilitated my access to the following (yet historically preceding) side of the food’s journey: Guinea-Bissau. By the end of the first three months in Portugal I had been able to move from Rossio to Damaia, a parish in the outskirts of Lisbon where a large number of African migrants live. In Damaia, one street in a socially deprived area of unplanned settlements is occupied by Guinean vendors of mainly food and mesinhu (‘traditional’ medicines) and became known to Guineans as ‘Bandim market’, the name of Bissau’s largest market.20 From there, I extended the fieldwork locations to other significant yet less frequently used spaces where practices of exchange and gatherings were performed, linking people and food in such performances, not only in Lisbon and its outskirts, but also between those spaces and similar ones in Guinea-Bissau. Three months later, I travelled to Bissau. Shortly after I arrived, my first host Miriam and her friend Binta asked me to read out to them the contacts they knew I carried in my notebook, which corresponded to the relatives of my research participants in Lisbon. To my surprise, Binta gave an amused laugh, followed by immediate information (including neighbourhood of residence) on most of the names I cited from my notebook. Her husband and mother, she then told me, were also migrants in Lisbon, the latter being one of the women selling food in Rossio. I was invited by the two enthusiastic women to get in Miriam’s car and look for the people on my list. During our tour, I made the first acquaintances with many of those who would later become my interlocutors and friends in Bissau. These encounters were accompanied by animated phone calls to Lisbon to acknowledge my arrival,
16
Food Connections
our encounter, and the receipt of the gifts I had brought to them as intermediary. Food continued, as it had in Lisbon, to mark such occasions, and was often offered to me in return for my help as gift carrier or for the news I brought from an imagined Europe. While travelling around the peripheral neighbourhoods of Bissau with the two women, the consequences of the rapid growth of the capital’s population (by 80 per cent between 1995 and 2005) were immediately obvious. Accounting for 25.5 per cent of the country’s population (INE 2013), nearly 90 per cent of Bissau residents live nowadays in those neighbourhoods, which stretch from the city centre where I lived to the unplanned settlements I visited regularly from that day, lacking the most basic services (MOPCUDGHU 2005). Many of those neighbourhoods were set across both sides of the Airport Road (Estrada do Aeroporto) – the main road that, leading out of the city, passes the airport of Bissau. The airport, where key food-related performances were staged three times a week, on flight days, became another important field site from the early stages of fieldwork in Bissau, as did local food markets and smallholdings in the capital and its outskirts, where most travelling produce originates. I began to follow the food on those three weekly days, starting in the smallholdings in the early morning, moving to the markets in the afternoon and to the airport of Bissau in the evening. Despite a multiplicity of other important field sites, it was Caracol food market that, like Rossio and Damaia in Lisbon, became my central field location in Bissau. Here I was honoured with the friendly title Maria di Caracol – a name that most of my research participants still address me with. Caracol, which started as a spontaneous agglomeration of sellers and has rapidly expanded since 1988 (Lourenço-Lindell 1995), is where most Guineans go to sell and buy the vegetables and fruits that will be sent in large quantities to their kin in Portugal. The majority of the smallholdings I visited were located around Bissau’s periphery and in the region of Biombo, adjacent to Bissau, since fresh vegetables and fruits require geographic proximity to the capital and its airport for a safe arrival in Portugal. Medicines and amulets made of tree or animal parts (known in Guinean Creole as mesinhu) secure particular importance for Guineans at home and abroad due to their material and spiritual properties also embedded in the land, therefore their role in connecting lifeworlds is also analysed in this book. The journey of these materials, like non-fresh food such as dried and smoked fish and seafood, kola nuts or palm oil, often start away from the urban centre. Occasionally, then, I travelled to the southern and the eastern regions of the country to meet those responsible for finding these products and transporting them to Bissau and, from there, to Lisbon. The surroundings of Bandim market in Bissau were also important field locations, as they play a central role in the relationship between food and migration, mainly due to their conglomeration of transnational agencies: the small businesses that facilitate the sending and receiving of food, and other products, between families in the two countries.
Introduction17
On the night of my return to Portugal, I had the privilege of finding several of my research participants at the airport of Bissau, as usual. Together with personal gifts wrapped with the same adhesive tape used to parcel up their regular travelling packages, I carried several food parcels to be collected and distributed at the airport of Lisbon by their relatives. As soon as I reached the arrivals terminal in Lisbon, I delivered the parcels and continued to spend time with old acquaintances and new contacts brought from GuineaBissau, benefiting from the status I had gained from having ‘experienced the land’ in Guinean soil. During the final three months, I explored further spaces of food exchange and consumption in Lisbon. The airport, in particular, unveiled important ways in which food experiences are materialized in spaces of intimate performative practice, as it had in Bissau, while being simultaneously linked to wider relationships. The material in this book results from observations, interviews, conversations and participation in the lives of Guinean women and men of different ethnic origins and age groups for the course of a year, in Guinea-Bissau and in Portugal. In line with the technique of ‘following the thing’ (Marcus 1995) that I adopted in this multi-sited ethnography, it was tracing Guinean food that led me to the people involved in making it circulate, from producers to buyers, sellers, intermediaries and consumers in the two countries.
Overview of the Book The chapters in this book follow Guinean food from its production sites in Guinea-Bissau (chapter 1), to adapted practices of food preparation and consumption on the destination side of the migration (chapter 2), foodrelated memories, imaginaries and aspirations across borders (chapter 3), transnational food gifts, reciprocities and trade (chapter 4), and food livelihoods and economic spaces in the two countries (chapter 5). Together, they merge the two ends of a migratory landscape with the various stages of the circulation of food, using multi-sited ethnography and a phenomenologicaloriented approach to transnationalism based on the experience of food, as well as its meanings and ensuing relationships. Chapter 1 presents a critical analysis of uses, meanings and politics of land in Guinea-Bissau, contextualizing it within the debate and conceptualization of land in Africa more generally, and in West Africa in particular. It discusses the relationship between land’s physical and social uses, including entwined dynamics of food production and forms of social organization. It also looks at how these dynamics are influenced by fluid notions of territory and ethnicity, historical processes of agricultural policies, and gendered practices of resource access and use, where the role of urban women farmers and food sellers within the transnational food chain is given particular attention. From the urban smallholdings where fresh vegetables are harvested to send across
18
Food Connections
borders to Portugal, the chapter moves on to discuss Guinean fruit orchards and woodland as spaces of production, where fruits and tree parts, as well as animal-based medicinal products that also circulate transnationally, originate. Here I analyse the particular land connections that these products embody, their ensuing uses and meanings, and the different roles played those who, from Guinea-Bissau, enable their circulation to spaces of migration. Chapter 2 moves on to the migration end and looks at the embodied experience of food preparation and consumption with a focus on the materiality and spirituality of food. While analysing the necessary adaptations resulting from migration, this chapter explores meanings and experiences of food in relation to health, risk and ritual. It also brings back ideas of land, ethnicity and territory and their association with specific foods, plants and related practices, and explores these based on Abarca’s (2004) concept of ‘originality’ rather than from a western concern with ‘authenticity’. It shows how Guinean migrants adapt their food-related experiences and transform that adaptation into a new lived reality, hence not necessarily questioning or lamenting a loss of the food’s authentic essence drawn from the origin land but, instead, responding creatively to the need to make sense of their transnational social world. Their new created reality hence continues to offer them protection in an unfamiliar environment, even when it includes the absence of certain foodstuffs. This chapter also reveals the roles played by those who, from both sides, help to maintain that protection. The celebration of Ramadan amongst Guinean Muslims in the two countries is used here as a transnational stage for exploring how the preparation and consumption of a special dish – moni – with which the fast is broken, helps to connect migrants and their kin across spaces. Following on from this example, the spiritual realm of food, resulting from the substance it shares with the spirits of the ancestors, is also investigated, drawing on the example of kola nuts and their ritual and healing proprieties, as well as on the local concept of djanfa – a harmful practice that the land and related objects and practices can protect against. I explore djanfa with a phenomenological approach to the wider critical debate on African witchcraft, building on Nakamura and Pels’s (2014) emphasis on the material qualities of what is often described, from a western perspective, as the invisible realm. Here I also introduce amulets as objects whose materiality, similar to food, draws on the land’s substance to provide wellbeing to migrants. The chapter therefore shows how food and other objects in Guineans’ cosmology, centred around the notion of land, mediate between spiritual power – and, often in an entwined manner, religious power – and human agency in order to help migrants and their home-based kin make sense of their world. It ultimately reveals how, in this mediation process, migrants’ lifeworlds are created through the lived experience of those materials and their changes. Chapter 3 uncovers the ways in which the past, present and future of Guineans’ transnational lifeworlds are brought together through past
Introduction19
memories and future projects made in everyday practices of exchange, consumption and preparation of food in street markets and other spaces of socialization in the two countries. Drawing on the temporal dimension of people’s lifeworlds, it addresses memories, imaginaries and aspirations of migration and return as spatial and material practices triggered by historical consciousness, and their embeddedness in complex juxtapositions of possibilities and risks. By analysing the influence of food in forming people’s understandings of their past and their projects for the future, this chapter also reveals new ways through which Guinean migrants experience and reinterpret the idioms of development, cosmopolitanism, gossip, jealousy and djanfa through food-related practices, and contextualizes this case study within a wider debate on African historical subjectivities, colonialism, modernity and mobility. Chapter 4 focuses on practices of food exchange and engages with the critical anthropological debates on gifts and commodities, the social value of food and the morality of trade. Within these debates, it adds a perspective that has so far been missing in studies of African economic anthropology and migration: one that combines both home and host societies when exploring the integration of local economies in a different socioeconomic space. The chapter looks at the obligations, reciprocity, debt, credit and trust that are part of West African structures of kinship and exchange relationships, centring the analysis on food and extending it across borders. Transnational agencies (ajensia) stage some of the stories narrated in this chapter, given the important role they play in facilitating such processes between the two countries and in reducing the distance that, in other contexts, would lead to change in the social value of the material (cf. Appadurai 1986). Those stories demonstrate that, for Guinean migrants and their home-based relations, transnational exchange practices do not always distinguish the role of food as gift or commodity, and that, whilst requiring adaptations, the social value of food does not automatically change with the process of exchange across borders. Here I draw on the debate that reconciles the economic, profit-oriented value of objects and their circulation with the reciprocity, sociability and spontaneity that are embedded in food and its exchange. The chapter also introduces new and more impersonal intermediary roles that are created in the transnational exchange. These new roles, which go beyond networks of co-ethnics and kin and are taken mostly by Chinese migrants in Portugal, do not necessarily change the social value of the food exchanged, nor do they lead to increased alienation between production and consumption. Likewise, the introduction of new spaces and new actors does not result in the loss of community-level forms of organization, even if ‘impersonal’ rather than ‘community’ forms of exchange are sometimes preferred in people’s negotiations as food traders or customers, in order to avoid the conflict and tension that might result from community obligations. The chapter therefore shows that the juxtaposition of the community and
20
Food Connections
impersonal domains of the exchange, which other economic anthropology models have proposed (Gudeman 2001), can be applied in a transnational context. It also shows that Guinean food consumers abroad maintain their connection with production, not just by adapting practices of food preparation and consumption, as explored in the previous chapters, but also through the intimate relationships that are kept between them and all other actors involved in the process of exchange in two countries, enabling lifeworlds to remain connected despite geographical distance. Finally, chapter 5 examines the transnational trade of food as a livelihood strategy for migrants and their kin at home. It draws on a critical analysis of the informal economy in both ends of the exchange, and looks at how wider structural processes at home and abroad, such as a lack of opportunities in the regular labour market and the financial crisis in Europe, have affected migrant food traders’ economic responses and strategies. The chapter introduces new spaces and actors in these activities, such as airports and the courier system with its carriers, airport staff and custom authorities, and reveals people’s own wish to remain engaged with the state on some occasions, and disengaged in others (Lourenço-Lindell 2004), in what Meagher (2013) defined as formalinformal linkages. The food markets of Rossio and Damaia in Lisbon, and Caracol in Bissau, which also illustrate these linkages, are brought back to the analysis in this chapter. With the phenomenological-oriented approach on which this book is centred, the physical market place and the market principle (Applbaum 2005) are examined here in combination, by exploring the ways in which the materiality of those spaces of exchange, including the relationships that are created in the spatial environment, influence the process of exchange. While the chapter shows that the increased demand for homeland food, resulting from the transnational element introduced in food trade activities, creates an economic opportunity for some, it also warns against celebrating any form of income growth through these self-organized channels, as neo-liberal assumptions tend to do. Rather, these channels are mainly part of people’s strategy for survival in a context of uncertainty. The chapter ends with the story of Wilson and his mother, Teresa: a narrative illustrative of the creativity and agency in livelihood strategies, which is simultaneously associated with income needs and the need for remaining connected to the land. This connection is experienced through the exchange and the relationships it generates by sharing smells, tastes and memories during Wilson’s itinerant food selling business in the periphery of Lisbon. In the concluding chapter, I position the findings of this ethnography in relation to the relatively scarce literature on food and migration in general and, in particular, within anthropology and social science research on Africa. I then suggest a ‘phenomenology of food and migration’ as a theoretical approach to the study of such an intricate and rich relationship. While food production, distribution and consumption have been investigated with a grounded historical approach to the relationship between material practices, power and meaning
Introduction21
since the work of pioneers like Mintz (1985) and Goody (1982), the combination of perceptions and experiences of food, when analysed at a transnational level, tends to focus only on the consumption end. In this book, instead, I explore meaning with a focus on the material – the food’s substance – and the spaces that, in the two countries, embody the experience of that meaning at the various stages of the food chain. I therefore conclude that although geographical distance affects people’s economic and social lives when these are set around the circulation of food, this circulation and the varied roles it creates for its participants also enables an intimate connection to be kept between both ends of the chain. While cautioning against the risks of applying a theoretical approach with western philosophical orientation to the study of Africa, the conclusion ends with a note on the possibility that other disciplines and study areas beyond anthropology (such as health and nutrition studies or development economics, which share a concern with food and migration in Africa) also make use of this guiding approach.
Notes 1. Interest in the field has resulted in a special issue of Food, Culture & Society edited by Harry West in 2011 (West 2011). 2. As the people of Guinea-Bissau do, I will refer to them and to their objects (including food) as Guineans or Guinean throughout the book. It should not be confused with the people and things of neighbouring Guinea-Conakry (République de Guinée). 3. Although the precise name of the square is Largo São Domingos, Guineans commonly refer to it as Rossio – the larger downtown area where Largo São Domingos is situated. I will use their own designation of the place and refer to it as Rossio throughout this book. Public awareness of Rossio as a space of socialization and exchange for Guineans is widespread and documented, hence the use of its real nomenclature in this book. However, all participants’ names (except those of my hosts in Bissau), as well as any information that might identify them, have been changed for anonymization purposes. 4. Lifeworld is a phenomenological notion that has had multiple philosophical definitions. Here I refer to Heidegger’s conceptualization of human beings’ inseparability from the world in which they live and exist, and from the things that surround them in everyday life (Heidegger 1962). 5. Drawing on his fieldwork from the 1990s, Machado (2002) referred to transnational food trade between Portugal and Guinea-Bissau as an incipient activity with a possible tendency to grow. 6. Vertovec (2009) offers a review of studies of ‘transnational habitus’. 7. Bourdieu and Wacquant (1992) provide a list of such critiques – mostly associated with the limited room it leaves for agency and social change – and their responses. 8. The term foodways has been used in the literature usually in relation to social and cultural culinary practices. In this book, when I use the term I am referring not only to practices of food preparation and consumption, but also of production and exchange.
22
Food Connections
9. In studies of food and migration, an exception is Abbots’ (2008) work on the relationship between Ecuadorian families separated by migration to New York, through fieldwork conducted in Ecuador. Another exception is Gaibazzi’s (2015) study of young Soninke men who remain in a Gambian village with a long history of migration. 10. In 2008, a special issue of Journal of Ethnic and Migration Studies was dedicated to transnational linkages and multi-sited lives between Africa and Europe (Grillo and Mazzucato 2008). 11. Part of the criticism of Marcus’s programme focused on its seductive character, which did not allow the first generation of multi-sited ethnographers much room for self-critical reflection. This recognition motivated a collection of essays edited by Falzon (2009), where the challenges and thrusts of multi-sited fieldwork are discussed by several scholars, including Marcus himself. 12. The controversy surrounding these designations also originates from different colonial traditions. The Fula, for example, are also known as Fulbe, Fulani or Peul, according to English or French nomenclature, just like the Mandinga are named Malinke, Mandinka or Mandigo. In this book, I will follow the most common designations used by Guineans themselves, which mainly follow Portuguese terminology. 13. In addition to these ethnic groups, who play the major roles in contemporary Guinean migration to Portugal, the Balanta are also worth mentioning. They outnumber all other groups, representing about 30 per cent of the population in Guinea-Bissau, and occupy a corridor between the Muslims of the hinterland and the people of the coast. 14. Independence was recognized by Portugal only one year later. 15. The civil war was followed by a series of political conflicts, with the assassination of the president and head of the armed forces in 2009, four coup d’états, and a constitutional crisis after the last elections of early 2020. 16. African Party for the Liberation of Guinea and Cape Verde. 17. Portuguese, despite being the official language of Guinea-Bissau, is used mainly in governmental domains and amongst the elite of Bissau but is not spoken by the majority of the population. The discrepancy between language use and education (which is still conducted in Portuguese, despite the poor command of the Portuguese language by many educators and its lack of use in everyday life) is what many see as the reason for Guinea-Bissau’s low ranking in universalized primary education, when compared with other West African countries. Data from 2010 indicated a primary schooling rate of 67.4 per cent at national level, and 23.5 per cent in secondary education. Amongst adults, illiteracy rates were estimated at 63.4 per cent in 2000 (MEPIR 2011; PNUD 2006). Most of the older participants at both ends of my ethnography were illiterate or had only received incomplete primary education, although the youngest had often completed secondary education. 18. Fieldwork in Portugal was therefore conducted in the Lisbon Metropolitan Area, which includes the city of Lisbon and its outskirts. 19. The building where the association operated until 1993 is now the Palace of Independence (Figure 5.3). 20. This street is located in a specific neighbourhood of the parish of Damaia, whose name is omitted in this book as per the participants’ request. It will therefore be referred to as Damaia.
CHAPTER 1 SPACES OF PRODUCTION
It was the end of the dry season when Mila – a Mancanha farmer whose smallholding (known in Guinean Creole as orta) was situated near the airport of Bissau – took me to see her crops. Pointing to the cultivated area where roselle leaves should be growing, she said, ‘badjki ka padi inda’ (roselle leaves haven’t ‘given birth’ yet). An unknown illness was also affecting djagatu (African eggplant) – she explained, with concern – causing a widespread death of the plant that season, and not allowing it to have fidju (children). Mila’s narrative and the language used reveal the intimate relationship between plants and social practices in Guinea-Bissau, where birth, health, death and fertility are shared life experiences, and the grammar that regulates the lives of humans, plants and animals, who originate in the same land, abides by the same set of rules (Fairhead and Leach 1994, 1996). ‘Tchon ki bo padidu nel’ (the land/soil where one is born) is an expression that Guineans often use when referring to the origin of their food, as well as to that of people. Conceptualizing land as the connection between all living things is common in other parts of Africa. Richards (1939) offered one of the first detailed anthropological essays on African land and food. According to her observations amongst the Bemba in then Northern Rhodesia, land was not measured, assessed or conceived in ‘European’ ways. Instead, the bush was seen as one unit ready to supply food, wood, ash, building material, medicines, plants or game, and the success in getting hold of those supplies depended on supernatural powers (1939: 234). To them, to be ‘lucky’ or ‘unlucky’ regarding the land’s supplies embodied a whole set of obligations and gratifications towards the chiefs and ancestors. Failing to keep these would result in a withdrawal of supernatural blessings in relation to the land. In similar ways, Temudo (2008) advanced the notion of ‘spirit of politics’ in Guinea-Bissau, according to which most people, irrespective of their Islamic or Christian influences, subscribe to a ‘basic cosmology’ where the land (tera) or soil (tchon) – both of which are often used interchangeably – are perceived
24
Food Connections
as inhabited by spiritual entities (known in Guinean Creole as Iran). The importance of the land in Guinean cosmology is thus related to the contracts that the first settlers in a territory made with these spirits, giving them the right to be called owners of that land. Sarró (2010) offered further insight into these relations in West Africa, demonstrating that for the coastal people of neighbouring Guinea-Conakry, the roles of landlords and strangers are fluid and interchangeable, since even first-comers must consider themselves strangers to the spirits of the place, ‘the ultimate autochthones of the region’ (ibid.: 235). Nala, a Catholic Mancanha farmer like Mila, first exemplified the ‘spirit of politics’ embedded in Guinean land relations when she showed me the red cloth hanging from a stick at the entrance of her orta, which was used to prevent crop theft. The red cloth was to become a common sight in other spaces of production that I visited. Symbolizing a contract with the Iran who ensured protection of the land, it discouraged theft through the threat of powerful sanctions imposed by that protective spirit. Beyond its importance as an economic resource, the land thus performs the vital function of producing life, generating relationships and enabling the spirits’ or the ancestors’ mediation of those relationships (Imbali 1992). In this chapter, I explore such intimate land relations with a phenomenological-oriented approach, which examines the ways in which different meanings and uses of land – made of lived forms and interconnections – come together in Guineans’ everyday lived experiences in spaces of food production – the spaces from where the products that travel to Europe originate. In order to understand these experiences, a brief note on land tenure organization in Guinea-Bissau, and Africa more generally, is necessary here. While a customary system based on the principle of social equity and the right to subsistence farming for all still prevails in the country,1 Guinean ‘traditional’ land tenure, as elsewhere in Africa, cannot be thought in terms of a coherent, homogenous and stable system of rules and beliefs (Lentz 2007). In fact, novel forms of land transactions have come into play and coexisted since precolonial times. In some areas, for example, Islamization challenged the existing spiritual authorities that protect the land and introduced changes in agricultural practice, even if, in the case of Guinea-Bissau, people’s basic cosmology was not radically affected and spiritual and religious domains continue to overlap amongst Islamized groups (Temudo 2009), as they do amongst the Christianized, like Mila and Nala. With colonialism, more layers of rights, institutions, perceptions, interests and strategies with respect to African land were created (Lentz 2007: 43), including in Guinea-Bissau. Writers like Chanock (1985, 1991) went as far as arguing that, in countries like Malawi and Zambia, a communal ownership of land was an ‘invention’ resulting from a joint venture between colonial administrators and the chiefs who held the land in trust, in order to serve
Spaces of Production
25
the interests of both. In West Africa, although commercial land transfers may have challenged the ‘traditional’ land sale prohibition – linked to the spiritual dangers of such transactions that could result in detriment of fertility and risks to the community – this shift has not been abrupt. In fact, vibrant land markets were developing in the West African forest belt in the second half of the nineteenth century and the beginning of the twentieth century (Lentz 2007). Moreover, there has always been a variety of often unnamed land arrangements and transfers in African agricultural societies, which are not necessarily market-oriented or based on money. They may involve, for example, long-distance seasonal migration and temporary share contracts or adjusted loans between neighbours (such as the ‘strange farmer’ pattern of Senegal, the Gambia and part of the Western Sahel, or amongst the Kenya Luo). These arrangements, in some cases, imply a certain level of restriction to planting trees or building houses, and, therefore, some contestation around notions of belonging to a territory (Shipton and Goheen 1992).
Land, Territory and Social Organization Understanding the historical relationship between land as agricultural practice, territory, belonging and specific forms of social organization in GuineaBissau, and in West Africa more generally, is essential to fully comprehend the embodied, emotional and material experiences of producing, exchanging and consuming Guinean food, and the way in which these interrelated practices transform Guineans’ economic and social life at home and abroad. Bohannan (1967) provided one of the first critiques of European views of African land through the example of the Tiv of Central Nigeria, for whom geography is seen as social organization and not as grid-type maps made up of bounded pieces of land. Considering territory in terms of social relationships in space provided the basis for lineage and grouping in Africa, and this geographical imagery differs substantially from western notions of territory and property. As Gupta and Ferguson (1992) have argued in relation to space more generally, cultures and societies do not necessarily map onto geographical territories, and space should be understood as ‘a kind of neutral grid on which cultural difference, historical memory, and societal organization are inscribed’ (ibid.: 7). In the context of the Baga of Guinea-Conakry, Sarró (2010) contended that we should get rid of the ‘map’ way of conceiving place in order to understand people’s geographical imagery, which is concerned with a common way of doing things more than with any fixed territorial boundaries. Likewise, Fairhead and Leach (1996) have shown how the physical and social uses of land are intimately related amongst the inhabitants of the Kissidougou region of Guinea-Conakry. In Guinea-Bissau, as largely in West Africa, these entwined relations between land and social organization have been addressed in relation to the
26
Food Connections
notion of ethnicity and its multiple influences, including the creation of the nation-state and the colonial experience. The nuanced ways in which people nowadays refer to their birthplace result partly from the colonial attempt to make order out of seemingly chaotic identities, and from the postcolonial inheritance of a European classification schema that assigned individuals to ethnic groups (Davidson 2003). Yet some authors question the extent to which such legacy was imposed, resisted or eventually internalized and assumed into a new subjectivity. Following Ranger’s (1983) work on the ‘invention of tradition’, Davidson contended that ‘nations need to establish a sense of the past in order to legitimate their presence in the present’ (2003: 43) and that in the context of politically unstable Guinea-Bissau, birthplace and territory have been used as markers of authenticity and belonging in what she defined as embodied ‘intrinsically-bound contradictions’ (ibid.: 52). As will become clear in this chapter, food is an important signifier of these regional or national belongings. In his analysis of the Kaabu,2 Lopes (2005) also emphasized the role played by colonialism in producing changes in the organization of space. Drawing on Amselle and M’Bokolo (1985), he contended that in precolonial times there was a skilful continuity of cultural practices, which colonialism disrupted by reinforcing differences and, in the case of the Kaabu, by imposing territorial divisions with the granting of the territory of the Gambia to England, Casamance to France and Geba-Corubal (in Guinea-Bissau) to Portugal (2005: 20). He argued that further complexity lies in the imposition of a formed state that is exogenous to Africa and that forces Africans to reunite a territorial patrimony and unify and integrate different people in a solid physical unity (Lopes 1982: 53–54). The question of the legacy of colonial borders and whether or not they should be maintained was later addressed by Heimer (2002), who challenged researchers on Africa to investigate the strength and consistency of ‘national identities’ that emerged from these imposed territorial divisions. Giving the example of Guinea-Bissau, he called attention to the fact that while a clear ‘national identity’ was revealed in the liberation war – where the history of resistance was the longest amongst the Portuguese African colonies – the postcolonial experience shows that ethnic, regional or religious identities did not lose any of their strength. Whether ethnicity existed before colonialism or is a product of colonial politics is subject to contestation. In the context of Guinea-Bissau, Jao (2003) drew on the example of the Mancanha to demonstrate how, precisely because migration precedes colonialism, the notion of ethnicity already existed in precolonial times with reference to a common land of origin, rather than being merely a colonial invention. To the Mancanha, the main conscious internal reference that distinguished them from their neighbours was, and still is, the geographical area of Bula (figure 0.1).3 Drawing on Carreira’s descriptions from the 1960s, where he observed that people’s own designations were directly related to the region, village or neighbourhood of origin
Spaces of Production
27
(Carreira 1964), Jao contended that locality, lineage or even surname might be the most important reference in the formation of ethnic identities. Going back to Bula, Farim or Bolama – territories of Mancanha settlements – was, for the Mancanha farmers of Bissau with whom I worked, a recurrent practice, usually to attend funerals of relatives or acquaintances. An occasion to meet with other Mancanha who are now dispersed across different regions, it followed the notion of tchon tchoma (the land is calling) – a return to the land of the ancestors and a reconnection of the spirit and the land after death (Gable 2006; Teixeira 2008). There are several layers of complexity and subjectivity surrounding the role played by colonialism and power relations in Africa in shaping what are now generally known as ethnic groups. What we can safely say is that, although exacerbated or altered by colonialism, relationships of power predate it. Not only did negotiations concerning new settlements between landlords and strangers start in precolonial times, but also the long history of migrations across West Africa and the wider movements through the continent bring subtleness and ambiguity to any attempt to historically identify ‘indigenous’ and ‘strangers’. This also helps to endow most ethnic categorizations that are known in today’s Guinea-Bissau with complex boundaries rather than clear-cut distinctions (Lepri 1986; Lopes 1987). We can therefore think of ‘ethnic groups’ or ‘ethnic belonging’ as imported designations that, however, refer to complex internal perceptions. If we focus on how Guineans themselves address this expression of identity, its relevance becomes evident. When introduced to me as Mancanha, Fula, Mandinga, Manjaco or other, the Guineans I met made use of the word rasa (race). In a similar way, but with reference to a national space, I was introduced to others as someone of Portuguese race. ‘Abo i di kal rasa?’ (what is your race?) was one of the most commonly asked questions in Guinean Creole that I came across while working in Guinea-Bissau and in Portugal. Moreover, I witnessed various encounters between Guineans in the two countries who, after identifying a shared ethnic affiliation (which usually meant a common region of origin and common acquaintances in the hometown or home village), began to address each other as familia or parenti (family or relative). This suggests the creation of kinship relations through geographic proximity, as Gardner (2008) has found amongst Bangladeshi migrants in the UK.4 Further complexity in such categorizations is added, however, by the differentiated ways in which Guineans themselves use ethnic nomenclature. Serifo, for example, was a Fula agency owner (ajensiador) in Guinea-Bissau – an intermediary role that consisted of organizing the process of sending and receiving food, remittances and other materials across borders for his customers. He once described Fula sub-divisions to me not as Fula-Forro, Futa-Fula and Fula-Preto, as external sources do, but as Fula-Leve (light Fula), Futa-Fula (from the Futa-Djallon region), and Fula-Forte (strong Fula) – the latter seen as the ‘pure’ Fula from Gabu (the Guinean part of the old Kaabu kingdom), like himself.
28
Food Connections
In her studies of the Mandinga, Johnson (2006) added the need to deconstruct western views of religion and ethnic belonging as two separate dimensions, since this division does not find equivalence in how people define themselves. Acknowledging the limitations of imposed categorizations, Johnson noted how Guineans often define themselves as either Musulmano (Muslim) or Criston (Christian) before providing an ethnic affiliation. Further complicating these categorizations, both Muslims and Christians maintain, though to different extents, pre-Islamic and pre-Christian practices and beliefs, as I have also found.5 As Bâ has put it, ‘[t]he symbiosis that came about [between Islam and African tradition] was so great that it is occasionally hard to make out what belongs to one tradition and what to the other’ (1981: 196). The influence of those ancient practices derives, as it does in other African contexts (cf. Bâ 1981; Gottlieb 1982), from the use of land as a social process of distribution alongside its use for agricultural production, which can provide a place for connection and return and establish continuity with the past (Ferguson 2013: 168). Here I return to the relationship that introduced this chapter – that which the land secures between people, plants, animals and spirits, following the ‘spirit of politics’ advanced by Temudo (2008). Ethnicity, despite its controversial value as a concept, can be useful to help us understand how people create their own sense of belonging to places and modes of being in the world, even if subject to continuous transformations and adaptations (Barth 1969; Eriksen 2010; Wimmer 2008). In the migration context, where connection with the land is physically disrupted and new encounters take place, understanding the significance of ethnicity in people’s daily lives gains noteworthy relevance, particularly in terms of how it may constrain or enable social action, help to make sense of the world and be an emotional and biographically grounded way of living, experiencing, perceiving and remembering (everyday) life situations (Karner 2007: 4). As Karner (ibid.: 34) put it: Is ethnicity not also about the most familiar experiences and practices that clothe people’s (early) lives, about sounds, sights and smells that surround us, become familiar, and will trigger memories whenever encountered again? Is ethnicity not also about the taste of familiar foods, the experienced rhythm of daily life, the multiple layers of meaning we detect and negotiate in our first language? Is ethnicity not also – on the level of such experiences, sensations and memories – simultaneously shared and profoundly personal? Analysed with a phenomenological-oriented approach, the stories in this book will demonstrate precisely the connection between an embodied and sensorial lived experience and a simultaneously subjective perception of belonging through food-related practices of production, exchange and consumption across borders.
Spaces of Production
29
Ethnicity, Food and Agriculture In spaces of production in the outskirts of Bissau, where the fresh fruits and vegetables that travel to Europe are grown, hence contributing to a transnational sharing of experiences, sensations and memories, harvesting is performed mainly by Pepel women whose ancestors settled long ago in Bissau,6 and by the Mancanha from the neighbouring coastal areas of Bula and Có. The Pepel, historically from Bissau, were entitled to their plots of land in the city according to customary law and to the contracts that the first settlers made with the Iran who inhabited that territory. Most of the Mancanha have, instead, either bought their plots or rented them from the local Pepel regulo (chief), to whom they pay between 3,000 and 10,000 CFA francs a year, depending on the extension of the land. During the rainy season, the Balanta are hired for rice cultivation in Bissau’s saltwater swamps, a period which was often described to me by Pepel and Mancanha farmers as tempu di Balanta (Balanta time). Life is thus organized according to dynamic agricultural practices and cultivation methods. The notion of tempu di Balanta illustrates the need to shift away from the conceptualization of ethnicity as bounded to a fixed territory to perhaps one of temporality, since people of different ethnic origins
Figure 1.1. Agricultural swamp area divided in plots. Zona Sete, Bissau, 2010. © Maria Abranches.
30
Food Connections
move geographically, often temporarily, for different purposes, such as those related to farming seasons. It is also an example of the relationship between ethnicity, agriculture and food. In fact, despite the fluid boundaries of such constructed categories, different ethnic groups share similar cultivation techniques,7 are specialized in particular products and cook distinct dishes, often dependant on the biodiversity of their tchon of origin. Tchon, which Guineans use to refer to land or soil, is therefore related to forms of social organization, agricultural practice and specific foods. Quinta, an elderly woman I met in Lisbon, explained this interplay with the example of her own life history. Her narrative emphasizes the connections between religion, ethnicity and territory – all of which are embedded in the concept of land – and their relation with food. It provides evidence for the need to get rid of a western-based linear understanding of these processes: It depends on the tribu [tribe]. Each tribu has its own dish… My father is Cape Verdean and my mother is a Christian from Bafatá. But I am fidju di Farim [daughter of Farim]. Our traditional food… We, children of Farim, what we like most is brindji di skilon [dish of fish broth with rice]. It’s a Christian’s food. The way food is used as a signifier of people’s sense of belonging – to the family, the village, the region, religion or the nation – varies depending on historical consciousness and context. The idea of a ‘national cuisine’ is more likely to dominate in the case of emergent or besieged nations (cf. Appadurai 1988; Ben-Ze’ev 2004) and, even in these cases, will often be entwined with regional and ethnic cuisines, sometimes highlighting difference through historical consciousness of the physical unities that preceded colonial territorial divisions. A sense of national belonging through food is also more common amongst migrants away from their homeland, as a result of the need to regain a sense of belonging in an unfamiliar and uncertain context. These experiences, as well as the way in which food also connects migrants to the historical consciousness of precolonial lifeworlds – such as ‘Fulaness’ or ‘Mandinganess’ – will be explored in more detail in the next chapter. Yet it is worth returning here to the perception of Fula’s history, discussed in the introduction, as nomads or invaders who settled in the Futa-Djallon region – which includes parts of present-day Guinea-Bissau and other neighbouring countries – which means they are often seen as ‘foreigners’, ‘travellers’ or ‘not children of the land’ (fidju di tera) by other Bissau-Guineans. When referring to their land of origin and related food practices, the Fula I met often meant Bafatá and Gabu – the eastern regions of Guinea-Bissau – but were also subjectively indicating the wider Futa-Djallon region as their ancestral land. People’s sense of belonging to an ethnic group is embodied not only in agricultural practice, a certain cuisine or a territory of origin, but also in particular ways of handling food, which relate to a sense of touch and a
Spaces of Production
31
way of being with the land. Fanta, a Futa-Fula woman living in Bissau who regularly bought foodstuffs to send to Lisbon, described the Mancanha’s relationship with agricultural products as one that involved more expertise and care: ‘the Mancanha’s badjiki and djagatu are better than those of the Fula from Bafatá [in the east]. The Fula know more about trading than about farming. The Fula, when buying foodstuffs, force everything into a bag and tie it. The Mancanha do it carefully in order not to damage the produce’. This view cannot be removed from a recognition of the Mancanha’s long history of farming as a way of being. Their migration, both internal and to Senegal during the nineteenth century, often followed the search for new arable land (Jao 2003), unlike other ethnic groups who specialized in different economic activities in the past, such as long-distance trade amongst the Mandinga or cattle raising amongst the Fula. Nevertheless, some Mancanha farmers in the Pepel territories of Bissau now have different aspirations, which require a detachment from agricultural work. Their participation in the transnational circulation of food from the production end generates an awareness of what life may be like at the other end and elicits particular migration aspirations. Occasionally, transformations in Mancanha women’s organization around the land incited by such aspirations led to criticism by women of other ethnic origins, who identified the Mancanha’s present unwillingness to harvest the land, as their ancestors used to do, with manifestations of laziness. As Teresa, a Manjaco woman who often travelled between Lisbon and Bissau, once put it, ‘according to tradition, the Mancanha were the ones to harvest the fields. Then the Pepel started to replace them, because the Mancanha don’t want to do it anymore’. Negative perceptions of agricultural work in Guinea-Bissau are, however, not just part of the Mancanha lifeworld, and predate the rise in activities related to the transnational exchange of food. The influence of historical processes in forming such perceptions has been analysed by Temudo and Abrantes (2013) in the case of rural Guinea-Bissau. The authors have shown that postcolonial failure in improving living conditions, the low price of agricultural products and obstacles to their transport and commercialization, as well as the perception that development funds are being usurped by the urban elite and projects’ staff, are amongst the main reasons for the negative views of agricultural-based livelihoods, especially amongst the rural youth (ibid.: 29). Yet these views are not limited to rural areas. The deterioration of living conditions amongst the urban population shapes similar perceptions in the city, where most of my participants lived, and results partly from a history of colonial state investments that generated an increasing distance between a small urban-political elite – which benefited from those investments – and the majority of the population. This disconnection was consolidated in the immediate post-independence period (Forrest 1992) and by the liberalization phase of the country’s history (Lourenço-Lindell 2004). This distance is embodied today in the geographical division of Bissau between
32
Food Connections
the urban centre (prasa), where the businesses and services that serve the small elite are concentrated, and the unplanned peripheral settlements with a lack of basic services like electricity, sewage or running water, where urban agricultural activities have expanded.
Agricultural Political Economy: From Colonialism to Liberalization Reforms In Guinea-Bissau, the Portuguese were unsuccessful in stimulating private investments as they did in other colonies, particularly in Angola and Mozambique. Large-scale imports and exports were in the hands of European trading houses, and the land concessions (pontas) of the nineteenth century – of groundnuts, followed by sugar cane and rice – had been taken mainly by Cape Verdeans, who had migrated to the mainland in order to escape a famine in the 1860s. Contracted Guinean farmers in these pontas kept their traditional production techniques and took de facto ownership of the land (Galli and Jones 1987). Colonial agricultural development actions started in 1945, introducing new varieties of rice and groundnuts, animal traction and commercial tree cropping, and diversifying food and export crops (Galli and Jones 1987: 35). However, they were hindered by insufficient government budgets, scarcely improved physical and technological conditions, paternalistic social relations and a lag in price increases, when compared with other West African countries (Galli 1995). After the start of the independence war in the 1960s, exports declined dramatically, and a new understanding of agricultural development, partly occasioned by the Green Revolution and its new technological package, went hand in hand with the suggestion of trying to eliminate the small middlemen in order to monopolize Guinean trade for Portugal. However, the incorporation of some of these new ideas in colonial practice was quickly undermined by the war. Until independence, therefore, the economy remained largely dependent on small farmers and, in general, strongly entrenched indigenous local power structures and social and economic practices outdid an infrastructurally undeveloped state (Mendy 1990). After independence in 1973, the PAIGC government pursued the nationalization of the economy and the control of agricultural trade, in some ways similar to Portuguese colonialism’s efforts to ‘nationalize’ trade (Galli 1995: 73). The economy was to be run through a network of 120 People’s Stores (Armazéns do Povo) and SOCOMIN, the government agency responsible for guaranteeing a fair exchange of agricultural products for imported goods. The intention was to channel part of the agricultural surplus into the urban sector and foster the creation and development of a thus-far hardly existent industrial sector. Yet the reality turned out to be quite different. Due to low crop purchase prices and the inadequacy of this system in covering a nation
Spaces of Production
33
of 3,600 villages, these outlet stores proved unable to offer the required diversity of products at affordable prices, and experienced increasing shortages of goods and corrupt practices, including profiteering by managers (Forrest 2002: 239; Galli and Jones 1987: 113). Consequences such as the stagnation of agriculture, impoverishment of farmers and increase of informal trade and migration were aggravated within rural communities. They were also compounded by a poor road infrastructure that hindered the transportation of goods to the city, regularly cutting villages off from commercial circuits. The liberalization and structural adjustment programme (SAP) reforms of 1987 benefited for the most part the already privileged elite owners of land concessions, rather than commercial farmers. Moreover, government policies forcing the direct exchange of imported rice with cashew nuts have also contributed to the impoverishment of farmers. These policies,8 which had first been generated by the competitive low prices of imported rice, proved unsuccessful due to a later rise in its cost and falling cashew prices. Rising levels of inflation and growth of external debt, in conjunction with a history of ill-designed economic and agricultural policies, demographic pressure and climatic changes have therefore limited commercial farming and intensified the deterioration of the population’s economic conditions. Meanwhile, as Forrest (2002: 261) put it, ‘most of the population obtained consumer goods and sold agricultural products on informal, inter-ethnic, local and longdistance trading circuits that harked back to precolonial times’. Although the penetration of capitalism into African subsistence economies through the growth of commercial agriculture and wage labour happened to a lesser degree in Guinea-Bissau when compared with other African countries, the transformations that occurred nevertheless produced particular consequences for urban women farmers like Mila and Nala. In feminist literature on Africa, an increase in commercial agriculture has been seen as generating especially negative consequences for women. Although it is recognized that these activities precede the advent of colonialism or the penetration of capitalism, they are understood to have developed in a way that allowed lower-class women only to ‘survive’ in the urban economy (Ahmed 1985; Brain 1976; Mueller 1977; Wright 1983). In GuineaBissau, for example, while men engaged in growing cash crops like cashew nuts, women had to dedicate long working hours to subsistence agriculture in order to provide food for family consumption. As brought to attention by scholars like Moore (1988: 79), however, it may be dangerous to ‘simplistically portray… men as the winners and women as the losers’, since women are not passive recipients of social change. Yet their response is not necessarily an opposition to change or new technology in food production; rather, as Wane (2014: 33) has shown in her study of contemporary indigenous food practices amongst Kenyan rural women, what they are against is a total rejection of the past that is embedded in the development paradigm, as much as it was in colonial and liberalization ideals.
34
Food Connections
In fact, under certain circumstances, the changes that occurred could also provide opportunities for capital accumulation, political influence or economic autonomy for them (Boserup 1970). One of the shortcomings of the feminization of subsistence agriculture theory was thus the dichotomy it set up between subsistence and commercial farming (Moore 1988). Women farmers like Mila and Nala, who harvest the land for subsistence but are also able to sell large quantities of their products to send to Portugal, illustrate this interplay. Indeed, in the context of Guinea-Bissau, where the majority of the population still works in agriculture, the end of state control over trade has created an opportunity for food production and small-scale trading activities in Bissau and its outskirts. Some authors have emphasized the partial success of these income-generating agricultural activities, mainly performed by women (Aguilar, Monteiro and Duarte 2001; Forrest 2002; Havik 1995), with Havik (1995: 25) stressing that: [t]o label the commercial circuits in Africa, where women stand for the majority of buyers and sellers, with the notions of ‘subsistence’, ‘informal’ or ‘parallel’ seriously stigmatized their study in Africa and other continents. Segregation according to gender patterns in the vocabulary of development has overlooked women’s role as dynamic promoters of social, cultural and political changes (my translation). Others, however, have focused on the increase in poverty, with the percentage of Guineans with a daily income below two dollars rising from 49 per cent in 1991 to 64.7 per cent in 2002, 69.3 per cent in 2010 and then slightly down to 68.4 per cent in 2020,9 and on the downgrading consequences of externally induced structural adjustment policies (LourençoLindell 2004). In order to explore the concrete ways in which these changes are experienced by women and men in Guinea-Bissau, I will now introduce the different spaces of food production – the first stage in the transnational circulation of food – and the products and actors in them.
Urban Smallholdings and Women Farmers The production of rice in the extensive swamp areas around Bissau, as well as subsistence and market-oriented smallholdings and spontaneous food markets, have expanded in the urban periphery of the city, where the landscape is nowadays made of vegetable gardens that appear continuously between built areas not far from the urban centre (figure 1.2). In peripheral neighbourhoods, which lack basic services like electricity, sewage or running water, the density of construction progressively diminishes until swamp areas used for agricultural land fully dominate the landscape (Lourenco-Lindell 1995). These agricultural activities are an example of how resource-management
Spaces of Production
35
practices have historically intersected, and continue to do so, with local and wider processes of socioeconomic change in Guinea-Bissau. If women’s urban smallholdings were initially a response to the dilemmas generated by colonial policies and structural adjustment programmes, they nowadays also respond to a new demand from the diaspora, targeting not only the local food markets of Bissau, but also those in Lisbon. Urban-grown vegetables and fruits comprise the key ingredients in the sauces (mafe) that accompany rice (bianda) – the main staple food in GuineaBissau – when meat and fish are not affordable. They are especially important for migrants, since it is often these sauces that, made with produce from Guinean land, endow exogenous food with Guinean taste and, consequently, social value. Badjiki (roselle leaves), djagatu (African eggplant), kandja (okra), malagueta (chilli) and sukulbembe (West African pepper) – as well as palm oil and groundnuts – are used to prepare sauces like kaldu branku (chilli), kaldu di tcheben (palm oil), kaldu di mankara (groundnuts) or baguitche (okra and eggplant), for example. Most Guinean vegetables that travel to Lisbon originate in Bissau and its surrounding areas. In addition to cost-effective advantages for those involved in making it circulate, the proximity of production sites to the city and its airport is of key importance, due to the need for these products to arrive fresh in Lisbon. A need for internal transportation over a longer distance would hinder the process, in a country where general electricity shortage inhibits proper refrigeration and poor road infrastructure makes transport of goods to the capital inefficient.
Figure 1.2. Urban smallholding. Cuntum, Guinea-Bissau, 2010. © Maria Abranches.
36
Food Connections
Migrants’ demand for homeland food results in an increase in the sale of larger quantities of produce. Women who harvest the land where the vegetables and fruits that travel to Lisbon grow are often envied by other urban farmers, who refer to them as being involved in bigger selling activities than their own. It is nonetheless crucial not to succumb to a romantic view of such changes, and these women do not necessarily see themselves as privileged. They are, like other women commercial farmers in Bissau, known in Guinean Creole as bideiras. Bida – the etymological origin of the term bideira – means life. Bideira thus embodies a whole notion of a particular gendered social world, referring not only to women farmers or vendors of small foodstuffs in markets, but to women who search for or make life (fasi bida). What is more, bideiras are not necessarily farmers, but also intermediaries who buy the produce from the farmers and resell it in the food markets of Bissau (Figure 1.3), adding extra complexity to the way in which money is generated and distributed. Linked to women’s income from the activity of bideira and the use of that income in the household are gender-distinct earning patterns, often translated in men’s larger but more intermittent income and women’s smaller but more regular earnings, resulting from what bideiras call bindi pikininu (small selling). Their small but regular income from commercial farming is mainly employed in basic supplies for their children and to help each other and their mothers, following practices of female solidarity commonly found in West Africa. Not without a touch of criticism towards the men’s use of money, Mila once told me: For us, women, it is just kansera [tiredness]. We can’t sit down. We have to come here [the orta and the market]; we can’t wait for our men. Because the men of today… One day they bring something home, the other day they don’t. So, we have to come here to make a little money to help our children, to pay for their school and buy them rice and clothes. And to help our mothers as well, because now they have nothing to help us with, so we have to help each other. We can’t leave it all in our husbands’ hands! As discussed above, the feminist debate from the 1970s and 1980s has called attention to these gender imbalances over workloads and income. More recently, the debate has been complicated by an awareness of women’s responsibility for the household’s food security and nutritional welfare (Pottier 2016: 159). A note on the concept of household is important to understand some of the inadequacies of food policies in this regard. When exploring gendered food budgeting issues in Africa, Pottier called attention to the need to overcome western conceptualizations of households as simple units. As African feminists have shown, a western obsession with the nuclear family and the conjugal roles of women does not resonate with the reality of extended families and more complex family relations in indigenous societies (Sudarkasa
Spaces of Production
Figure 1.3. Bideiras selling fresh vegetables and fruits. Caracol market, Bissau, 2010. © Maria Abranches.
37
38
Food Connections
1986). In Africa, not only are household and family not overlapping spheres (Guyer 1981; Whitehead 1981) but also, in Guinea-Bissau, the concepts of fogon (stove) and moransa (residential area) add extra layers in women’s lived realities. Moransa is the basic unit of social and territorial organization of most ethnic groups and refers to the residential area where the extended family of one to three generations live. Each moransa is, in turn, composed of several fogon – a relatively independent unit of organization of production, processing, consumption and distribution (Temudo 2011). As Mila explained, women bideiras’ income from selling produce in bulk to be sent abroad often has to be used to cover the needs of the extended moransa. Another new element that results from bideiras’ participation in the transnational circulation of food is their emerging migratory projects, as I will examine further in chapter 3. Being the producers of the foodstuffs that perform an ever more desired travel to Europe helps to reinforce these projects. In Caracol food market of Bissau, where bideiras sell their crops, readymade slogans announcing the freshness of the vegetables and a guaranteed safe arrival in Europe could often be heard. ‘Odja i badjiki pa Europa!’ (Look, here’s badjiki to send to Europe), as bideiras announced in the market, revealed an awareness of their vegetables’ final destination. This awareness, associated with images of a wealthier life in Europe, often strengthened the desire for migration. Involvement in the transnational exchange of food had a particular impact upon Mancanha farmers. Mila, for example, had established a close relationship of exchange with the Fula relatives of migrant traders, who bought her produce to send to Lisbon. Her conscious recognition of her vegetables’ final destination clearly played a role in intensifying her migratory projects. At the end of my first visit to her plot of land, as she accompanied me to the main road to find a local transport that could take me home, Mila asked me to help her follow her crops on their weekly journey to Portugal. ‘Can you help me?’ She asked me in an insistent tone. ‘Who is going to take me? There are no jobs here, you see? There, I could do anything! All my kandja and badjiki go to Portugal. Why can’t I make it too?’ Despite the difficulties encountered in accomplishing these projects and aspirations, urban women farmers who participate in the transnational food trade from Bissau construct their own strategies for alternative gendered scenarios, from which they can benefit (cf. De Boeck 1999). Although they often have to use their income in the extended moransa, hence obtaining limited personal gain, they assume a position of control and are able to allocate their resources by forming – and terminating, when needed – relationships with other intermediaries in the business, who depend on them to guarantee availability of Guinean food in Lisbon. This connection with the food’s final destination, which influences their imaginaries of life abroad, hence playing a key role in activating aspirations and projects of migration, is based on a web of rights and obligations with others who participate in the chain.
Spaces of Production
39
Food from Orchards and Waters In addition to fresh vegetables, there are other products that, equally important in Guineans’ foodways, also reach the migrant community in Lisbon. Since they can endure longer conservation periods, they mostly originate further from Bissau and often go through more middle levels between production and final consumption. Although that distance has a particular impact on the creation of new strategies and intermediary roles of transporting goods to the city, and on the exchange relationships established, people in these relationships remain, in many cases, connected through the land and the materiality of the exchange. Some of these foodstuffs are, like fresh vegetables, associated with particular forms of social organization, production techniques or regions of origin. Examples of such food products include palm oil and palm kernels – the main ingredients in kaldu di tcheben, one of Guinea-Bissau’s most iconic dishes – groundnuts, cashew nuts and other popular fruits such as Guinea gum vine (foli), baobab fruit (kabasera), néré (faroba) and velvet tamarind (veludo) – the last three sent as dry pulp to be prepared as juices. In addition to these, mangos are one of the most desired food products by Guineans, and the mango season in Guinea-Bissau is especially awaited, just as it is highly celebrated in Lisbon when the first fruits start to arrive, sent by those at home. When they are ready for harvesting, and the demand from the Guinean migrant community in Lisbon generates a significant increase in the total of food packages sent, the process of mango exchange becomes embedded in social and economic practices that connect income generating needs with desired tastes and smells. At the start of the season, my hosts in Bissau, for example, called their two children, residents in London, as they ate the first mango of the year. ‘It’s symbolic’, they explained, ‘we had to let them know’. The phone call intensified the two youngsters’ eagerness to eat what they called ‘the best mango in the world’. It was a temporal ritualized emotion, experienced – or awaited, in the case of migrants – every year around the same time. Their desire, however, wasn’t fulfilled until my return to the UK. When they finally ate the two mangos that I carried as gifts from their parents, they defined the fruit’s taste as ‘the taste of Guinea’. As was often the case with other foodstuffs and cooked dishes, mango was then associated with the ‘taste of the land’ – an idea materialized in the expression, widespread amongst Guineans, ‘Guine i tera sabi’ (Guinea is a tasty land). Dija was a Mancanha bideira whose activities during the mango season also illustrate the financial and material importance of this desired fruit. When I arrived at her house in Cumura, about six miles from Bissau – after several delays due to her need to attend ritual ceremonies away from home, in Mancanha land – the neighbours approached me from the backyard and, explaining that she was currently not home, invited me to wait until she returned. I sat down waiting as they prepared cashew wine by grinding the
40
Food Connections
fruit and leaving it to ferment. It was not until she arrived with her husband, their children and large basins filled with mangos that I realized where they had been. Although she had invited me to visit her vegetable orta on the other side of the main road, working at the state mango orchard (granja) had become her main activity since the start of the season. There, she and her family regularly bought mangos to resell in the streets of Bissau or directly to the transnational traders who then shipped them to Lisbon. ‘More than fifty women come every day’, she explained, excitedly, indicating the predominant role that particular fruit had taken, at that time of the year, in the lives of other bideiras like herself. After I accompanied her to her vegetable orta as planned, Dija offered me three mangos to take back home and invited me to come back the following week to visit the granja. I went back as agreed. This time we walked together to the mango orchard, talking about the different types of mango and their health benefits. ‘Mangu di faca is the preferred type amongst migrants’, Dija explained, proudly. At the granja, she introduced me to the different workers and gave details of their roles and of the process of exchange. Chefe de posto was the person in charge of the orchard. In the office, other staff managed the large scale where women queued to have their mangos weighed. At the lowest level, young boys from the surrounding tabanka (villages) were hired to climb the mango trees with a long stick and pick the fruits from the highest branches. Down by the trees, two women caught the mangos that the boys threw down with a stretched cloth held from both sides. Dija, like other women, collected and bought mangos from the state orchard for 200 CFA francs per kilo. Reselling each fruit, which weighs around half a kilo, for 150 or 200 CFA, she managed to double the money invested. That day, Dija bought more than forty mangos to sell in bulk to Teresa and Djariatu, two women involved in the transnational food trade from the Bissauan side. They would soon reach Lisbon’s informal market. The mango season illustrates an example of intermediary roles – sometimes seasonal – that are created by migrants’ demand for food, beyond those directly related to harvesting and selling one’s own produce. But mango trees also fulfil other material roles, such as providing shade. In Caracol market, women vegetable sellers were positioned around the shade of the two mango trees that occupied the area. When indicating their selling place (mesa),10 the bideiras used one or the other mango tree as a key reference point, as did their clients. On other occasions of prolonged sitting, the shade of a mango tree was, too, the preferred location, and the tree was often referred to me as a source of giving – the shade and the fruit that, during that season, ‘falls on our hands’. Other examples of intermediary roles are the transporting, packaging and carrying of the material – between locations in the city, from countryside to the city, and across borders. Food products acquired in large quantities from storehouses that sell in bulk in the city, or from the weekly markets
Figure 1.4. Women catching mangos at the state mango orchard. Cumura, Guinea-Bissau, 2010. © Maria Abranches.
Spaces of Production 41
42
Food Connections
(lumu) outside Bissau, elucidate how such roles are operationalized. Teresa, for example, travelled each week to the different lumu – the feiras grandes (big markets) that were described in colonial accounts and that rotated on a weekly basis, as they do today, between regionally strategic locations, allowing for commercial exchanges to take place between rural and urban or peri-urban areas (Carreira 1960). In Bula, where we went together one day, she bought a basket full of palm kernels to send to Lisbon, as well as kandja, badjiki and dried seafood to resell in Caracol. Storehouses materialize another intermediary space. One of the storehouses in Bissau from where products were bought to send to migrants in Lisbon was that of Inussa, who was a Susu lokatero – the name given to those responsible for the transport of rural producers’ goods to the city – from the south. He and his associates provide a good illustration of the complex interethnic mix that characterizes the southern region where they, as well as the products they brought to Bissau, came from – Tombali. Amadu, one of his associates, was Fula, and managed the storehouse. Malam was of Nalu origin, and had specialized in transporting kola nuts, abundant in his region of origin. Karamó was also Nalu, specialized in the palm oil business. When asked about the reason for such well-defined product specialization, they simply said, ‘this is where my luck is’. The organization of food-related livelihood strategies and markets, where product specialization plays an important role, will be explored in chapter 5. Yet the reference to ‘luck’ that they continued to bring forward in further conversations about their economic activities also meant that this organization, as Richards (1939) observed long ago in northern Rhodesia, is embedded in a relationship with the land’s living things that blessed their business, and should therefore not be broken. Inussa’s job as lokatero consisted of a rented truck, a storehouse where the products were kept in Bissau and an organized weekly journey to the difficult-to-access southern regions, acting as facilitator or intermediary between the producers in the south and their clients in Bissau, who would later resell these products in the local markets or in bulk to the transnational traders’ kin. In addition to palm oil and kola nuts, different species of dried or smoked fish and seafood (djafal, bagre, gandin, kuntchurbedja and oyster, amongst others) were bought from storehouses such as Inussa’s, to send to Lisbon. Although artisanal fishing is practised in the estuary of the Geba River, where the city of Bissau is situated, these urban fishing activities are not large enough to enter the transnational trade of food, and most fish that travels to Lisbon originates in other parts of the country. It therefore goes through more layers of intermediary levels than urban vegetables in the exchange chain. Yet close relationships are also established with these food products at their centre, not so much between producers and final buyers, but between the latter and intermediaries like Inussa. On one of my visits to Inussa’s storehouse in Bissau, I was introduced to Buba Skalada, whose name was a friendly diminutive associated with his task of bringing species
Spaces of Production
43
of dried and salted fish (known as skalada) from the south. As we started talking, we realized that he sold this product each week to Safiatu, one of my participants in Bissau who then shipped it to Lisbon. Soon after, over a drink in the evening, Safiatu exclaimed, unsurprised, ‘so you met Skalada already’, and the close-knit relations that link food producers, traders, consumers and a wide range of small intermediaries that operate in Guinea-Bissau and in Portugal became clear, even in the case of non-fresh produce. Despite the greater physical distance between production and final consumption in the case of these products, and the consequent need for more levels of distribution roles, the exchange relationships established between intermediaries and clients still followed trust-based arrangements comparable to those created between farmers and buyers of urban vegetables. Furthermore, just as urban vegetable production is associated with the Mancanha, certain types of fishing-related activities, for example, are also part of the social organization of specific ethnic groups and territories. An example of this is selebsonh – a small type of dried fish also known as pis Balanta (Balanta fish), since it is mainly fished and sold by women of this ethnic group, in the southern regions of Guinea-Bissau. Transnational traders and consumers in Bissau and in Lisbon are aware of this fish’s origin and its link to a particular tchon and to specific fish preparation techniques. Once, pointing at the recently arrived packages of small dried fish that her daughter had sent her from Bissau, Kadi – a Guinean food seller in Lisbon originally from Bolama Island – exclaimed with a longing smile, ‘in Bolama we have really good fish; I miss the way we used to eat it fresh. Having it dried like this is more a tradition from the south’. Kadi’s emotive reference to Bolama as her homeland, which she associated with a certain food tradition (fresh fish) more than with an ethnic group of origin, is illustrative of the power of food as a signifier of belonging. Her story also illustrates the need to understand ethnicity as a fluid concept. Having initially told me she was Fula, given her father’s Fula origin, she later laughed at the idea of ‘ethnic belonging’ and said, ‘I am Bolanhes [‘Bolamese’, an invented adjective that defines those from Bolama Island]. I am everything’. She was then not only referring to the complex ethnic mix that characterized the female lineage of her family, but also to the larger historical process entailed in the history of Bolama Island, which has been inhabited in different periods by Africans and Europeans of many different origins who never fully dominated the area (Pélissier 2001a). Unfortunately, ‘Bolama style’ fresh fish could not endure the long travel to Portugal without proper refrigerating conditions. However, the circulation of dried fish across borders also faces obstacles, mostly linked to processes of agricultural inspection, and only small quantities are able to circulate. The same obstacles are faced by plant medicines, whose healing properties make them an essential material for Guinean migrants in Lisbon to cope with the separation from Guinean land.
44
Food Connections
Woodland Medicinal Products Most Guinean plants are endowed with a variety of healing properties that are used for an array of medicinal practices. From fruit trees, for example, it is not just the fruit that is used for consumption. The seeds, leaves, branches, roots, trunk and bark have specific qualities, which, when combined with different preparation and application methods (for example boiled, dried or uncooked, either to drink or to clean the body with), fulfil specific medicinal purposes. As elsewhere in Africa, a wide range of Guinean forest products and their effects and uses have been studied anthropologically (FrazãoMoreira 2009, 2016) and identified and labelled by European and African technicians in collaboration with traditional healers and wise men (Gomes et al. 2003).11 Experiences of benefiting from different medicinal plants were varied amongst my participants, especially migrants, who saw some diseases as resulting from a change in food habits and lifestyle, usually involving an accumulation of stress and anxiety, hence requiring more specialized knowledge. The use of appropriate medicinal plants for specific treatments comprised an important part of daily conversations amongst migrants. The wide range of effects they might produce and the required knowledge that only some people possess, either inherited or learned, contribute to some variation in perceptions, which were, however, drawn together by a general experience of bodily comfort brought by these products. Like fish, woodland plant medicines are shipped and sold dried, which makes them last longer and imposes fewer restrictions on their territorial origin, by diminishing the need for proximity to Bissau. Since Fula and Mandinga migrants from the eastern regions of Guinea-Bissau play a key role in making these plants available in Lisbon through their family-based transnational trading networks, most travelling medicinal plants originate in the eastern savannah woodlands of Gabu and Bafatá, rather than the dense forests of the south. Infali was a Mandinga healer who managed a ‘pharmacy’, as the open space for mesinhu (medicines and amulets made of tree or animal parts) transactions was known in the Bandim area of Bissau where it was situated. His pharmacy was a central intermediary location in the transnational circulation of mesinhu, and much of what arrived in Lisbon was bought there. On my first visit, he gave me a tour around the square and showed me the different types of plant and animal products, explaining each associated effect. The products were brought from Bafatá and Gabu regions, either directly by specialized hunters and herbalists, or by other intermediaries who travelled to those regions to get the products and resell them in the city. Infali performed his healing job in his house at the back of the square where his pharmacy stood or, at times, on a mat outside. In his pharmacy, two different kinds of mesinhu could be found (figure 1.5). The first kind, made of tree parts, is known in Creole as mesinhu di tera
Spaces of Production
45
(medicine ‘from the land’), which brings in the power of the land and the trees that grow in it. It is especially important for migrants in a distant land, Infali explained, to alleviate the corporeal adversities brought on, amongst other things, by cold temperatures and unknown food, but also, depending on the application method, problems resulting from jealousy and threats of djanfa (a practice that, similar to sorcery, is intended to harm others). The second kind is not only made of plants but also animal products (such as wild pig tooth, lizard or hare skin, turtle shell, cowry shells, porcupine spines, gazelle, buffalo or ram horns), and used as amulet. Amulets, like plant medicines or in combination with them, can help to solve problems that go beyond health-related issues, such as work, family quarrels, intimate and emotional life, children’s success in school and, amongst migrants, problems resulting from their migrant condition, such as difficulties in obtaining official documents or work permits (Saraiva 2008: 262). They can also provide protection against djanfa or be used to express gratitude or make payments. Infali defined both types primarily as mesinhu di saude (health-related medicines) and mesinhu di asuntu (subject-related medicines), the former being more often used by kuranderu (healers), who can be men or women, and the latter by muru (Muslim diviners), usually an activity that is exclusive to men. Being a Muslim Mandinga himself, he did not initially refer to
Figure 1.5. Dried and liquid tree parts (medicines) and animal products for amulet-making. Bissau, 2010. © Maria Abranches.
46
Food Connections
the work of djambakus, who perform divination practices as the muru do, but who rely on bush spirits rather than on God and Qur’anic texts. When I asked, however, his answer was straightforward: ‘the muru’s work is valid, but the djambakus’s is not. They don’t have statutu [status]… They don’t have proper training, you see? They do what they want. The muru is real because they use the Book, and they learn how to use it in school’. Yet, as elucidated earlier in this chapter, the distinction between spiritual (related to the spirits of the ancestors) and religious domains is not always as clear-cut as Infali’s narrative might initially lead one to believe. In fact, I was first taken to Infali’s pharmacy by my friend Dembó, a Muslim Biafada who had spent the morning telling me about his family’s troubled relation with the Iran of his village of origin near Bambadinca, in the eastern region of Bafatá. Since the situation was affecting his future, especially creating obstacles to his plan of migrating to Portugal, Dembó used Infali’s skills as diviner and healer to free him from the curse. The treatment, he explained, involved reading verses from the Qur’an and using specific plants endowed with the power of the bush spirits from their land of origin. Dembó’s narrative and the different powers associated with the materials used help to reinforce the argument that although Guineans often present themselves primarily as Muslims or Christians, the ancestral spirits and their connection to a vital ancestral land generally find a space in the lives of both (Johnson 2006). Infali, too, after having claimed not to believe in the assistance that non-Muslim herbalists and healers are thought to gain from forest spirits, pointed to a man sitting nearby and introduced him to me as a renowned hunter with a sacred relation with the forest. On the one hand, the Mandinga, like other Muslim Guineans, make frequent sharp distinctions at the level of discourse between Islamic healing practices, which draw their powers from God and the Qur’an, and non-Muslim ones, which rely on the bush spirits as a source of power. On the other hand, in practice, as these narratives show, the material and spiritual properties of Guinean land, both of which are important to most Guineans, mean that the two domains overlap. Before Infali had to leave to assist a patient who had arrived in the meantime, he went inside to fetch a cardboard box with three bottles containing a dark liquid, and explained, ‘this is mesinhu for pancreas ailments. It must not be boiled. The roots stay in water for five days, and only then can the liquid be drunk. The correct dosage is one cup three times a day, for three days. If you take it for longer, it causes nausea and vomiting’. After a pause, he shed some light on the adaptations necessarily generated by migration: ‘this kind of mesinhu is difficult to ship to Europe. Customs might accept the dried roots, but not the liquid infusion. The problem is that if one sends the roots, people there might not know the exact dosage’. His nephew Quemo, who managed an ajensia (agency) nearby, was responsible for shipping, amongst other things, family gifts of mesinhu across borders. Confirming the frequent practice of sending mesinhu without prescription, he upheld his uncle’s view
Spaces of Production
47
of the problems related to the use of these medicines by migrants: ‘they should preferably be sent with instructions on how to take them, or after a phone call where the patient in Portugal learns the correct usage once the plants arrive, yet this is not always done. Only people with power can give these instructions, like my uncle’. Moreover, Infali and other specialized healers in Bissau were concerned with the fact that although some mesinhu could not be given to branku (white people), due to what they considered ‘different constitution and habits’, Guinean mesinhu traders in Lisbon might nonetheless sell them indiscriminately, in view of economic gains. Infali’s concern with migrant vendors’ possible unethical commercialization of medicinal plants could be partly related to the historical view of the Fula – who prevail in the migrant trade business – as foreigners who have long been travelling large distances in search of better business opportunities and trade, which is seen as having devalued their relationship with the land and transformed the basis of their economy, making it shift towards the notion of individual property (Silva 1953). These accounts provide evidence of the complex and entwined meanings of land for Guineans. The fact that the Fula are often not regarded as fidju di tera (children of the land) means that they are perceived as lacking that special relation with an origin soil. Rather, they are seen as seekers of a more individualized wellbeing and material wealth, historically also linked to how they are described in the literature as a nomad semi-feudal group dominated by a hierarchy of chiefs, nobles and religious men who lived from added value extorted from the lowest categories of farmers and artisans (Lopes 1982). The experiences of Fula and Mandinga food and mesinhu traders in Lisbon, and the adaptions that take place with migration, will be explored in the next chapter. *** In Guinea-Bissau, harvesting, transporting and selling food to be sent to Lisbon – and, although in smaller amounts and with less regularity, also gathering and selling woodland medicinal products with the same end – result from migrants’ need to compensate for the physical distance imposed between them and their land, which is a key part of their lifeworld. Processes of socioeconomic change, such as those that took place with colonialism and the SAP reforms, have affected agricultural development in the country, as well as the urban social and economic landscape, with an expansion of urban smallholdings and food markets. Yet more recent transformations result precisely from that migrant need, which has generated additional demand and allowed for a variety of roles to be fulfilled in spaces of production in Guinea-Bissau. Urban farmers and food sellers (who often participate in both tasks) are mostly women, who therefore play a key role in connecting Guinean migrants’ lifeworlds to their land. Urban fresh vegetables, which
48
Food Connections
women like Mila and Nala harvest and sell, are the central food product in migrants’ foodways, due to their capacity to travel regularly and to their predominance in the sauces that comprise Guinean migrants’ everyday meals. Guinean land generates not only resource-use livelihoods in cultivation areas, wild vegetation, water and forests, but also other aspects of socioeconomic life, such as shared foods and shared belonging to an ancestral origin. However, shared food-related practices in spaces of production, just like in spaces of consumption across borders, are not necessarily a result of ethnic identity. Rather, following the phenomenological orientation that I adopt in this book, they can be seen as a performance of expressions of that identity (Fortier 1999). Food, space and belonging are products of dynamic rather than static practices. They are entwined processes involving embodied, emotional and material experiences, visible in Kadi’s desire for the fresh fish of Bolama, for example, or in the celebration of the mango season amongst all Guineans. Performativity matters in the making of self through practised, dynamic and interactive spatial assemblages (Rose 1999), hence in the understanding of ethnicity and belonging through action (Bell 1999; Edensor 2002; Fortier 1999). Performativity theories, inspired by the work of Butler (1988) on gender and sexuality, and the role food plays in those performances, are relevant tools with which to analyse such processes of belonging, in particular how they change with migration, which will be the focus of the next chapter. In her work, Butler (ibid.: 521) has shown how a phenomenological theory of the gendered body requires us to look at performativity in relation to both the meaning of acts and the actual acts through which meaning is performed. As Sahlins (1985) points out in his introduction to Islands of History, in Hawaii one may become a ‘native’, i.e. by right action if, for example, eating from the same land, thus becoming part of that land’s substance (1985: xi–xii) (author’s emphasis). Watson (1983) named such performed identities as Lamarckian – as opposed to a Mendelian model of fixed traits transmitted in a succession of generations – when describing the people from Papua New Guinea’s eastern highlands. In his ethnographic account of the Tairora, Watson also stressed the role of the environment in creating identity, drawing attention to the importance of common links to the land such as food that springs from the local soil, in a performed construction of people’s identities. Following Watson’s and Sahlins’ observations, as well as the phenomenological argument that people’s everyday lived experiences and connections to the land are embedded in both meanings and experiences of food, the next chapter will look at how perceptions and practices of food preparation and consumption across borders are adapted to recreate the land and produce a new embodied experience.
Spaces of Production
49
Notes 1. Even if the Land Law and the changes introduced in 1998 define land as belonging to the State and create a market value for land, with economic utility being required when it is used by the community. 2. An old kingdom whose territory is presently divided between the Gambia, Guinea-Bissau and Senegal. 3. In today’s administrative divisions in Guinea-Bissau, Bula is known as a ‘sector’, which includes several villages. It is believed to be the ancestral origin of all the Mancanha, although there are different stories about the origin of the group. One legend says that the name Mancanha derives from Emancanha – the first son of Brahima (the son of a Fula king) and Mbula (a Mandinga princess). Because their parents belonged to enemy groups, they ran away and had their son Emancanha in a safe fertile land that is now called Bula (Fonseca 1997). 4. Relevant to this discussion are hometown associations. Amongst Guinean migrants, these associations are usually named ‘friends’, ‘relatives’ or ‘children’ of a certain region of Guinea-Bissau. Carreiro and Sangreman (2011) and Có (2004) offer some detail on their constitution and role (in Portuguese). 5. Additionally, Dias (1999) has explored the role of new evangelical churches that have more recently spread in Guinea-Bissau. 6. The city of Bissau is also known as tchon di Pepel (Pepel land). 7. Temudo has written extensively about the Balanta’s techniques of rice farming (e.g. Temudo 2011; Temudo and Abrantes 2013). 8. They included the intensification of cashew nuts production, which became the country’s primary export (Galli 1990). 9. Although the reliability of such indicators might be questioned, they were reported in the World Bank’s most recent development indicators available for poverty headcount ratio (World Bank 2021). 10. Literally meaning ‘table’, the word mesa is used, in the context of Guinean markets, to describe a small selling place that is usually rented, not always implying a physical stall but sometimes only a piece of plastic laid out on the floor or an improvised plastic crate, with the goods displayed on top (figure 1.3). 11. In the broader West African context, a six-volume revised edition of The Useful Plants of West Tropical Africa (Burkill 1985) provides a detailed description of a variety of plants and their uses.
CHAPTER 2 MIGRATION, BODY AND ADAPTATION Preparing and Consuming Food away from the Land
When I arrived at Kadi’s house in the periphery of Lisbon, I was greeted,
as usual, with enquiries about my health and my family’s. ‘Kuma di kurpu?’ (How is your body?) was normally followed by a description of the body’s condition, which can either be ‘fine’ (‘kurpu sta diritu’), ‘lacking feeling’ (‘n’ka sinti kurpu’), or ‘refusing’ (‘kurpu ka seta’), the last two indicating illness. The food Kadi was preparing for lunch could be smelled from the corridor. Before heading back to the kitchen to finish cooking, she invited me to the living room and asked an old woman whom she addressed as aunt to entertain me with stories about Guinean food. She had become a good friend and knew my interests well. The old aunt was Quinta, her brother’s mother-in-law, who was visiting from Bissau. Kadi’s brother, resident in England, had arranged a trip to Portugal for his mother-in-law to treat severe varicose veins in hospital, under a bilateral cooperation programme between the Ministries of Health in Guinea-Bissau and Portugal. Health treatment was one of the most common reasons for migration to Portugal amongst the Guineans I worked with. Most of the time their migration ended up involving longer stays than the short period envisaged by this type of visa, largely due to the need for continuity of treatment, but also to the lengthy bureaucracy involved in renewal procedures, which sometimes meant overstaying their visa. Moreover, the purpose of getting treatment in hospital often concealed more important migratory intentions, which also perpetuated the stay. Eventually, finding a job that would enable a shortterm accumulation of savings was a prevailing parallel intention, especially amongst younger migrants. Migration to receive health treatment was also frequently entwined with a sense of risk related to weakening one’s body upon arrival in Europe – a threat that seemed at first sight to contradict the
52
Food Connections
aim of improving health conditions. Although all the Guineans I met emphasized this risk, it was the elders, like Quinta, who most intensely experienced bodily weakening in an unfamiliar environment, away from the strength customarily offered by the invigorating Guinean land and its vital foodstuffs. Kadi’s lunch was a rich goat meat dish with rice, chilli, baguitche and pounded okra – made with the vegetables that her daughter had sent from Bissau and part of which she sold in Damaia. Lunch was accompanied by the television showing the afternoon news and enthusiastic conversations that involved the old aunt’s complaints about ‘European food’.1 The two women praised what they considered the more natural quality of Guinean food, due to the absence of chemical fertilizers or feedlot livestock in its production, and showed particular suspicion towards the meat they consumed in Europe, describing it as ‘unnaturally fat’. Then, suddenly, a last-minute news showed on television – a well-known Portuguese actor had died of pancreatic cancer. Shocked by that news, they followed attentively the television report on the life of the actor and his friends’ reactions to his death, occasionally sighing and expressing pity. Once the subject changed on the screen, they debated such unfortunate common cases of cancer in Europe and related them to food habits. While arguing for the need to have regular check-ups in hospital in order to avoid serious complications, they also considered that Guinean foodstuffs, as well as mesinhu di tera, played an important role in preventing or treating such illnesses. They recounted how a Portuguese woman they knew, who also sold food in Damaia, had her cancer successfully treated with mesinhu made with dried banana leaves brought from Guinea-Bissau by a fellow Guinean migrant. This example is relevant not only to understand the relationship between food, body and health for Guineans, but also to acknowledge the ways in which, in Lisbon, some Guinean plant medicines might be perceived as benefiting ‘white people’s’ bodies as well – a practice which was, as seen in the previous chapter, regarded with more suspicion by Guinean healers like Infali, in Guinea-Bissau. This chapter will draw on the sensorial experience of food and other plant- and animal-based objects and medicines to address the material, religious and spiritual negotiations and reconciliations that Guineans undertake in order to make sense of the bodily disorientation and uprooting that comes with migration.
Food, Health and ‘Originality’ Within a material culture perspective, it was the work of Douglas and Isherwood (1979) and Bourdieu (1984) – followed by the extensive work of Miller (e.g. 1994, 2001, 2008, 2012) – that brought consumption into anthropological enquiry, by regarding consumer objects as carriers of meaning that enable individuals to make sense of their social world. The relationship between the consumption of food – as an object of material culture – and
Migration, Body and Adaptation
53
migration has been at the core of a growing – although still limited – number of studies (Abbots 2016; Ben-Ze’ev 2004; Brown and Mussell 1984; Hage 1997; Kershen 2002; Law 2001; Parveen 2017; Petridou 2001; Ray 2004; Wilk 1999).2 Amongst Guinean migrants, consumption of homeland food is a way of enabling a link to be kept between the person and the land, which is essential in their lifeworld. One of the most important reasons for maintaining that connection, for migrants, is to minimize the health risks that result, conversely, from their physical separation from the land – a risk that Strathern (1988) identified amongst Vanuatuan migrant labourers in the Melanesian context, who consumed things derived from the European world. ‘European foods’, as Kadi and Quinta defined them, cause weakening of the body due to their different substance, which refers both to their constituent elements (such as the chemically manipulated means of production emphasized by the two women), and to the material and spiritual realms that, in the case of Guinean food, are part of the close relationship with the land where food is grown. Return migrants are particularly aware of the health risks and needs of those who live abroad. Echoing Quinta’s complaints, Augusto, now living in Bissau after having been resident in Portugal for nineteen years, described an intense bodily discomfort felt during his first years away, which he saw as caused by the changes in climate and food habits. He evoked the ‘absolute insufficiency of European foods’ felt upon arrival in Portugal, and recounted the corporeal adversities experienced during his final year of high school in a small town of west-central Portugal, when eating in the school canteen with other students. His constant sickness alerted the social services’ staff, who called a Guinean doctor coincidentally resident in the same town. Following medical advice, Augusto prepared a bottle of lemon juice with Guinean chilli and onions (konserva) to pour on his meals, in order to gradually get used to the new food without going through an abrupt disruption of the familiar taste and substance that sustained and nourished the body. Another example is that of Teresa, who suffered from diabetes. Despite living in Portugal since 1993, she went regularly to Guinea-Bissau, where we met, and where she regained strength. Once, she proudly recounted how she surprised her doctor in Lisbon after a four-month stay in Bissau drinking baobab fruit juice three times a day, from where she returned with her blood sugar levels significantly reduced. Kadi and Quinta also highlighted what they considered an important external recognition of the healthy proprieties of Guinean foodstuffs: Our products have vitamins that have already been confirmed by groups of Italian doctors who work in Guinea, through Caritas.3 Caritas has even created a centre in Guinea, where they teach about food practices, about what is more nutritious for children and also for adults, and they really recommend many of our products, like lalu [pounded baobab leaves], which is very rich in vitamins and very good for children. Kandja, badjiki
54
Food Connections
and djagatu are also very rich in vitamins, and djagatu, for example, is used against diabetes as well. It is proved. On the one hand, a corporeal rejection of unfamiliar foodstuffs that, as in the case of Augusto, made sickness and nausea a common reaction, indicated not just what migrants perceived as unhealthy substance but also an actual bodily disposition that they described as ‘incompatibility’ and explained as ‘our bodies can’t eat it’. On the other hand, ‘it gives us more energy’ and ‘it makes us feel better’ were commonly described embodied sensations from consuming homeland food, especially important in the case of illnesses that migrants knew they were more susceptible to suffer from. Diabetes and cardiovascular disease were some of the most cited by Guineans themselves, and their incidence amidst migrants has been recognized in the literature on nutrition as resulting partly from changes in the environment, nutrition transition and stress (Misra and Ganda 2007). Within this recognition, some studies of health and nutrition have emphasized the need to look into migrants’ own perceived medicinal value of the food they consume in order to improve health care programmes (Burns 2004; Pieroni et al. 2007). An in-depth anthropological investigation that further enquires into the perceptions and experiences of the material and spiritual substance that constitutes food for migrants, and how that materiality and spirituality are adapted, offers important insights on local knowledge and people’s own agency in dealing with the consequences of food change with migration. Indeed, the risk prevention qualities of homeland food were seen by all as an outcome of its materiality and spirituality. Yet, just as land is, in GuineaBissau, associated with a ‘spirit of politics’ (Temudo 2008) and forms of social organization rather than strict geographical boundaries, what is perceived as Guinean food amongst migrants does not always originate in Guinean territory. With migration, as this chapter will demonstrate, encounters with different foods and the lack of availability of certain products generate inevitable mutual influences and need for negotiation, especially following an initial stage of bodily rejection. Focusing on the ways in which meanings of things are transformed across borders, authors like Howes (1996) and Appadurai (1986) argue that goods can be recontextualized and endowed with different cultural meanings from those involved in the original production. Long and Villarreal (1998) also explored this by looking at how products used in Mexican cuisine can be seen as part of Mexican identity, tradition and ancestral heritage by consumers abroad, whereas at the production site the same products were instead perceived as transmitters of modernity, given their growing importance in the global market and cross-border export, which are seen as signs of modern industry. Reversely, Guinean migrants provide exogenous food with Guinean flavours through preparation techniques and the addition of particular sauces, the latter representing one of the most common adaptation strategies amongst migrants, as this chapter will show.
Migration, Body and Adaptation
55
As one of them put it, ‘even if I buy the meat… the chicken here, I have to cook it in the Guinean style. I cook a Guinean sauce with palm oil, for example. I don’t know… it has to be spicy’. Here I draw on Abarca’s (2004) suggested use of ‘originality’ rather than ‘authenticity’ in the case of food, since it is, as other expressions of culture also are, always changing through creative adaptations that result from encounters with new foods. While authenticity may be an important emotional and political issue for some (Linnekin 1991), for example in the case of indigenous movements, where it helps provide leaders with authority to speak and be heard (Jackson 2019), the dualism authentic/false is a reductionist one. In fact, as Linnekin (1991: 447) also argued, all traditions are invented – western and indigenous – as they are ongoing human creations that ‘reflect contemporary concerns and purposes rather than a passively inherited legacy’. When debating cultures, Sahlins (1999: 411) contended that the notion of authenticity is contrary to the human social condition, where ‘no people are the sole or even principal author of their own existence’, and criticized the idea of ‘loss of authenticity’ as a legacy of bourgeois self-consciousness. In fact, although discourses and practices of authenticity can fulfil political projects such as the indigenous movements mentioned above, the Guineans I worked with rarely referred to ‘loss’ even when food adaptations were necessary, instead incorporating those adaptations in their ‘original’ foodways after the more severely adverse bodily experiences as newly arrived migrants. Moreover, they intentionally consumed ‘typical Portuguese’ meals on occasion, for example codfish at Christmas, therefore also showing that people negotiate what is important to them and that the presence of western goods in small-scale societies can be integrated in people’s choices rather than being a risk to cultural traditions or meaningful social relations (Howes 1996; Wilk 1999). Authenticity is, as Spooner (1986) pointed out, not so much related to genuineness but to the interpretation and experience of genuineness. The stories presented in the remainder of this chapter contribute to the anthropological debate around people’s role as active agents in instilling their own meaning into objects (Appadurai 1986; Howes 1996) and striving to create ‘material and social routines and patterns which give order, meaning and often moral adjudication to their lives’ (Miller 2008: 296). While food is at the centre of these stories, other objects of the Guinean lifeworld will be introduced here – amulets in particular – due to their material importance which is equally derived from the land, the animal origin of their constitution, and their similar need for adaptation with migration. The materiality and spirituality embodied in these objects, as well as in certain foods, generate experiences of bodily protection that are particular important in cases where the body has been weakened by djanfa – a practice intended to harm others that will also be discussed here and situated within a wider debate on African ‘witchcraft’.
56
Food Connections
Migration and New Food Encounters Although a large quantity of Guinean vegetables, dried fish and seafood, fruits and medicinal plants travel from Guinea-Bissau to Lisbon, destined for trade or as a gift from Bissau-based kin, new products found in Portugal can also be used by Guineans with similar protective intentions. Together, they establish a mutually constitutive relationship that allows for the importance of the land to be pragmatically negotiated. Produce of what was of exogenous origin to me could be considered Guinean food to my participants, reflecting the way in which a search for authenticity might ignore people’s complex interpretations of the value of things (Appadurai 1986; Howes 1996) and overlook the many ways in which they not only interpret but actually create new foodways critically and creatively.4 A recurrent example of the negotiations that take place with new food encounters is materialized in fish. Bentana (Guinean tilapia) and bagri (Guinean sea catfish) are widely consumed fish species in Guinea-Bissau. In Lisbon, however, similar fish is mostly acquired frozen in fish wholesale markets in the outskirts of the city, or fresh from Cape Verdean women peddlers, who obtain it from Portuguese vendors in wholesale facilities.5 The uncertainty of the fish’s origin – often, in fact, farmed in aquaculture – was, however, not important for Guinean migrants, who appeared indifferent towards its source and always called it ‘Guinean fish’. Cook and Harrison’s (2007) pioneer study of the connections between production and consumption across borders found similar experiences amongst Caribbean migrant consumers of ‘West Indian’ hot pepper sauce in North London, for whom the exact origin of the hot peppers they used did not seem to matter. For Guineans in Lisbon, the sauces with which fish would be cooked and the technique of smoking it, widely used in Bissau and reproduced in Lisbon, as well as the practice of sharing meals, were strategies used to remake the taste and value of the food in a way that was consistent with their reordered material world. Carlos was one of the many Guineans who specialized in smoking the fish that, amongst the migrant community, continued to be known as bagri, although the species may vary slightly. When I met him in the fumeiro (smokehouse) he used for his practice of smoking fish – a courtyard behind his house in the outskirts of Lisbon – others were gathered around occupying old sofas, chairs and tables that equipped the poorly paved area. They were mostly men, all Guinean residents in the neighbourhood. The strong smell emanating from the fumeiro reproduced the familiar sensations from their neighbourhoods in Guinea-Bissau, as did the vivid sounds of their enthusiastic conversations while sharing some of the fish that could be spared before Carlos took the rest to sell. Maize undergoes a similar process of ‘original’ adaptation (Abarca 2004) and reveals, additionally, the fluidity and multiple influences entwined in the
Migration, Body and Adaptation
57
concepts of ethnicity and territory that were explored in spaces of production in the previous chapter. Unlike fish and vegetables, whose consumption is prevalent amongst all Guineans, maize is, alongside millet and sorghum, one of the key crops in the diet of the Fula from the eastern regions of Guinea-Bissau. It is therefore a homeland product that assumes particular importance in contemporary ethnic identities, making them comprehensible and worthwhile by endowing them with expression and value (Fardon 2008: 235). Hence, while the widespread adaptations of fish for Guineans might be linked to what Ben-Ze’ev (2004) considered, amongst Palestinian refugees, to be the need to retrieve memories from a wider context, maize is a signifier of regional differences and its adaptation is therefore experienced in particularly performative ways. Like fish, the maize grains (corn) that are available in Lisbon do not originate in Guinea-Bissau, or in the Fula eastern territories of the country. Rather, they are locally produced and usually bought from Portuguese or Cape Verdean women sellers before being grilled by Guinean Fula women on charcoal grills and resold within the community. As with fish, it is the reproduction of familiar preparation techniques that endows corn with Guinean, or Fula, taste. However, in the case of corn, Fula women fulfil the intermediary role of buying the produce to resell amongst other Fula, which strengthens the sense of belonging to a place and a mode of being in the world – their ‘Fulaness’. Moreover, its consumption is associated with everyday community practices such as open-air grilling in spaces of exchange, consumption and conviviality shared by other Fula. The sweet aroma wafting from the grill, as the older women in their wax prints received the corn from the vendors with the unhurried rhythm and motion of their bodies, relieved them of the tiredness and estrangement of their lives as migrants. Its hot taste, they explained, recreated in their bodies the experience of the dry savannah regions of Eastern Guinea-Bissau where they were born. Conversations in those moments were often about home-based relatives who still grow this food crop in their villages of origin. Younger generations of Guinean women and men, most of whom were born in Bissau, listened to their stories while sharing the sweet and hot taste and smell of corn. Other key grains in the Fula diet, such as fonio6 or millet, are, when available, usually brought from France by Fula traders involved in wider transnational networks. The uncertain provenance of these products was, like the ones just described, not questioned by the Guinean Fula migrants I encountered. Dála, for example, was once sitting in her regular selling place when another woman placed a large plastic bag filled with millet by her feet. She turned to me with a smile, as she anticipated the occasion, that evening, when she would pound it to make flour for her meals. ‘It comes from France’, she explained, as she enthusiastically praised the other woman for an enterprise that allowed a larger quantity of African products to be available in Lisbon. Migrant women assume here, as do the bideiras in Bissau through their
58
Food Connections
relationship with other intermediaries (chapter 1), a position of control by transforming food preparation and exchange ‘into a creative and emotionally powerful agent of connection’ (Counihan 2009: 124). The phenomenological perspective that I adopt here brings up the need to recognize that people do not understand and experience things as two separate realities, as is often assumed. Instead, in the process of understanding or making meaning of their new experiences in Lisbon, Guinean migrants are actually creating an experienced reality that makes sense to them. As Parveen (2017) also found amongst South Asian women in England, homeland recipes are adapted depending on what is more readily available. While a different ingredient may risk a change of taste in some dishes, women’s competence in preparing them does not compromise the authority of both cook and food. Amongst Guineans, too, the practice of preparing and sharing food produces homely emotions that remind migrants of the village – such as the heat sensation provided by the grilled corn or the smell of smoked fish from Carlos’s fumeiro – and help to make sense of a transnational world. Unlike the emergence of an Indian or Palestinian cuisine, both of which are entwined with regional and ethnic foods, but preferred as an expression of cosmopolitanism in the case of the former (Appadurai 1988), and as a way to avoid inappropriate local patriotisms in the case of the latter (Ben-Ze’ev 2004), Fula consumption practices suggest that regional and ethnic dishes and identities are important for Guineans, as they are amongst other African migrants elsewhere (cf. Williams-Forson 2014). However, the
Figure 2.1. Grilling corn in Damaia, Portugal, 2010. © Maria Abranches.
Migration, Body and Adaptation
59
notion of ‘Guinean food’ is equally important, and both Fula and Guinean food are creatively and originally recreated through the use of exogenous land, just like ethnic categories are themselves constructed through multiple influences and transpositions. Moreover, most Fula migrants in Lisbon had migrated to Bissau several years before, and many of their home-based kin now lived in the capital, having been pulled away from the countryside due to their migrant kin’s investments, especially in house building, in Bissau. This also creates further layers in the distance imposed between migrants and their home-based kin and land. Another example of such negotiations is that performed by Sali, a Guinean restaurant owner in downtown Rossio, who specialized in cooking and selling yams outside in the late afternoon, when her work in the restaurant slowed down. Despite buying the yams from a wholesale market on the outskirts of Lisbon, the act of distributing and sharing that familiar food with other Guineans compensates for the uncertainty of the produce’s origin. Representing, above all, an obvious strategy of accumulating savings for Sali, this also enabled her to contribute to reinforcing community ties through her practice of yam exchange. Cidália, an enthusiastic dweller in spaces of socialization and Guinean food exchange and consumption in Lisbon, narrated her participation in the annual city festivities (Festas de Lisboa) as an opportunity to sell her Guinean meat cakes and longuisa (a thin kind of sausage) to a wider clientele. Prepared with a combination of local meat and Guinean herbs and chilli, she proudly described them as ‘a success amongst the Portuguese’, illustrating the process of embedding Guinean food in wider social and cultural landscapes. Simultaneously, she recounted her own consumption of ‘typical’ Portuguese sardines and bifanas (pork sandwiches) – the trademark of these festivities that take place across Lisbon’s bairros (neighbourhoods) during the month of June. ‘In my bairro I always participate in the festas. I drink beer and eat sardines and bifanas. I love it!’, she started by saying. Later, however, alluding to the use of Guinean sauces, she rectified her taste for ‘Portuguese food’ at the festas, and added, ‘I have to season the bifanas my way, though, because here they only use salt’. She described adding the spicy and sour taste of chilli and lemon as an embodied and sensorial part of cooking – the act that will bring pleasure to the experience of eating, as well as a sensation of certainty that the body needs in order to feel healthy and safe. The relationship between food and land assumes its utmost disconnection when it comes to the large-scale industrialized foods that Guinean migrants generally name ‘European’ or ‘white people’s’ food – such as the ‘European meat’ that Quinta alluded to with concern about its effects on health. Industrialized foods do not rigidly correspond to a point on a map. Guineans’ notion of ‘European foods’ is built upon the idea of industrialized processes of food production where consumers are physically distanced and
60
Food Connections
relationally detached from the food’s soil of origin, which is often undistinguished, more than it is linked to any idea of ‘European land’. Holtzman (2009) found similar views amongst the Samburu in Kenya, to whom purchased foods – also called ‘gray foods’ – were considered of an indistinct blend and vague colour, devoid of the qualities of customary pastoral foods. However, for them, it was the preparation and consumption techniques of ‘gray foods’ around a single cooking fire, more than the food itself, that disrupted a gender-age system of separate eating habits based on respect and shame, and therefore deteriorated relationships. For Guinean migrants, instead, it was the possibility of maintaining familiar preparation and eating practices that made ‘European foods’ more tolerable. In Guineans’ commensality practices, rice – the main staple food in GuineaBissau and the key ingredient in Guinean dishes – offers an interesting example of processes of adaptation that take place both at home and abroad. In Lisbon, rice is normally bought in Chinese shops at more affordable prices and comes predominantly from Thailand, in 25-kilo bags. In Guinea-Bissau, too, rice is largely imported, following ill-designed governmental policies that resulted in, amongst other effects, inefficient circuits of commercialization of local rice and the rise of cheap imported rice (Temudo and Abrantes 2013). Nevertheless, given that ‘Guinean land’ is not necessarily bound to political borders, rice continues to act as a marker of national cuisine. Indeed, since Guinean migrants’ practices of food preparation and consumption embed the exogenous substance of rice, corn, millet, fonio, bentana or bagri into their everyday lived experiences, these foods remain an expression of national or ethnic land for them. Transcending a preoccupation with authenticity and looking, instead, at the ingenuity with which migrants respond to their everyday needs and deprivation is essential to understand what really matters to people. Moreover, although these products are technologically different from the production processes associated with home, some (such as fonio and millet sachets, or frozen fish) help to save preparation time, as Renne (2007: 621) also found amongst West Africans in the United States, where time constraints impose limits on choice. Maimuna, a Fula woman originally from Gabu and a seller of Guinean food with the help of her daughter from Bissau, illustrates the inadequacy of imposing well-defined boundaries between Guineans’ perceptions and experiences of different foods. Maimuna was concerned with her poor health condition, which had led to her migration to Portugal twenty years before. To ensure effective treatment, she combined hospital care with an attention to the medicinal properties of all foodstuffs. On one occasion, I passed the coffee shop near her flat and saw her sitting at a table outside with two friends. She was having a jar of lupini beans – legume seeds customarily eaten in Portugal as snacks. Inviting me to join them, she explained her food choice: ‘I was told it’s good for diabetes, and since I like it, I decided to try’. She picked them from a large jar and savoured them unhurriedly, calling my
Migration, Body and Adaptation
61
attention to their sensorial and material components, by associating their taste with a pleasant sensation and their composition with the high alkaloid content of the uncooked seed. While this is normally eliminated during preparation by soaking the lupini in water for several days, she nonetheless added her own water to the jar to ensure that no toxicity remained.
Preparing Moni during Ramadan: Religious Foodways in a Transnational Setting Certain food products are used in bodily rituals related to birth, marriage or death, illustrating the connection between the material and spiritual realms of food in Guineans’ lifeworlds. ‘After giving birth, for example, banana leaves are boiled and the mixture is used for the woman’s traditional bath that is intended to clean and close the body again’, Quinta explained during lunch at her niece’s house. Marriages also require the sharing of special meals to guarantee fertility for the bride. Manda kabas (the marriage proposal, translated as ‘send the calabash’) involves the offering of certain products that perpetuate the ties between both families, such as drinks, palm oil, rice, tobacco and kola nuts. Funerals and death rituals, too, entail the sharing of food as a way of not only reinforcing solidarity and obligations between relatives and friends, but also assuring the maintenance of a harmonic relation between the living and the dead. When families are physically separated by migration, the continuation of these rituals acquires particular importance, as does the need for adaptation. The religious event of Ramadan offers a good illustration of the role of food in ensuring continuation with migration, as well as of the adaptations needed. Amongst migrants, ensuring the reception of specific food products from Bissau is one of the first concerns when it comes to ritual performances. Sometimes the ritual is carried out simultaneously in the two countries, or divided into two different parts, one to be performed in each country, therefore contributing to reinforce the links between migrants and their kin in both spaces. Some foods play a key role in allowing those links to be kept during such celebrations. Ramadan is an important religious festivity for the majority of Muslim Guinean migrants in Lisbon; as it occurs simultaneously in Guinea-Bissau and in Portugal, and involves the preparation of a special dish, it helps us to understand these links. In my year of fieldwork, Ramadan started during the second part of my ethnographic research in Lisbon, in the summer of 2010, and its relation to fasting introduced new meanings and experiences of space and food during that period amongst the Fula – who comprise the largest part of Muslim Guinean migrants – and the Mandinga. Guinean bars and restaurants located in Rossio and Damaia – the main areas of the city and its outskirts where migrants socialize and exchange food from home – started
62
Food Connections
serving dinners, while reducing the number of lunches, adjusting the business to the new temporal rhythms of commensality introduced by this ritual. Nevertheless, despite the reduced movement of buying, selling and consuming foods during the day, Guineans still regularly attended these locations for social gatherings. Conversations regarding Guinean food were then not only about its qualities and availability or scarcity in Lisbon’s markets – dominated by the Fula themselves – but also about the experience of its absence in daily practices of commensality. On the first day of Ramadan, the ambience in Rossio was quieter than usual. Some of the elders lamented the extremely hot temperatures of that summer in Lisbon, which, they believed, made fasting more difficult. ‘Kalur na matan!’ (The heat is killing me) was their recurrent answer to my greeting questions. I had more often heard migrants complain about the cold European weather, whose negative consequences for the body were partly managed with the help of Guinean food. Yet they now mentioned bodily weakness and headaches provoked by what they considered a terribly dry heat, and expressed longing for the Guinean rains and humidity that could be felt at home at this time of the year. ‘This heat is dry and stays in our body. We feel as if we have fever’, one of them explained. ‘In Guinea, the heat makes us sweat and that’s good, because it cleans our skin’. Sweating has been associated in other West African contexts with the flow and movement that express health (Fairhead and Leach 1996). Blocked sweat, on the contrary, as with other human and ecological ‘blockages’ or other obstacles to the flow of crop production and human reproduction (Gottlieb 1982), provokes illness, confirming the predominance of a socioecological order in Guineans’ lifeworlds. Moreover, the fact that it was summer meant having to wait longer hours until sunset, when they could break the fast. Despite these complaints, however, Ramadan had been wished for and prepared in advance. Moni – a dish prepared with millet, sugar and water – is the first meal typically consumed by Guineans when breaking the fast after sunset. It is traditionally a Fula dish, said to be originally consumed by the djila (itinerant traders) due to its energetic and nutritional qualities. Just as these food properties were needed during the long-distance travel for trade amongst the djila, they are now beneficial during the Islamic fast in Ramadan, to compensate for lost calories. On the days preceding the beginning of Ramadan, the origin of the ingredients and the preparation of moni became common topics of discussion amongst the Fula women sellers in Lisbon. As Ramadan finally started, they organized their trade activities in order to ensure the availability of enough millet during the next month. Some acquired large quantities of pounded millet (sometimes corn, if availability was limited), usually made in Senegal and sold in manufactured sachets by African shops in Lisbon. Others had the millet pounded and sent by their relatives from Guinea-Bissau. Another adaptation – in addition to
Migration, Body and Adaptation
63
the occasional use of corn – was the replacement of liti durmidu (curdled milk), which is normally added to the recipe in Guinea-Bissau, with yogurt or cream in Portugal. In either case, its preparation, exchange and consumption embodied moments of shared satisfaction where the ingredients’ origin seemed to play a less important role than the practices involved. In Damaia, complaints could also be heard about the hot weather. Here, however, there was more activity than in Rossio. Given the peripheral location of this neighbourhood and the consequent reduced police control over exchange activities, women moni vendors used the street to prepare it, reproducing the outdoor spaces of food preparation that had been part of their experience in Guinea-Bissau. Sitting together in circles with one large basin each, they used their hands to mix millet flour with water until small round grains were formed. The touch of the mixture was experienced as a familiar sensation that contributed to bring their transnational lifeworld together during an annual religious event that was simultaneously celebrated in Guinea-Bissau and in Portugal. It was also animated by lively conversations about life in both countries and by the comfort found in the certainty that relatives at home were performing similar practices. After preparation and packaging was completed, however, women moni vendors also expressed tiredness and weakness while sitting with the final packages for sale, associating the sensation with the long hours involved in the process under such intense dry heat and with the effects of fasting. Many Guinean Muslims who were not involved in food trade – more often Mandinga families – preferred, if possible, to have the pounded millet sent directly by their relatives from Guinea-Bissau, in order to have it available for the rest of the month and to avoid having to buy it at higher prices from the Fula vendors in Lisbon. The first family I was invited to join to break the fast was a Mandinga family who had 25 kilos of pounded millet stored in several bags in the kitchen. It had been brought by a fellow villager recently arrived in Portugal, temporarily hosted at their place. That evening, I arrived slightly after sunset, and the fast had been broken fifteen minutes earlier. The oldest brother, who had invited me, had already started his prayers. Since he was the eldest and most respected member of the family, he could not leave the prayer room, and I was welcomed by the younger brother, Bacar, in the kitchen, who protested at not having been informed of my visit, or he would have waited for me to break the fast together. I sat with him and his wife at the kitchen table with my own bowl of moni with cream, listening to their enthusiastic explanations of Ramadan practices and the excitement of having just received enough millet from their home village to consume during that festive month. After eating moni, I shared with them a long meal of kaldu di mankara and kabasera and veludo juice, while Bacar talked about his kinship relations in Portugal and in Guinea-Bissau, which he described as ‘complicated’. Bacar remembered his relatives in Bissau – his second wife and children – with an expression of protest. Frowning, he complained
64
Food Connections
that it was a marriage arranged by his older brother, which resulted in the additional obligation to remit at least 350 Euros every moth, even when he was, as then, unemployed from the construction sector. The connection he experienced with his village of origin in Eastern Guinea-Bissau, which was, as in the case of the women moni vendors in Damaia, reinforced by the emotions generated by receiving homeland millet, did not find an equivalent in his relationship with kin in Bissau, from whom he felt more distanced. In fact, for this family, as for other Mandinga and Fula migrants, Ramadan was both a celebration of a widely shared religious identity as Muslims and of the reunification of geographically distant families in ‘Mandinga land’ – Eastern Guinea-Bissau – through a temporal religious practice. Following the inadequacy of viewing religious and ethnic belonging as two separate dimensions (Johnson 2006), which I discussed in the previous chapter, the Mandinga and the Fula, who are often perceived as strangers amongst other Guineans in Guinea-Bissau and in Portugal, took this chance to celebrate, through the ritual preparation and consumption of a particular meal, their ‘Mandinganess’ and ‘Fulaness’ on a global scale. Ritual meals connect participants to the nonhuman realm, but they also perform critical social functions (see Mintz and Du Bois 2002), reaffirming relationships with other kin or ethnic-based relations (familia or parenti) from the same region of origin.
Djanfa and Kola Nuts: The Material and Spiritual Realms of Food Just as certain foods are particularly illustrative of the entwined religious and ethnic adaptations that take place with migration – like cream or yogurt instead of liti durmidu or, occasionally, corn instead of millet in moni during Ramadan – other objects resulting from new encounters reveal the juxtaposition of multiple religious and spiritual influences in people’s lifeworlds, when risk and uncertainty prevail. Maimuna, for example, was not only attentive to the healing properties of the food she consumed, as described above, but also to the combination of these with other protective objects. She had two miniature statues of catholic saints that had been blessed by the Pope when he had visited Lisbon a few months before. The two statues were laid carefully on her bedside table. Kissing one gently, she said, ‘I asked my sister to give me this one. Here people don’t help each other as they do in Guinea. I know she [the saint] is protecting me’. In her efforts to protect her body, weakened as a result of djanfa that someone had put on her, Maimuna, like most Guineans I met, combined the materiality of objects (such as food, but also the saint figurines she had adopted in Lisbon) with their spiritual realm. She had therefore also asked trustworthy relatives in her hometown in Guinea-Bissau to perform a protective action with the help of the local Iran, revealing the way in which
Migration, Body and Adaptation
65
protection can be imbued in the land. She explained that jealousy and djanfa had always been a part of her life due to the success of her family since the time of their early ancestors, who reigned in the Gabu region. Djanfa is part of the Guinean lifeworld, grounded on a basic cosmology where the land is inhabited not only by people, plants and animals, but also by spiritual entities to whom one might appeal to exert a powerful material influence upon the world. In this book, rather than making use of the translated vocabulary of ‘sorcery’, ‘witchcraft’ or ‘magic’, I follow Guineans’ own use of the terms rumour, jealousy and the practice of djanfa (which, resulting from those rumours and jealousy, is intended to harm others). In the literature on Africa, the use of the terms witchcraft, sorcery or magic is the result of a colonial inheritance, whereby these concepts were applied to beliefs and practices from the colonial frame of reference (Wallace 2015). Murrey (2017) warned of the danger of ‘spiritual othering’ that may result from studying processes of witchcraft in and from Euro-American universities. The tendency of western academics to value witchcraft stories often hides a motivation to subvert modernist certainties, whereas in the context of everyday life in Africa these stories are not necessarily the subject of intellectual curiosity, but ‘a fait social total that is an ongoing part of life with which people have to deal in one way or another’ (Ciekawy and Geschiere 1998: 3). In his critique of practice theory, Piot (1999: 19) argues that in societies where people are not just tied to other human beings but also to the nonhuman world of spirits and ancestors, where the visible body is connected to an invisible one, they may use these beings to amplify their powers and affect their relationships with other humans in ways that are not captured by much Euro-American social theory. A common way to address an academic anxiety with producing othering stereotypes through the study of witchcraft has been to centre it on the notions of modernity and contemporaneity. However, as Kapferer (2002) argued, a risk of superficiality comes with seeing current practices of sorcery and magic as modern responses to current crises. In an anxious move to get rid of a colonial past, this change in theoretical orientation tends to ignore the diverse structures and processes that form such practices (ibid.: 16). In his study of the Haya of northwest Tanzania, Weiss (1996) has shown that sorcery practices are not concerned with contrasts between former ways of life and the hazards of the modern world. Instead, they illustrate the connections ‘between bodies and commodities, semantic values and economic transactions, rural livelihoods and urban travels, as well as local “experiences” and global “events”’ (ibid.: 219). Whereas the misapplication of such translated terms has been addressed in the literature (Pels 1998; West 2005), some authors point out a more complex problem of translation, which includes not just the correct use of words but also a transformation of the senses (Nakamura and Pels 2014). Since Evans-Pritchard’s (1937) symbolic approach to African witchcraft,
66
Food Connections
focused on systems of belief, anthropologists have increasingly agreed that ‘witchcraft’ is not just a belief about the world but a patent feature of it. In this regard, Murrey (2017) proposes creative and non-conventional communicative ways to further decolonize the production of knowledge on the topic, which include written, oral, aural, embodied, expressive and visual (ibid.: 165). Phenomenological approaches have been increasingly adopted (Jackson 1989; Kapferer 1997; Stoller 1995),7 with Nakamura and Pels (2014) proposing a materialist approach that emphasizes the material, as opposed to the immaterial or illusory quality of magic. They interrogate how materials, in combination with space and attributes, mediate between the spiritual power and human agency and action. The importance of the material was evident in Guineans’ practices of protecting against djanfa. When I first encountered Kadi’s aunt over lunch, she looked at the cowry shells that adorned my Guinean sandals – a gift I had received from her niece in Bissau – and expressed enthusiasm for what was, to me, a simple decorative object. ‘It is very good that you have those. Do you know what they are for?’ she asked me then. Following my hesitancy, she broke the silence and said, ‘they’re for your protection!’ Just as cowry shells are used as healing and protective objects, some foods, like kola nuts, can fulfil similar purposes. Later that afternoon, Kadi excused herself and left the house to get two kola nuts she had left in the car. While she was away, her aunt explained, ‘I asked her for those kola nuts. One is for me to eat, because I really like kola. The other one is for simola’. Simola consists in a payment, through material objects, for the assistance of the spirits or God in freeing the person from trouble or djanfa. Quinta had dreamt of opening a kola nut that had gone bad inside and associated the image in the dream with her ailment. ‘It’s because of my foot. Something is wrong inside my body’, she said. Dreams, like the land, link human beings to the spirits of the ancestors and might predict the future or offer orientations on how to proceed in certain situations (Je̜ drej and Shaw 1992). Following Quinta’s interpretation of her dream, she would later go to a crossroad near her niece’s house and place a kola nut at the intersection, while offering a prayer. ‘It’s a place of passage and junction of different paths’, she explained. ‘Spiritually, the kola nut will be multiplied and solve the problem in my foot’. Abu, a Muslim Fula man who was part of the small urban elite of Bissau, also narrated a personal episode that involved the offering of kola nuts, revealing how the materiality of objects, dreams and simola crosses over the notion of class: There are people who have dreams… This week, I had to buy one hundred kola nuts and offer them to the mosque, because someone had a dream where I was doing it. Even with all the high education that one gets, one has no choice. It is as if I am mandjidu [tied]. I can’t deny it – kola nut has an important meaning.
Migration, Body and Adaptation
67
Kola nut is one of the food products that better illustrates how human body and spirits are intimately related as a result of a shared land. The materiality of the fruit and the practices that involve it are able to guarantee continuity and stability of that relationship with migration. On the one hand, being an addictive, stimulant fruit, kola nut plays a key role in comforting the migrant’s body. On the other hand, it materializes the spirituality of things and the role played by those things in generating or maintaining social relations. If, as Mauss (1950) advanced in the context of Maori law and religion, given things are endowed with hau (spiritual power) that is passed on from giver to receiver, kola nuts embody a double meaning in the act of giving. Not only can they be sent from Guinea-Bissau, like other foodstuffs, as gifts to the migrant kin, but the act of offering them symbolizes respect and is part of greeting or thanking rituals. They also initiate or strengthen existing relationships, and are performed, in both countries, as elsewhere in West Africa, in courtship practices, formal proposals, weddings and funerals (Drucker-Brown 1995). As Quinta’s story has revealed, they play a central role in simola, a payment that might be linked, as it was in her case, to health-related problems. The spiritual and corporeal significance of kola nuts was confirmed by Abu, who explained, ‘kola nut is one of the main fruits consumed in Portugal. It is not only a good stimulant, especially needed due to the cold weather in Europe, but it is also extremely important because it symbolizes respect and forgiveness’. Amongst migrant men in particular, kola nut’s stimulant proprieties are also key in treating a condition of sexual impotence described to me as common during the first period of migration, which followed a change in climate and food habits and a general weakness of the body. The materiality of kola nuts and related practices of exchange and consumption also reveal how the realm of the ancestral spirits is entwined with religion. Seeing the terms witchcraft, sorcery or magic as separate categories from religion, or the sacred and the profane as two separate realms, is a view rooted in a Judaeo-Christian discourse of good and evil, whereby ‘the occult’ is condemned or seen as a primitive stage in the history of human religious thought (Kapferer 2002; Wallace 2015). Instead, Kapferer (2002) suggests the abandonment of such categories and argues that the spaces of magical and sorcery practices are largely amoral, or contextually moral, and their force is often a result of their connection with other realms, such as that of religion. Kola nuts, due to their role as payments to God or the spirits, illustrate such connections. Quinta, for example, who had introduced herself to me as Christian, also made use of the spirits – through the offering of a kola nut – to mediate her plea for good health. Further examples were recurrent in my observations and conversations with the elders in communal spaces of Guinean conviviality in Bissau and in Lisbon, which were often interrupted by a neighbour or a passer-by who would stop to make an offering, following a muru’s (Muslim diviner) recommended treatment. Like kola nuts, especially their
68
Food Connections
white variety, other white products such as rice, milk, salt, candles or paper were used as simola on those occasions. Once, in Rossio, a man came by with two cartons of milk and handed them to Djara, one of the elderly women who daily occupied that space of food exchange, addressing her in Fula. Djara, another woman and the two men with whom I was sitting in a circle, under a tree, raised their hands with the palms facing up and joined them in the air, while one of the men offered a prayer in Arabic. The other woman in the circle took my hands and placed them in the right position, asking me to join. At the end of the prayer, the donor thanked them and left. Then, pointing to the man who had prayed, Djara explained to me, ‘I asked him to do the prayer because he’s the oldest man. In the prayer, he thanked the simola and wished for its donor to attain whatever he was asking for’.
Transnational Healing Practices: On Being a Muru away from the Land The Guinean ‘basic cosmology’ (Temudo 2008), whereby religion and the spiritual entities that inhabit the land are conciliated, is also observable in the Guinean healers’ work and in the plant- and animal-based objects that they
Figure 2.2. Shipment of kola nuts. At a storehouse in Bissau, 2010. © Maria Abranches.
Migration, Body and Adaptation
69
use, which confirms the inadequacy of separating deceptively opposed terms. Moreover, these materials are, like the power of the land itself, creatively and pragmatically adapted by Muslim healers in Lisbon, who predominate amongst Guinean migrant healers. Seco was a Mandinga sapateiro and healer. Sapateiros, literally translated as shoemakers, are Mandinga leather workers whose primary occupation is to sew the amulets or charms (the mesinhu di asunto introduced by Infali in the previous chapter) prescribed by healers as part of a treatment for a variety of corporeal, spiritual or social ailments, with a leather wrapper and string (figure 2.3). The leather sachets Seco sewed to be used as amulets (which I will refer to, as he did, as mesinhu) are usually fixed around the patient’s belly or neck. In Portugal, however, they are more often kept in a purse or in the pocket, in order to avoid the suspicious or invasive gaze of others. This need to adapt the way in which mesinhu is used indicates that the same material which is often aimed at protecting against the adversities resulting from migration can simultaneously be a reason for those adversities. Indeed, the use of amulets can generate a discriminatory gaze from non-Guineans in Lisbon, although they are commonly worn by Guinean migrants to protect them precisely against the threats of discrimination that is experienced in the new country. They are, too, an example of the material adaptations that take place. Seco’s cousin Baciro, a healer and mesinhu maker like himself, lived in Bambadinca, a small town in eastern Guinea-Bissau. From there, Baciro occasionally sent animal skin to Seco, alongside ram and other bush animals’ horns, which he bought from specialized hunters in the local weekly market, before having them sent to Bissau and, from there, to Lisbon. Yet most of the material Seco needed for his job could be easily found in Lisbon: Lisbon has many shoemaking shops that sell most of the leather we need: sheep, goat and cow. Here, the leather is even more perfect, due to mechanical tanning processes. In Guinea-Bissau it is done manually, and the leather is harder… Here we can buy black leather already prepared, there’s no need to manually dye it… Most of our work is done more easily here than there. There, for example, they cover the mesinhu with pieces of cloth that need to be ironed. Here, we found this new cloth tape, which is much easier to use. Similarly, working tools such as needles, screwdrivers and strings are bought in local shops. In addition to the practical advantages of using tools and materials found in Portugal, as described by Seco, the costs of having them sent from Guinea-Bissau, which include the price of transportation to Bissau and from there to Lisbon, as well as the risk of having them detained by customs authorities on arrival, would render the business unprofitable. Although Seco performed the work of sapateiro and healer simultaneously, these are usually two different jobs. For Muslim healers and diviners,
70
Food Connections
the difficulties of adaptation with migration are more deeply felt, as are the obstacles related to finding the necessary material away from Guinean land. Objects such as the cowry shells that adorned my sandals and protected the body – also used by healers in their divination work – were easily acquired in small plastic sachets from any of the several Chinese storehouses in Rossio and, like the food described above, did not require an ‘authentic’ Guinean tchon understood in western terms. It was, however, the power of the Qur’anic inscriptions contained in the little leather sachets sewed by sapateiros – which are believed to truly protect or help the bearer of amulets (Mommersteeg 1988) – that entailed more difficulties when being used far from Guinea-Bissau. One of the reasons for such difficulties is that the religious powers of skrita (writing) work alongside the material power of the land.8 As Seco explained, ‘it is more about skrita. All other materials are complementary to the writing of certain verses of the Qur’an. We learn how to do it from the elderly healers’ secrets. And for it to be perfect, unlike the work of sapateiros, it is easier to do it in Guinea-Bissau’. Proximity to Guinean land is important, for example, when the ailment or problem that the patient suffers from requires charms to be buried under the fire where women cook, or at the entrance of the patient’s house, on the ground (tchon), in order to benefit from the land’s strength. Adaptation, in these cases, requires creativity, as Seco went on to explain:
Figure 2.3. Materials and tools used for amulet-making in Lisbon, 2010. © Maria Abranches.
Migration, Body and Adaptation
71
We try to be creative. Where can we bury it here that is as hot? Sometimes we try a heater, or a car’s motor. But it won’t be under the ground; it won’t be in contact with the soil. Where can we find a place here where people cook on the tchon?... How can we dig our floors, or the walls of our houses? In Guinea-Bissau, you can do it with a knife, and no one will feel it… Reality is different here, and we need to adapt. Animals used in healing and divination rituals, unlike the leather used to cover charms, also present difficulties in terms of access. Galinha di matu (Guinea fowl), for example, is commonly used in such rituals. On one occasion, Seco needed to perform a healing ritual to free a patient from djanfa, which involved cooking galinha di matu in special water where sacred ink from skrita had been infused. The alternative, in this case, was to divide the ritual between both countries. With a phone call to Bissau, Seco asked another relative, Dembó, to buy the animal and offer it to an elderly woman. In Portugal, the patient would dilute the ink in water and wash their body with the liquid, instead of eating the chicken cooked in it. The djanfa would thus be successfully removed through the use of both geographical spaces, spiritually and materially connected via this ritual. Sometimes, due to other migration-related obstacles, Seco’s cousin Baciro, like other healers in Guinea-Bissau, performed his healing practice at a distance. Baciro gave an example of one such performance: Next week I will work for an emigrant, so that he can pass an exam for university… I will do the work from here and he pays me only after he passes the exam… Working at a distance is more complicated though. If he needs to offer a simola of goats, he won’t be able to do it there, so he has to send money over here so that someone can do it for him. Amidst these adaptations, migrant healers and sapateiros like Seco criticized others who, following migration, seemed to suddenly gain more powers than they had at home. The fear of dishonesty in the work of migrant traders of mesinhu had been brought up by Infali and his nephew in Bissau, who had shown concerns about the unethical shipment of plant medicines without prescription, as well as its indiscriminate commercialization in Lisbon with no appropriate knowledge of dosage (chapter 1). Likewise, some migrants were criticized for publicizing false healing and divination knowledge, while their expertise was, in Guinea-Bissau, restricted to the work of sewing charms. Seco, for example, saw the act of distributing flyers, which had become common practice amongst Guinean migrant healers to advertise their work, as morally inappropriate. Announcing themselves as ‘astrologists’, many had seen in this practice a way of targeting a wider clientele, which the word muru would not attract. In these adverts, some added titles such as ‘master’, ‘professor’ or ‘scientist’. Seco, however, contested this practice. In his view,
72
Food Connections
it was an immoral deceitfulness, a commercial opportunism embodied in the impersonation of learned men, which, in fact, they were not. Instead, having studied halfway through secondary education and having practised as a teacher in Bissau, he considered his studies to be an advantage in his work as a healer. He understood the combination of his formal and religious education as a tool for being able to practise a ‘modern’ kind of healing: I don’t publicize my work, because I have a different ambition. I want to be a modern muru; to combine the two knowledges. I have an academic knowledge, thank God. Many muru don’t even know how to write their names. I am different from them. If someone needs, I can write a Qur’anic verse in Portuguese. That’s my secret weapon. This academic background did not remove the power that had been granted to him through inheritance and by means of learning the elders’ secrets, as well as from attending an Islamic school and receiving the powers of the Qur’an. Having studied the Qur’an with a master in the Gambia, he did not dismiss the importance of his religious education from his discourse. Any western-based wish to understand from which domain his authority as a Mandinga healer originated cannot be fully satisfied here. Instead, his discourse moved constantly from the importance of formal education to the key role played by his Gambian Muslim master. It is precisely Seco’s notion of ‘modern muru’ that provides evidence of the way in which healers challenge the definitional boundaries of ‘traditional healing’ in myriad ways, by adopting new methods and materials, borrowing ‘traditions’ from other times and places, and even inventing new traditions from scratch (West 2007: 39). Seco’s religious education in the Gambia also illustrates the importance of the land seen in terms of social relationships in a space that is not confined to nation-state borders, but to the notion of what might be called, in Seco’s case, his ‘Mandinganess’. The importance of a ‘Mandinga land’ for Seco, understood historically as related to a wider common tchon that includes several states which now neighbour each other, is demonstrated in the story of one of the materials – a horn – used in his healing practice, given to him in the past by his Mandinga Gambian master. ‘When I was learning with him, he gave me a horn to use in my practice, but it is not powerful enough anymore’, he explained. As a result of the horn’s loss of efficacy, Seco now needed to go back to the Gambia and find the master who had taught him, since he was currently having to complement his work with a large number of prayers offered throughout the night. ‘I don’t sleep enough, because I have to pray all night to compensate. I am too tired. I need to exchange this horn for a more powerful one, but only the master who gave it to me in the Gambia can now change it over’. In Portugal, Seco had clients of various origins, ranging from different African nationalities to Brazilian and Portuguese. Aware that not all of them
Migration, Body and Adaptation
73
were Muslims, he made use not just of Qur’anic texts, but also of the Bible, which he kept on a shelf in his room, side by side with the Qur’an. Yet this interplay was, for Seco, not just a strategy to gain more clients. Alongside the obvious wish to satisfy his varied clientele, he also considered it to be his own special power, derived from his good Portuguese reading skills and wider cultural knowledge, which complemented the ‘secret’ powers of the Qur’anic inscriptions contained in the little leather sachets. For him, it was the advantage of his educational attainment that, combined with his knowledge of religion and of the power of the land, strengthened his authority as a healer. Seco’s narrative shows, on the one hand, that the view of the exogenous imposed nature of school as the locus of reproduction of postcolonial state ideologies and social inequalities (Mbembe 1985) should not make us overlook the active relationship that Africans also establish with school, appropriating it as a strategy for attaining a new form of social prestige (Bordonaro 2009). On the other hand, seen through the lens of migration, it shows that migrants need to find a comfortable position where sense is made of the worldviews of others without rendering their own views of the world nonsensical (West 2007). *** A phenomenological approach to the study of food and migration in West Africa looks at the materiality of everyday products and practices that help to facilitate corporeal adaptations to a different environment and way of life (Gardner 2002). It offers important insights into how changes in Guineans’ food practices do not necessarily produce new meanings but rather a new embodied experience, whereby any changed meanings are incorporated into people’s lifeworlds. When certain foods are not available, familiar preparation techniques, such as the method of smoking fish, grilling corn or using spices and sauces made with ingredients sent from Bissau, as well as customary consumption practices like sharing meals in spaces of group socialization, can act as adaptation mechanisms that endow exogenous food with Guinean value through the familiar experiences they generate. Although ‘Guinean land’ and the food that shares its substance – as much as the historical consciousness of a Fula or Mandinga land – are an essential carrier of wellbeing for migrants, their protective and healing proprieties can continue to be acquired through those adaptation mechanisms. Exogenous land can hence be ‘made Guinean’, as Strathern (1988: 81) found amongst the Vanuatuans, who ‘vanuatized’ European things to their own ends in the Melanesian context. Just as the term fidju di tera (children of the land) is not only reserved to the native-born, what is considered kumida di tera (food from the land) does not always correspond to native crops grown in Guinean soil. In many cases, however, Guineans in Lisbon perceive ‘European food’ as ‘contaminated’, due to the chemicals used in industrialized methods of
74
Food Connections
production. Quinta’s narrative regarding the relationship between ‘European food’ and cancer and other illnesses in Europe was illustrative of such perceptions. Yet it is important to note here that this discourse also intersects with a current European view of organic food as healthier. Industrialized foods are lamented by many Portuguese, as they are by Guineans, showing that, on the one hand, the positionality and relationality of such perceptions and experiences might bring different lifeworlds closer than what is often imagined. On the other hand, as Williams-Forson (2014) has shown in the context of Ghanaian migrants in the United States, migrant voices are often absent from ‘eat local’ campaigns. Using the concept of ‘cultural sustainability’, she called attention to the importance of acknowledging that such discourses can create strong inequalities in cases where preserving food culture is vital for the wellbeing of people. Despite the processes of adaptation described, Guineans tend to experience weakening of the body with migration, particularly in the period that immediately follows the move away from the land. The relationship between body and land amongst Guineans is also evident in the exchange of greetings that I started by describing in this chapter. As in other West African countries, ‘agreeing’ or ‘accepting’, for example – which are ways of responding to greetings – indicate health as much as success in agricultural activities, revealing the way in which social and ecological relations are entwined (Fairhead and Leach 1996: 144). ‘Refusing’, too, designates a sense of illbeing that, amongst newly arrived migrants, is often caused by bodily rejection of certain foods. In these intimate connections, the realms of religion and the spirits of the ancestors that inhabit the land also help to ensure protection to migrants. Healing rituals and their adaptation with migration forge a unique path in the biography of foods, plants and other objects, as well as in people’s relationship with those products. Just as migrant healers adapt the materials and tools used in their charm-making and healing work, they, like all Guineans, also make use not only of different beliefs about the world, but also of different ways of making worlds (Geshiere 1997; Moore and Sanders 2001; West 2007). The next chapter will explore such ways by focusing on memories and imaginaries that are generated and transformed with the movement of food.
Notes 1. Quinta, like most other Guinean migrants, referred to the food found in Portugal as ‘European food’ or kumida di branku (white people’s food). 2. Interest in the field has resulted in a special issue of Food, Culture & Society edited by Harry West in 2011. 3. Caritas Italiana is an ecclesiastical charity, represented worldwide.
Migration, Body and Adaptation
75
4. See Bielenin-Lenczowska (2018) for an analysis of Macedonian migrants’ new creative foodscapes in Italy. 5. See Fikes (2008, 2009) for an account of Cape Verdean women’s fish selling in Lisbon. 6. A widely cultivated cereal in the eastern savannah woodlands of Guinea-Bissau. 7. Moore and Sanders (2001) offer a detailed review of the history of anthropological approaches to ‘witchcraft’. 8. Seco referred to the powers of skrita as sigridu (secret), suggesting that they derived from sources that go beyond God and the Qur’an.
CHAPTER 3 TEMPORAL CONNECTIONS The Making of Memories and Aspirations through Food
When I arrived in Rossio on a Sunday afternoon, Meta gestured an invitation
for me to sit with her under the shade of the same tree. She was a recently arrived young Mandinga woman who used that space on weekends to sell homemade ondjo juice, konserva and other prepared foodstuffs.1 Djara and Dála – two elderly Fula bideiras – were occupying the same area. From the two elders, Meta usually received support and guidance in different matters of her new life in Portugal. In return, she offered them help in their food selling business by covering their absence whenever necessary. That afternoon, they were discussing Meta’s concerns about the need to find a more stable job than the temporary replacement position she had at a local supermarket. Shortly after I arrived, two women dressed in fancy bazin garments came and sat near us.2 They spoke to each other in French, mainly about their businesses. One of them was Adama, a Guinean Fula known to Djara and Dála, who joined us for the rest of the day and continued to share her experience of what she considered a successful African clothes trade, not only due to the profit made but also, relatedly, to the material and mobility involved in the exchange. She emphasized the quality of the fabrics – the same kind of bazin garment she was wearing – and her regular travels to Guinea-Conakry and Senegal to buy the material intended for selling in Portugal, where she had been living since 1980. Later in the afternoon, as usual, Sali came down from her restaurant with cooked yams to sell, and most of us bought one. Adama hesitated at first, but with Sali’s and the other women’s insistence, she finally took one. Holding the yam that had been passed on to her with a suspicious look, however, she complained, ‘don’t you have a knife?’ Sali passed her a knife. Yet the friendly and mocking dispute continued when she wasted part of the yam as she was trying to peel it. The other women reproached her. Sali took the knife back
78
Food Connections
and peeled the yam herself, as Adama, smiling, justified herself: ‘you know, I’m from prasa’. Projects of migration and return were discussed at length during the rest of that afternoon, motivated by imaginaries and memories embodied in materials such as Adama’s clothes and Sali’s food, and their orientation towards unrealized futures. Keane (2005: 194), drawing on George Herbert Mead, uses the example of the chair to reflect on the way in which objects invite the embodied actor to perform certain types of future action (e.g. to sit down) by virtue of their form. Likewise, homeland food, due to its familiar tastes, smells and ensuing bodily sensations, instigates new intentions and projects of return amongst migrants, as this chapter will demonstrate. Clothes like Adama’s, too, open up new possibilities of migration and movement, and performances such as the one she enacted, combining food and clothes, help to shape the imaginaries of other migrants, just as similar performances upon return influence aspirations in Guinea-Bissau. Firstly, the encounter between Adama and Sali’s yams exemplifies the capacity, and perhaps danger, of material things to change the person. Eating them, as well as the particular way of eating, could risk downplaying Adama’s differentiated style, visible in her clothing, jewellery, language, successful business, capacity to travel and association with the urban elite of Bissau (prasa). Additionally, while critical – though in an amiable way – of her apparent distance from a customary food consumption practice amongst Guineans in Portugal, the other women in Rossio commented longingly on the return trips that successful traders like Adama were, unlike themselves, able to perform. This distinction is also linked to the value of the material exchanged – which will be explored in the next chapter – since the key social value of food for migrants does not equate to its economic value when compared with materials like clothes, whose trade is capable of generating higher income, resulting in more mobility. Adama’s example may at first sight seem to illustrate what Ferguson (1999, 2002) defined as Africans’ enthusiasm with ‘modernity’, or the worldliness and cosmopolitan style explored by Gable (2006) amongst return migrants in rural Guinea-Bissau, which are partly materialized in the idea of an increased mobility in modern landscapes. Adama’s lifeworld, in fact, emphasizes mobility and its materiality, combining objects (food and clothes), places and movement in dynamic, performative and temporal ways (Munn 1992). Yet her example of mobility should not be seen merely as an embodiment of modern global networks (Hannerz 1996) and flows (Appadurai 1996). On the one hand, it is important to be cautious, as recent scholarship is, against the privileging of cosmopolitan mobility and an idealization of movement that depends on the exclusion of others. On the other hand, stasis is as much part of mobility as is movement (Sheller and Urry 2006), and some forms of mobility involve the desire for permanence. Indeed, although Adama talked about her regular trips to West Africa with satisfaction, while eating Sali’s yams she also referred to that desire, expressed in her future plan to
Temporal Connections
79
permanently return to Guinea-Bissau, although claiming that she was ‘not ready yet’ due to the money she was making with the sales. In the context of Guinea-Bissau, as elsewhere in Africa, an understanding of mobility that focuses on movement only might therefore not be entirely suited (Mavhunga 2014). Here, rather than looking at mobility and immobility as two positions whereby one is the absence of the other, I seek to explore, as Gaibazzi (2015) did in rural Gambia, how some forms of immobility (like Djara’s and Dála’s inability to return) are transformed into dynamic and reinvented positions. Although seen as a constraint by the majority of Guineans in Bissau and in Lisbon, the lack of prospect to travel is remade, through food, in memories and aspirations that make them participate in the wider world. Following a phenomenological perspective that understands cognition as embodied in the experience of making reality, this example illustrates how memories and aspirations, materialized in and triggered by performances involving food, help to temporalize people’s realities, whereby the past, present and future become intrinsically connected. As Stones at al. (2019: 50) put it, retrieving the past and imaginatively projecting the future can become part of the routines of migrants’ embodied daily experience, as two physical worlds fold into each other. The dailiness of food practices and spaces of food consumption, like eating yams in Rossio, provides food with a key role in these connections. Food, as Sutton (2001) described in his study of the Greek Island of Kalymnos, is as a provider of temporal rhythms. Everyday contexts of buying, preparing and consuming it structure both daily routines and more long-term rhythms (ibid.: 16), which are, in the case of the women who accompanied Adama that afternoon, represented in their life projects and experiences of migration and return through the small-scale food trade in which they are involved.
From Guinea-Bissau: Imaginaries of Migration through the Circulation of Food In Guinea-Bissau, the imagination of a better life in Europe by smallholders, bideiras, agency owners and staff, and others involved in the task of sending Guinean food to Portugal, is materialized in the constant movement of that food, to which they contribute. Mila’s narrative (chapter 1) already demonstrated how small producers aspire to follow their crops to Lisbon. To some of them migration was such an important part of their plans that they could name and describe places in Lisbon where they had never been – Rossio and Damaia in particular – only by having heard stories of their foodstuffs’ final destination. The travel of food plays an important role in activating or deepening this desire, and influences perceptions of Europe in complex ways. The stories that follow, of Yasin and Dembó, illustrate some such ways amongst a younger generation, for whom migration aspirations are more relevant.
80
Food Connections
They reveal how these projects, which may (in Yasin’s case) or may not (for Dembó) be accomplished, are influenced by imaginaries of ‘development’ and wealth. These stories also show that such imaginaries are temporally shaped: they are historically situated, drawing on the colonial past, while also resulting from people’s current involvement in the transnational movement of food and plants and coexisting with the fear of bodily risks and threats that I explored in the previous chapter, related to an unfamiliar environment and unknown food.
Yasin Yasin – Kadi’s daughter, who sent the foodstuffs her mother sold in Lisbon – was twenty-three years old when we met in Bissau. While her ethnic background was, like her mother’s, a complex mix that made it difficult to ‘map’ in the constellation of Guinean ethnic groups, her young age was representative of the majority of Guineans who, amongst the people responsible for sending food to their relatives in Portugal, aspire to migrate. For the elders, instead, the threats that unfamiliar environments and foods pose to their body make most of them fear migration more than, or as much as, desire it. ‘Europe is not made for old people’ was an idea that I frequently heard in people’s conversations, including those with Yasin. Older people involved in the process of sending food to Portugal would also more often complain of the exhaustion, risks and difficulties associated with their roles of farming, selling, buying, packing, transporting or finding carriers, especially the complex relationships of exchange and trust that were necessary to safely complete the process, which I will explore in the following chapters. While young people like Yasin might express similar protests, these were more often associated with the obstacles that such tasks created on their other life projects, especially studying, due to their time-consuming nature. The successful way in which Yasin’s migration story developed is not a typical one. Unlike her, most of the young Guineans I met in Bissau had not succeeded after years of attempting to migrate. Yet it allows for a better understanding of the journey and experience of both sides of the migration for those who have long been imagining that same journey through the movement of food. Yasin was already gathering all the documentation needed for her visa application when we met in Bissau. When we first discussed her future migration to Portugal, she had told me that she ‘just want[ed] to see what it looks like’. Later, as our friendship developed and the complex – and, to her, frustrating – bureaucracy linked to the visa application became a regular topic of our conversations, Yasin started to constantly refer to an imagined Lisbon. When we shopped for vegetables in Caracol market, or as we met at
Temporal Connections
81
the airport in the evenings, where she went to have her food parcels shipped, the intensity of her project seemed to gain strength. These were particularly important spaces for the creation and intensification of imagined futures (Abranches 2013a, 2013b). At the airport of Bissau, where Yasin went on the three weekly flight nights to have her food and mesinhu parcels sent to her mother in Lisbon, the movement of those parcels, which she wished for herself, vividly materialized her desire for migration. To Yasin, as to younger senders in general, having food sent three times a week impacted upon this desire in particular ways, since it represented a family obligation that imposed constraints on their personal lives. Emphasizing those constraints, Yasin started presenting her intention as more than briefly visiting Lisbon. One night, we were sitting outside the check-in area at the airport with her aunt Aua, who was a migrant in Lisbon on a temporary return, looking for carriers for her parcels. Experiencing the risk-related anxiety of that particular task (which will be further explored in chapter 5), she explained that, while for many of her older fellow food senders this was considered a job from which they were compensated monetarily, it was not worthwhile to her. On those days, which started early in the morning, she rushed from phone calls to the bideiras (to negotiate the produce needed) to trips to the local food markets and then sorting, packing and coming to the airport late at night. ‘This is not a job! Today I had to miss classes again, and my mother doesn’t understand that’, she sighed. Then, more optimistic, she added, ‘I hope I can get there by August’. Her aunt Aua then suggested that if she ever got a residence permit, once in Portugal, she would not come back. Yasin smiled and, after a short silence, asked how hard it was to get one. Yet alongside this desire to migrate, which was turning into a longer-term project, she also shared with other Guineans the image of Europe as a place that aged people, resulting from the weakening of the body described in the previous chapter. When I showed her the photos of her migrant relatives, which I had taken with me to Bissau, I was surprised at her expression of disappointment as she said, with a grimace, ‘They look so old! Europe really ages you’. This conscious fear of bodily deterioration was more noticeable amongst the elders, but not exclusive to them. It went hand-in-hand with the general will to improve one’s economic life situation with migration or, as was more evident in Yasin’s case, to pursue an individual project that could liberate her from the family obligation of sending food for her mother’s trade activities in Lisbon. Yasin’s migratory project was not accomplished in August but only four months later, in December. When I met her in Lisbon the following summer, she was, as her aunt had predicted, already applying for a residence permit. This would not be difficult to get, she hoped, given her mother’s Portuguese passport, obtained on account of her long-term residence in Portugal. Following an initial stage of what Yasin described as severe adversities
82
Food Connections
related to the emotional and physical effort required to cope with life in an unfamiliar place – especially feelings of loneliness, insecurity, fear and discomfort under the cold weather of December, which unsettled her premigration aspirations and imagination of Lisbon – she was now considering staying for an indeterminate period. This intention was linked to the opportunity of what she then phrased as ‘moving towards development’, a notion that she, like other young Guineans, perceived as associated with the possibility of continuing her studies. ‘Now I like it more here, due to development. I think I still have a lot to learn’, she then told me, before revealing her plan to study computers.
Dembó Dembó, introduced in chapter 1, was a thirty-two-year old Biafada, responsible for sending the materials that his relative Seco needed for his charmmaking and healing job in Lisbon. He got most of these materials from Infali’s pharmacy in Bandim market in Bissau. Dembó was, of all the Guineans I worked with in Bissau, the one who most insistently spoke of his desire to migrate. Unlike Yasin, however, he did not have a close relative in Lisbon who could help him in the process, although Seco contributed with money whenever he could. Dembó often complained of the lack of solidarity of his migrant uncles and aunts, who he imagined to have a wealthy life, yet were unwilling to help him – financially and with the necessary procedures – to migrate. Conversely, this desire to migrate was, like Yasin’s, embedded in the consciousness of bodily weakening that Europe represented, which he also perceived as resulting from the disruption with the vigorous Guinean land and nutritious foods, as well as from an unhealthy lifestyle and unfamiliar environment. Aware of that threat, he considered Seco’s healing job in Lisbon and his own responsibility for shipping plant medicines and materials for amulet-making to be of great importance, partly echoing Gaibazzi’s (2015) findings in the Gambia, where ‘sitters’, or those who do not migrate, seek recognition by assuming responsibilities towards the community (ibid.: 159). ‘Many people there get several diseases, I know’, Dembó said. ‘But they can’t always come back to receive treatment, and this is why it is important that we, here, send them mesinhu’. Yet objects travelling in the reverse direction, especially those displayed by return migrants, incited Dembó’s imagination of migrants’ wealth in ways that counterbalanced his perception of health-related risks. As he explained, anxiously, ‘sometimes I feel like I’m losing my mind, because I think of my bida di koitadi [poor life] here, and I see others who are in Portugal or Spain and have houses and cars… One of them built two or three houses here; others, when they come to visit, they bring ten different sets of bed linen’. The tension experienced here was, as Jackson (2013: 165) put it in his essays
Temporal Connections
83
on existential anthropology, linked to the existing gap between the objects on which Dembó secured his imagination and search for satisfaction, and the external situation of his existence, which creates an intersubjective mode of being that is based on meaning and experience. Like Yasin, Dembó defined his lack of technological knowledge and skills as associated with his suffering – a result of his condition of being stagnant and lacking development – which he hoped to overcome with migration and the study of computers. As he put it: I am suffering now, but it will get better. I am going to study computers. Nowadays, if you don’t use computers you are like a mobile statue, you won’t know what’s going on in the world… I know that one day… I don’t know which day, at what time, or in what year, but I’ll be in Portugal. Or in Europe! No one can be punished forever. This reference to punishment is associated with the influence of djanfa, which, for Dembó, as for others in Guinea-Bissau and in Portugal, had direct consequences on the success or failure of migration and return projects. In his view, it was the djanfa put on him by an uncle that had been hindering his long-term plan to migrate. After his mother’s death, Dembó’s uncle inherited the woman’s cattle and farm, according to tradition. In order to keep his inheritance safe, the uncle made a contract with the village Iran, according to which Dembó would die if he ever tried to get part of his mother’s possessions. Taken by fear that Dembó might claim them if he was ever to gain more power with migration, his uncle’s contract with the Iran included a ‘clause’ according to which Dembó would not succeed in migrating abroad – the young man’s long-term aspiration – unless permission was given by the uncle himself.
Reconciling Possibility and Risk in Projects of Migration A phenomenological approach to the study of migratory imaginaries and aspirations, which sees them as not just endowed with meaning but as a historically situated intrinsic part of Guineans’ lifeworlds, can give us important insights into the temporal dimension of lived experiences. It can also elucidate the way in which future projects are embedded in everyday juxtapositions of possibilities and risks, as illustrated in Yasin’s and Dembó’s stories. In Togo, Piot (2010) drew attention to the way in which people are ‘longing for a future’ through the appropriation of commodified imaginaries and desires, clearly visible in the widespread requests for exit visas. Similarly, Guineans involved in the transnational circulation of food and mesinhu construct their own realities based on imaginaries that link future projects with present and past experiences. Their projects also exemplify the complex
84
Food Connections
interplay between desired and threatening outcomes of what some, Like Yasin and Dembó, define as ‘development’, and its connection with the land. On the one hand, development is understood as a resource in people’s struggle to overcome either the feeling of koitadesa – a Creole notion that indicates the perception of being unfortunate and poor, as was the case for Dembó – or the obstacles resulting from family obligations related to sending food, which inhibit the accomplishment of their individual aspirations, as for Yasin. On the other hand, it is also perceived as having a negative impact upon people’s corporeal wellbeing, linked to the distance it imposes between them and their land. Possibility and risk are therefore entwined in people’s perceptions and experiences of migration, confirming the ambiguity that prevails in one’s engagement with social and economic transformations, whereby a desire for change is combined with deep anxiety about that same change (Mitchell 2001). Such ambiguities also reinforce the importance of acknowledging that ‘development’ means different things for different actors (Escobar 1995; Grillo 1997; Mills 1999) and is experienced differently depending on situation and context. Within the reconciliations that Guineans adopt to make sense of their life, it is also important to recognize the role played by the colonial past in the construction of the ‘development’ imaginary. As Fonkoué (2019) argues, this heritage weighed on the post-independence project of nation building in Africa, by turning it into ‘developmentalism’, ‘a top-down and forced-march “modernization” that deprived the citizens of opportunity to participate in shaping the future of their country’ (ibid.: 14). Faced with failure and lack of opportunities, people like Yasin and Dembó can easily erase or distort their memories – or, amongst those of young age like them, their imaginaries – of the past, hence associating development with a certain order and stability that, however, was built at the cost of individuals’ rights to participate in public life. The way in which young Guineans like Yasin and Dembó associate development with technology also calls for further reflection. Scholars like Diouf (2000) and Mavhunga (2014) have drawn attention to the need to move beyond talking about technology in Africa as external innovations that have either victimized Africans or created the need for them to appropriate and adapt. Instead, they argue, it is important to recognize African technology in its own context, and Africans’ involvement ‘in a process of exchange, emitting their own things in exchange for those of the outside world’ (Mavhunga 2014: 11). When technology is defined as a way of doing that combines skills, innovation, creativity and knowledge, we can clearly see how people involved in producing, exchanging, transporting or sending food and mesinhu – like Yasin and Dembó – are creating it. Yet in the case of these two youngsters, a developmentalist discourse is also appropriated, to some extent, as a way to express their frustrations, needs and aspirations, as Bordonaro (2009) also found amongst the Bijagó youth. An imported view of technology was,
Temporal Connections
85
to Yasin, Dembó and others, a channel through which to succeed in their ‘development project’ in Europe. Studying computers was, for them, a way to escape the feeling of koitadesa and to participate in a wider project where acquiring knowledge played an important role. The concept of modernity has also been accused of embeddedness in neo-colonial constructions of the world and of working as an ideological tool in global power relations (Escobar 1995). However, as some scholars have argued, people are unlikely to fully resist or criticize the technologies and conveniences of modernization (Moore and Sanders 2001: 13) and, especially for youngsters like Yasin and Dembó, for whom related corporeal risks can perhaps be more easily overcome, fears of ‘development’ must not be read as an unequivocal rejection of ‘modernity’ (Mitchell 2001: 5). When considering Guinean foodways, however, as elsewhere in Africa (cf. Wane 2014), the introduction of neo-colonial based technology can be particularly concerning for the people involved. Despite motivating projects of migration like Yasin’s and Dembó’s, when used in agricultural production technology was, for many Guineans I talked to, a risk brought in by ‘development’, due to the absence of an intimate connection with the land that defines it, and consequent generation of unhealthy food and weakening of the body. In his work on food security, Pottier (1999) stressed the fact that developers conceptualize their intervention amongst food-insecure people in ways that resemble colonial policies, assuming that people want to ‘progress’ and ‘develop’ by moving away from what they consider ancient practices that do not suit modern conditions. The official debate on world food security, Pottier contends, has become divorced from the realities of food-insecure people. They may, then, resist change, when it threatens their ecological and social worlds (Wane 2014), or respond in inventive ways when those threats are inevitable, as they are amongst migrants, by, for example, endowing exogenous foods with local meaning in order to minimize the bodily weakening that those foods generate (chapter 2). This debate is useful to help understand how cosmopolitan performances, such as that illustrated by Adama at the start of this chapter, are not necessarily a sign of a kind of western modernity or, as Piot (1999) has argued, that cosmopolitanism is as much African as it is European. If by cosmopolitanism we mean that ‘people partake in a social life characterized by flux, uncertainty, encounters with difference, and the experience of processes of transculturation’ (ibid.: 23), then Africans’ encounters with elaborate systems of solidarity, demands and obligations, constant migration of relatives and also with the spirits of the ancestors make them as cosmopolitan as any European. Moreover, cosmopolitanism, in the case of migrant traders like Adama, can be strategic and temporary, emerging ‘out of the need for vulnerable individuals and groups to make a living in an environment characterized by insecurity’ (Kothari 2008: 500), hence being part of a lived reality that disrupts the predominantly western view of cosmopolitanism. As
86
Food Connections
Moore (2011: 74) has questioned, and in line with the discussion on African technology, why do African subjectivities always seem to be transformed by what comes from outside, whereas nothing new seems to come from the continent? Drawing on Werbner (2002), Moore also questions the notion of ‘new’ in current postcolonial studies, as if interdependence and mutual entanglement had not been part of African subjectivities before. Conversely, however, as Yasin’s and Dembó’s aspirations also show, saying that everyone is cosmopolitan might obscure the experience of differentiation and marginalization that coexists with conviviality (Fardon 2008: 252). For Guineans at home, the desire for migration is a result of that experience. It is therefore influenced both by objects and tales that express cosmopolitanism and modernity and that travel from Europe to the home country – the financial and social remittances analysed by scholars like Levitt (1998) – and by the ‘traditional’ food and mesinhu they send in the opposite direction, which will offer them protection upon arrival. At the receiving end, too, return projects are built through the movement of food, as I will now explore.
Between Locations: The Dilemmas of Return Just as migratory projects were central in conversations with people like Yasin and Dembó in Guinea-Bissau, discussions with Guinean food traders in Lisbon were frequently about return aspirations and more concrete return projects. I also encountered several of them later in Bissau, when they were temporarily visiting. Returns generate privileged situations from where to observe how the two lifeworlds that migrants inhabit come together. In this approximation, there is a clear interplay, amongst Guineans, between performances of success and wealth on the one hand, and threats of djanfa on the other, embedded in jealousy, gossiping and the perception of ‘being stuck’ for those who stay. Food and mesinhu interfere greatly in these experiences, as does the material of the environment where these homeland products are produced, exchanged or consumed – the spaces materialized by the travel of food. The airport of Lisbon, like that of Bissau, is an iconic place that materializes the link between the two homes, entwining past, present and future (Abranches 2013a). For first-generation migrants, homeland food is one of the first things to be missed and one of the reasons highlighted for a desired return. Packages of Guinean foodstuffs arriving at Lisbon airport three mornings a week evoke the longing for home and an imagined return from a visit to Bissau amongst the migrant food traders who collect them. Encouraged by the materiality of that space of constant arrivals of not only people, but also food, return projects were discussed at length, often accompanied by a recently arrived bag of products such as cashew nuts, which could be immediately opened and shared between those nearby, while waiting for further packages to arrive.
Temporal Connections
87
On one of those mornings, Aua – Kadi’s sister, and Yasin’s aunt – arrived in Lisbon with her nineteen-year-old son and three trolleys filled with luggage, after a five-month return trip to Guinea-Bissau. Kadi and other relatives welcomed her on arrival. Following an excited exchange of greetings, Aua opened her bags and distributed the contents with her sister: mangos, Guinea gum vine, okras, roselle leaves, different sorts of dried fish and seafood and, amongst other things, a five-litre container of palm oil which had leaked in the bag. This episode took place before Yasin’s successful migration, and Aua had the photos I had taken of the two of them in Bissau. While trying to sort and distribute the food products spread at the entrance hall of the airport and hurriedly clean the red stains from the palm oil left on the floor, Kadi looked at her daughter Yasin’s pictures with an expressively longing smile. Like her sister, she wanted to spend a period of a few months at home. In her case, however, a return was not advised by doctors in Portugal, due to her unstable health condition resulting from a snake bite during her last visit home in 2005. ‘There is more to be said on that matter, though’, she considered, before continuing, ‘I had a good life here until 2005. Someone who was jealous of me there made the snake bite happen’. This reference to jealousy and djanfa highlights, on the one hand, the imagination of an easy accumulation of wealth in Europe held by relatives in Guinea-Bissau. Migrants’ display of material objects such as the houses and cars mentioned by Dembó above help to shape that imagination. On the other hand, the threat of djanfa adds to the economic, legal, bureaucratic and political obstacles imposed on migrants’ return projects. ‘I can’t go back to Guinea-Bissau now because of someone’s evilness’, Kadi concluded with a longing expression, still looking at the photos. Most of the time, however, the home-based kin’s imagination of a wealthy life in Europe did not correspond to the everyday experiences of the migrant abroad who, rather, tends to lose social status in societies of destination. This discrepancy, defined by Niewswand (2011) as ‘the status paradox of migration’, was named by a Guinean migrant food seller at the airport one day as ‘the emigrant myth’. As he explained: If one goes back to Guinea-Bissau to make business, one has to start slowly, due to the emigrant myth. When I lived there, I didn’t have any problems with carrying heavy stuff, for example. Now, if I do it there, people will talk. They will think that I should hire someone to do the heavy job. They don’t know that the work I do in Europe is much heavier, and they create all this confusion around it. Also, if one goes there and stays for more than two months, they will start wondering, ‘why doesn’t he go back to Europe…?’ This is the emigrant myth. The ambiguity therefore lies in the fact that, if return migrants are not expected to do hard work due to expectations of social mobility with
88
Food Connections
migration, adopting exaggerated acts of worldliness is not a welcomed practice either. Ferguson (1999) identified similar dilemmas amongst Zambian urban workers who returned to the village, finding themselves obliged to disguise their cosmopolitanism. As with the Zambians that Ferguson worked with, powerful and effective sanctions are imposed on Guinean returnees who ‘show off’, consisting of ostracism, gossip, withholding of help when in need, burning down of houses, assault and djanfa, such as that experienced by Kadi. Moreover, as Binsbergen (1988) also found amongst Manjaco returnees in Guinea-Bissau, disguising one’s cosmopolitanism could be a way to avoid demands from relatives at home, who count on their migrant kin’s visits to provide them with extra gifts and pay for expenses. This – which Guinean migrants referred to as pidi, pidi (ask, ask) – was, more than a conscious reason to avoid cosmopolitan attitudes, another motive often given for postponing returns. While planning or imagining their return, Guineans were conscious of the conflicts that certain attitudes could cause. Fatu, a Susu migrant woman resident in Damaia who had been involved in the transnational food trade in the past, gave the example of what she named the ‘café lifestyle’ that she adopted in Portugal – an expression that many use as a portrayal of a Portuguese way of life – as a practice that would not be well accepted in Bissau. ‘I got used to this café lifestyle, but people in Guinea would talk’, she said. ‘They immediately recognize someone who came from Europe, even if you dress in African clothes. And they gossip. They really gossip a lot’. Yet in Portugal she had adopted the ‘café lifestyle’ precisely to escape the gossip and rumour that are part of the social life of Damaia’s ‘Bandim market’, where many Guinean residents in that area socialize. Attempts to avoid gossip and rumour, which are based on words that might have the force to make the spirits act more quickly on those who they are directed at (Bâ 1981; Cole 2001), are therefore made at both ends of the migration, in order to keep away from eventual forms of djanfa. However, Fatu and others considered the risks to be higher in Lisbon. ‘Guineans here talk too much, and that only leads to trouble’, Fatu explained. Experiencing a situation of temporary unemployment, she spent the afternoons at a local café looking for job offers in daily newspapers. At the café she knew the waiters and many of the regular customers, most of whom were Portuguese. She described them as ‘mostly old men and women with nothing to do. They feel lonely and just sit here looking at what’s going on. They like me, and I like them. They’re harmless’. Like Fatu, Dála saw a tendency for self-centredness amongst Guinean migrants, which she associated with what many migrants themselves view as an uncontrolled desire for that wealth accumulation that had been part of the migratory project, but was contrary to Guineans’ common solidarity at home. As Dála put it:
Temporal Connections
89
Back home Guineans really help each other, but here they are… Gosh! Here they’re always looking, always talking too much! People are really goodhearted there. But not here, here they change. Here they want more and more, but they can’t find it. So, they always want to know what you are wearing, what you are eating, what you are doing… it’s too much! For these reasons, her daughter Samira’s plan to temporarily return to Bissau was to be kept a secret amongst the migrant community in Lisbon. For twenty-five-year old Sandra – descendent of a complex mix of FutaFula, Pepel, Geba Christians and Cape Verdeans – who had lived most of her life in Portugal and decided to return permanently to Bissau, the potential reintegration dilemmas identified by Fatu – what Robertson et al. (1994) called the effects of the ‘corrosive travel’ that migration embodies – were deeply felt. Four years before we met she had decided to move back to Guinea-Bissau to live with her mother, while studying Law at a private university in Bissau. She hoped to specialize in Environmental Law and find a job in the area, which she considered promising in Guinea-Bissau. Sandra, however, did not have an easy reintegration, and called me constantly to go out for a coffee in prasa, which she considered an opportunity to relive the ‘café lifestyle’ that she missed, and an escape from the things that she found hard to get used to. ‘Just sitting here and having a coffee as we’re having now’, she confessed, ‘is something I cannot do by myself. If someone sees me they will immediately call me bandida [criminal, immoral]. My friends from university only get together at each other’s houses, but I really like going out for a coffee. In Lisbon, my friends and I would spend an entire afternoon just sitting at a café’. Finally, she concluded, ‘oh, I really miss that!’ Sandra’s and Fatu’s preference for a differentiated social space – the café – characterized by a distinct lifestyle, to use Bourdieu’s (1984) terms, but also by consumption practices that carried the risks associated with unfamiliarity, embodied a sort of cosmopolitanism that was not well accepted by their Guinean acquaintances in Bissau. This, however, does not imply a Guinean ‘traditional’ view whereby any sign of ‘modernity’ is punished with gossip or djanfa. On the contrary, cosmopolitan lifestyles are condemned by the same Guineans who simultaneously see the fact of having a migrant relative as an element that distinguishes them from others in a positive way. The interplay between embracing change and maintaining (or recreating) continuities, sometimes in what may seem conflicting ways, is part of Guineans’ daily lived experiences both at home and abroad, and is not necessarily contradictory. Likewise, as discussed in the previous chapter, contemporary practices such as djanfa cannot be seen as either ‘traditional’ or ‘modern’ under a western frame of reference and a colonially inherited idea of ‘development’ (Geshiere 1997). Due to the privileged position of homeland food in instilling proximity between migrants and their land, food and land-related
90
Food Connections
products and practices play an important role in deconstructing these dichotomies and alleviating the dilemmas entailed in the return.
The Role of Guinean Food and Mesinhu in the Return Unlike lifestyles such as the frequenting of cafés, or other material goods like the houses, cars and bed linen described above by Dembó, exogenous food was not viewed as ‘modern’ or ‘developed’, or as a signifier of worldliness or success amongst migrants. On the contrary, it was the endogenous organic method of production of Guinean food – and ensuing connection with the land – resulting from Guinea-Bissau’s lack of agricultural and economic development, that led to its perception as healthier and preferred to kumida di branku (white people’s food), also amongst migrants. It was, too, the reason for return migrants to regain strength through a return to Guinean land and the consumption of Guinean food without the need for adaptations. In fact, unlike the first-generation Korean residents in Japan studied by Lee (2000), I have never encountered a first-generation Guinean migrant who, upon return to Guinea-Bissau, saw their penchant for Guinean food disappear through what Lee identified as bodily transformation.3 On the contrary, upon return, food is one of the first compensations for a period spent away. Yet some displays of success also involved food. When performed by return migrants or during return visits, the practice of feeding others, for example, created an image of success related to the generation of status, as Janowski (2007) also found in Southeast Asia. The example of Ansumane, an elderly Fula trader of Guinean food in Lisbon who was on a temporary return during my stay in Bissau, illustrates one way in which such imaginaries can be shaped and experienced. We had met in Lisbon, where he had told me about the new house he was having built in Bissau – one of the main material embodiments of migrants’ success, as Dembó’s narrative revealed above. When I arrived to see the construction progress, he directed me to the porch, and there we sat together for the rest of the afternoon. At a certain point, a group of children arrived, loudly, and joined us. Gently smiling, Ansumane watched the women go inside and get a bowl of rice with white sauce to give to the children, who sat in a circle near us, eating. ‘I always do simola to the children in the neighbourhood’, Ansumane explained, proudly, referring to his success as a migrant, which gave him the possibility to do so. ‘Every day around this time I offer them a bowl of food’. Ansumane’s distribution of food amongst the children of his neighbourhood exemplifies one way in which local practices can be reinvented as a sign of cosmopolitan success, especially when involving a vital and distinctive material like food (Sahlins 1985). A different performance was once enacted by Samira – Dála’s daughter – at the end of her three-month visit to Bissau, which had been planned in secret
Temporal Connections
91
to protect her from the gossiping of other Guineans in Lisbon. When accompanying her sister to Caracol market in order to buy the foodstuffs that her mother had asked her to bring back to Lisbon, she walked slowly and kept her distance with an uneasy smile and quietness amidst the loud voices of sellers and clients, embodying what could be interpreted as the expression of an outsider. Seeing me, she explained that she was not accustomed to the dynamics of the food market. ‘I never come here’, she said. ‘No, these places… not me!’ Yet Samira’s negative comment did not seem to upset the bideiras who heard her. Instead, they joyfully recalled having known Samira since she was a baby and emphasized what seemed important: ‘she is going to take our foodstuffs to Portugal. Guineans in Portugal can’t live without our food’. The bideiras’ response to Samira’s performance of apparent detachment from this space of food exchange is illustrative of ways in which cosmopolitanism, like tradition, can be brought into ‘focal awareness’ or be momentarily ‘out of focus’ (Munn 1992) according to people’s conscious choices. Playing a distinct role, plant- and animal-based mesinhu are, due to their material and spiritual properties, especially important in alleviating the dilemmas of return. Serifo, for example, was a Muslim Fula whose forced return was considered a result of djanfa, due to his mother’s act of ‘showing off’ the remittances she received from him. After his father passed away, Serifo made the decision to remit equally to his mother’s and his uncle’s (the father’s brother) families. Although there are varied configurations of gendered power relations in Guinea-Bissau, linked to its complex ethnic diversity (chapter 1), the Fula are predominantly patrilineal and patriarchal societies, whereby the father’s lineage retains power. Serifo’s decision was therefore seen by the uncle – who had already seen his marriage proposal to his late brother’s wife, according to tradition, refused by the widowed – as a defiance to his expected role as head of the extended family, and a sanction was imposed. The use of plant mesinhu was as a way to free Serifo from the djanfa that had been put on him: I was caught in Portugal and repatriated to Guinea. When I arrived [in Guinea-Bissau], people told me it had been djanfa – someone from here wanted me to be repatriated, because I was sending all this money to my mother and she was a woman, so my uncle thought that she was just showing it off… I told my mother that she should have been more discreet. There are so many people trying to migrate, and they never manage… I managed, but the others are still trying. She shouldn’t have had that pride… ‘Oh, my son is in Europe’, you see?… When I came back my mother gave me mesinhu to take and to bathe with, so that the evil would go away. Serifo’s reference to mesinhu was not absent from the complex overlap of everyday experiences, meanings, beliefs and strategies that are part of
92
Food Connections
Guineans’ social realities. When explaining the effects of mesinhu in alleviating his return, he started by saying, ‘I don’t believe in those things, but my mother really wanted me to take them. I never believed it’. Then, after a short pause, he added, ‘I have only used it before I went to Portugal because there was a wise man in the village who was really wise; whatever he said that would happen, happened. So I told him I wanted to go to Portugal but I had no money. And his prayers helped me to go’. In this narrative, Serifo is referring to mesinhu used for asuntu (subject), the ‘subject medicines’ prescribed by Muslim diviners and introduced by Infali in chapter 1, which can help solve problems that are not necessarily health-related by drawing on both the spiritual (linked to the land) and religious domains. Mesinhu di saude (plant-based health medicines) were even more commonly used by migrants returning temporarily to Guinea-Bissau, in order to reduce the corporeal weakness experienced while away. These products, used in combination with food, nourish both physically and spiritually those that experienced uprootedness with migration (Gemmeke 2018). When it comes to gender and return, it is important to acknowledge a gap in the literature that seeks to understand African women’s return migration, which partly results from the historically longer migration of men compared to women, and the consequent more widespread return of men.4 Likewise, there is limited knowledge of how returns affect home-based women in Africa.5 The narratives above shed some light on some of the obstacles faced by women wanting to return and on the experiences of those who stay or who return temporarily. The examples of Samira and Serifo’s mother, for example, demonstrate how women might also follow what Gable (2006) argued for Manjaco men – a praise of their worldliness and new wealth resulting from their own or their relatives’ migration and ensuing remittances, despite the risks of djanfa. The inflicted sanction on Serifo as a result of his mother’s attitude can also be seen as an outcome of particular gender relations amongst the Fula, whereby the male lineage of the family is considered a more legitimate receiver of the remittances sent. Yet alongside the important role played by gender – as well as age – in how returns and migration might be desired or lived differently, my findings suggest that the basic cosmology that endows land, food and plants with invigorating and healing properties is a key element for all.
From Portugal: Remembering Guinea-Bissau in Lisbon’s Food Markets In Portugal, migrants’ memories of Guinea-Bissau are contextually and historically made in the new spaces of food exchange and food sharing they inhabit. Several studies of memory have adopted a spatial and material perspective (Boym 2001; Carsten 2007; Cole 2001; Connerton 1989; Stewart and Strathern
Temporal Connections
93
2003), where the importance of the senses has also been emphasized. In the context of migration, Stoller (2009), for example, described ‘embodied memories’ as infused in familiar objects, narratives and senses. Boym’s (2001) concept of ‘diasporic intimacy’ is especially useful for examining Guinean migrants’ practices of remembering home in Lisbon’s food markets, where exchanging and sharing food (examples of that sensorial intimacy through familiar taste and smell) help them to make sense of their lives in a foreign environment. Lambek and Antze (1996) contend that we need to look at the cultural vehicles through which memory is embodied and objectified. Although food is one such cultural vehicle, anthropological studies of food and memory do not abound.6 While existent scholarship has tended to focus on everyday contexts in which food is bought, prepared and consumed as producers of memory (Law 2001; Parveen 2017; Sutton 2001), or on the reproduction of food production practices such as planting and kitchen-gardens that women used to carry out prior to migration (Tolia-Kelly 2010), the relationship between memory and acts of sharing, giving, reciprocating and exchanging food in spaces of everyday socialization has been considerably less investigated. In places like Damaia and Rossio, memories are activated or reinforced through that exchange, which is accompanied, too, by food preparation and consumption practices, such as grilling corn, smoking fish or preparing moni during Ramadan, as seen in the previous chapter. On these occasions, memories are materialized in powerful bodily sensations – feelings of pleasure, but sometimes also weakness and exhaustion that connect the warmth and sweetness of the grilled corn, the touch of the millet flour mixture or the intense rhythmic patterns of their work, with the places, people and events involved in creating those feelings in Guinea-Bissau and in Portugal. Memories are also activated through the reproduction of socialization practices that Guinean food traders and their clients used to perform in the food markets of Bissau. The act of gossiping, analysed above, plays a role within these socialization practices. The news brought in by a newly arrived client was always received with excitement, and often included reports on recent marriages and deaths at home, as well as outcomes of family quarrels or court cases of common acquaintances. News about recent developments in Guinea-Bissau’s tumultuous political life was amongst the most sought after, and most Guinean street sellers and shopkeepers had small radios constantly playing RDP Africa,7 whose programmes always stimulated new conversations on Guinean politics amongst its listeners. Moreover, memories triggered by spaces of food exchange are also embedded in a historical consciousness (Ben-Ze’ev 2004; Sutton 2001), as well as in the experience of land as ethnic territory and its association with specific foods or plants. One Guinean food shop client once illustrated this relationship by narrating the story of some of the neighbourhoods of Bissau – a city that was originally the territory of the Pepel. Ernesto was a client, but
Figure 3.1. Tree bark and roots, palm oil, mangos, baobab fruit and velvet tamarind (dry pulp) for sale in Damaia, with a radio on top playing RDP Africa. Portugal, 2010. © Maria Abranches.
94 Food Connections
Temporal Connections
95
also a neighbour and friend of Fodé, who owned the shop on the outskirts of Lisbon. He had been sitting in silence at the back of the shop, as Fodé shared his latest business-related preoccupations with me. Then, slowly, he entered the conversation, transforming it into an enthusiastic sharing of stories of Guinean ethnic groups and territories, different ancient practices and beliefs, healing qualities of Guinean plants, fruits and roots from a variety of trees, several recipes and methods for different sorts of maladies and lively accounts of past encounters. The history of Bissau, he contended, was associated with the importance of specific trees. As he explained: Pilum [a neighbourhood in Bissau] is called that, but most people don’t know why. It is called Pilum because there was a tree there… the tree that gives a small round yellow fruit called mandiple [spondias mombin]. In Pepel language, mandiple is pilum. So, they would say, ‘I’m going to get pilum’, and when the Portuguese came they gave that name to the area. There are many beautiful stories like that… Mindará, the neighbourhood where Bandim market stands, the Pepel called it Osombra [shade]. Why? Because in the times of our big men, of our ancestors, that was an area with only very big trees. But these names are getting lost now… Ernesto’s narrative is illustrative of Appadurai’s (1996) notion of ‘nostalgia without memory’, evoking the way in which memories are embedded in oral histories that reconnect descendants with their ancestors in ways that make people look back to a world that they may have never known, therefore never lost (ibid.: 29–30). It is also illustrative of the way in which memories are embedded in the history of a colonial past (Cole 2001). In other spaces of food exchange and socialization around Lisbon, similar associations between food, space and memory could be observed. Cidália and her husband, who I met in Damaia’s ‘Bandim market’, introduced me to one such location. They were sitting in Aminata’s regular selling place when I first met them, helping to watch over her mesa (stall) during a brief absence of the young vendor. Yet they, too, had a bag of smoked chicken for sale by Cidália’s feet. Cidália explained that her husband specialized in smoking chicken, and she in selling them. ‘I sell these in Monte Abraão market on Saturdays’, she told me. ‘I shall take you there sometime’. I visited Monte Abraão with Cidália a few Saturdays later. It consisted of a weekly open market in one of Lisbon’s periurban municipalities, not far from Damaia, and was another space of performative food memories for Guineans, although used with less regularity. Cidália considered this a business opportunity, since competition between Guinean sellers was lower here. This opportunity, alongside the occasion for socializing it generated, made her prefer this place of social and economic activity. As she put it, ‘we don’t have a licence to sell in this market, but here we can sell discretely, to
96
Food Connections
our parenti (relatives), while we djumbai’.8 When we arrived, however, her husband Braima, who was already there with a group of Guinean friends, kept the bags of smoked chicken out of sight, under a table, at an outdoor bar near the entrance. His occasional quick scan of the area suggested that, alongside the advantages of this market for socializing and exchanging, a certain anxiety was also part of the experience of selling without a licence, which resulted from the need to stay on constant guard and alert, materializing the ‘somatic modes of attention’ coined by Csordas (1993). At the bar, beers, draught wine, bifanas and burgers were being served. Cidália was the descendant of a complex ethnic mix from Bolama and, being Christian, she enjoyed drinking wine and eating pork sandwiches (Portuguese bifanas). The Guineans that gathered around the bar at Monte Abraão on Saturdays were, unlike the predominantly Fula and Mandinga that dwelled in Rossio and Damaia, of different ethnic and religious backgrounds. Although people like Cidália and her husband moved through all these different spaces, this was a weekly occasion for others, who did not use Rossio or Damaia on a daily basis, to socialize. I joined them at the bar for the rest of the morning, during which several Guinean clients and friends stopped to buy Cidália’s chickens or sell homemade baobab fruit ice-creams to the children. These interactions, which were conducted mostly between Guineans amongst an apparently inattentive Portuguese majority at the bar, were only rarely limited to the economic exchange. Although constant energy seemed to be dedicated to finding clients, making them attend to their own body movement – which was, in this case, cautious and discreet – and to the embodied presence of others in the surroundings (Csordas 1993), memories of similar djumbai in which they used to participate in Guinea-Bissau were often activated. Braima and Cidália explained that the place was known amongst them as bar di Tuga (bar of the Portuguese), while others called it bar di Guineense (bar of the Guineans) due to their regular presence there. On the back wall, a poster with the words Rota dos Sabores (Route of Flavours) could be seen alongside the flags of six Lusophone countries: Cape Verde, Mozambique, São Tomé e Príncipe, Guinea-Bissau, Angola and Brazil. Space was used here as a material environment for memory practices and also for creating a sense of belonging, which both Portuguese and Guineans did in relation to each other and to others (Barth 1969; Eriksen 2010; Wimmer 2008), through practices of socialization that involved the consumption and exchange of different types of food. Yet the materiality of this particular space and the reference to ‘othering’ that Cidália used to describe it also supports a critical view of Portuguese postcoloniality – a type of postcolonial construction attempting to claim that: [t]he material Portuguese empire has now dissolved into a transnational and heterogeneous symbolic community of Portuguese-speaking
Temporal Connections
97
nations linked by common historical, linguistic, and cultural bonds, under the umbrella concept of Lusofonia [Lusophony], which is institutionally embodied by the CPLP or Community of Portuguese-Speaking Countries. (Arenas 2015: 358) This idea of a ‘community of nations’, represented in the flags at the bar, follows a deep-seated myth of Portuguese exceptionalism through the term ‘Lusotropicalism’ – the perception that its colonial enterprise has been more benign and racially tolerant as a colonizer than other European powers (Anderson, Roque and Santos 2019). Yet this stands in sharp contrast with experiences of racist attitudes, such as the following, described by one of Braima’s friends: Once, on a bus in Lisbon, a man told me, ‘go back to your country’, but I told him that I could bet I knew more about this country than he did. For better or worse, I was born in Portuguese land [colonial Portuguese Guinea] and, until now, I remember the names of the rivers and railway stations of Portugal. I learned them at school, even though I didn’t even know what a railway station was back then. Then, smiling, he added, ‘the elders on the bus agreed with me’. Being from Bolama, like Cidália, he exemplified, as others from the same region, the complex way in which ethnic imagination is constructed by Guineans from the island, where the history of the tchon’s first owners has more nuances. ‘My father is Manjaco and my mother is a Geba Christian. I have Portuguese grandparents. That thing about race doesn’t make any sense to me. I don’t have a race’, he said. This story illustrates, on the one hand, a particular situation of racial discrimination. On the other hand, it exemplifies the complex spatial and historical configuration of memory, where the consciousness of a colonial past might be used as a tool with which to deal with that same discrimination. I will now examine these configurations in more detail.
Memory and the Colonial Past The social and historical constitution of memory emerges in remarkable ways when we look at the complex processes of remembering and forgetting a past of violence associated with the colonial period (Cole 2001). While several scholars have been concerned with a postcolonial ‘memory crisis’ and the striking boom of ‘colonial nostalgia’ in Africa (Bissell 2005; Ferguson 2002; Stoller 1995; Werbner 1998), others have more recently drawn attention to the need to focus instead on ‘colonial forgetting’ (Hannoum 2019) or a ‘politics of amnesia’ (Fonkoué 2019) – processes that result from a removal of memories
98
Food Connections
of oppression, violence and humiliation from dominant postcolonial narratives. As Hannoum notes, forgetting is ‘neither a dysfunction of memory nor an absence of recollection’ (2019: 369), but, instead, a strategic way to reorganize memory that serves the present by giving meaning to events or controlling their contingencies. In the example narrated above by Braima’s friend, memory of the colonial past, removed of violence, can be seen as a way to respond to, and cope with, the violence of the present in Portugal. On several occasions during fieldwork in Lisbon, I heard accounts where the colonial period was narrated as a better time. Often, sitting at their mesa, Guinean food sellers and their clients discussed tempo di Tuga (the time of the Portuguese) as ‘the good times’, expressing critiques of Guinean post-independence politics and the idea that ‘the Portuguese left us all good things’, while ‘the PAIGC did nothing’. For Guineans, however, this ‘nostalgia’ was not for the colonial economic, political and cultural repression, which was, in fact, removed from their narratives. Likewise, the widespread neglect from Portugal towards then colonial Guinea, where the population saw hardly any improvement in living standards, infrastructures, local manufacturing, health or education (Galli 1995; Lobban and Mendy 1997), was also ‘forgotten’. Instead, it was a way to criticize the current unstable political situation of Guinea-Bissau, which they described as a cause for embarrassment and shame, contrary to the sense of satisfaction, wellbeing and pride that Guinean food brought to them. Dála described her memories of the colonial past from the perspective of her own participation in the independence war as a washerwoman for the Portuguese, emphasizing the role of Fula women under colonialism in Guinea-Bissau – a role that was linked to the Fula chiefs who served as tax collectors for the colonial state (Idahosa 2002). She explained: I like to work. Ever since tempo di Tuga, I left the house to be their washerwoman, so that I could earn some money and help my family. Half of the Guineans were with the Portuguese. The other half – the Balanta and other tribes – would ambush us and set the bush on fire, and we had to run home. Then a Portuguese soldier stayed and guarded the place while we washed. I was in the war from the beginning to the end, for 13 years, washing the Portuguese’s clothes! Every day we’d go to the bush to wash, since there was no well in the house. We made the war of the Portuguese, we worked so much for them – every day washing, ironing… every day! Alongside ‘nostalgia’ for the colonial past, this narrative also seems to suggest an acclaimed memory of the protective Portuguese soldier, against the danger of the ‘Balanta and others’. In the same vein, while Dála had not yet received the compensation she was entitled to for her work for the Portuguese during the independence war, she censured other Guinean
Temporal Connections
99
women for their lack of solidarity, rather than seeing the Portuguese state as responsible for her present situation: I once asked a lawyer, and he told me I should get witnesses. And there are so many others here [in Lisbon] who were also washerwomen! The thing is that there they were known by a different name, which is not the same name by which they’re known here. Sometimes, if I manage to identify one, she says, ‘fine, fine’, but never does anything. Here people don’t help their compatriots. They change with migration. It was not until later in the conversation that the PAIGC was presented as also to be blamed for the difficulties she currently faced, but there was no reference, in her narrative, to colonial repression:9 ‘I was asked if I had a washwoman’s card. I had the card, I had a Portuguese passport, a Portuguese ID… Everything! But when the PAIGC came they took it all. They took all our documents’. This complex configuration of memory, which, purged of memories of the brutality of colonialism, can be a strategy of resistance for migrants, is thus also a way to condemn the failure of the nation-state after independence and in present-day Guinea-Bissau that, as in other African countries, continues to benefit mostly national elites (Hannoum 2019). Such critiques are, much like Hannoum observed amongst an older generation of Moroccans in France, including his own father (ibid.: 378), motivated by the distance between the majority of Guineans and the politico-military elite that have continued to use their positions for personal gain rather than furthering democracy and working for the improvement of people’s living conditions (Embaló 2012). It is precisely the widespread sense of foreboding and ever more precarious daily life that people are facing, in which the neo-liberal reforms and the negative consequences of the Structural Adjustment Programme played a role, that makes them idealize the past. Amongst the bideiras of Bissau, similar manifestations of resentment regulated by ‘then’ and ‘now’ were commonly present in daily conversations. Yet while this resentment was often expressed in the corruption and political instability of the present, there were rarely concrete examples of what was better for them in the past. On 1 April 2010, a failed coup d’état took place in Bissau. When I went to Caracol food market the following day, it was operating with no apparent disruptions. The permanence of the aromas, tastes and textures of the food products, as well as the aesthetics of their display at the women’s mesas, seemed to compensate for the political turmoil of the previous day. A group of women food sellers and their clients discussed the incident around a large bowl of chicken and rice, while claiming to be used to such ‘political and military foolishness’ and to the ‘unwise people who run the power’. Teresa, who was visiting from Lisbon, seemed more upset at the event. She shook her head with dismay and said, ‘I was
100
Food Connections
never in Bissau before when such things happened. Yesterday I was already here [at the market] when it all started. I had to leave hurriedly and walk all the way back home. It was scary’. The others, however, reassured her, and moved on to discuss her son’s marriage and what was still missing – only a few drinks now – in the kabas to be sent to the girl’s family. In Bissau, as in Lisbon, expressions of nostalgia for the past were more evident amongst the Fula. This, however, cannot be viewed only in line with the cooperation of many Fula with the Portuguese and their shared historical interest in fighting against the ‘animists of the coast’ (Lopes 1982; Teixeira da Mota 1951), as explained in the introduction. More than that, as Bissell (2005: 240) has argued for the Zanzibari, it is associated with the imaginative resource that the colonial past provides, ‘a realm rich in invention, critical in possibility – for people struggling with the present, hoping to secure what can no longer be found in the future’. Nostalgia, like other forms of memory practice, can only be understood in particular historical and spatial contexts, and does not necessarily mean the wish for a return to the past, of which concrete positive remembrances cannot actually be identified, illustrating what Shaw (2002) defined as the difference between discursive and practical memory. Between both poles, Shaw contends, there are several intermingled forms of memory that do not necessarily contradict each other. For Guineans, too, ‘colonial nostalgia’ is used strategically and contextually, interwoven with diverse social memory practices and with the historical myths produced by ‘Lusotropicalism’ – the idea that Portuguese colonialism was relatively peaceful. To Braima’s and Cidália’s friend in Lisbon, the consciousness of colonization was a strategic tool against discrimination. To Dála, colonial memory might have been a way to deal with the difficulties faced by the lack of solidarity amongst her fellow Guinean migrants, as well as with the difficulties that she went through after independence. Often, this ‘colonial nostalgia’ was combined, in Lisbon, with a more general longing for the homeland, which was brought forward through powerful sensory elements like the smell or touch of food and related practices of preparation, consumption or exchange. Just as this longing was selective and episodic in order to help the community in diaspora make sense of a new environment (Parveen 2017), so was ‘colonial nostalgia’ part of a strategy and desire to simultaneously forget and remember. Feelings of nostalgia were also, at that time, exacerbated by the economic crisis in Portugal and the deterioration of migrants’ living conditions. This interplay, where life in Guinea-Bissau is simultaneously longed for and its politics criticized to the point that colonial times are portrayed as better, is part of the apparent contradictions that are, however, not perceived as such by Guineans themselves, who find their own ways to reconcile them. In Guinea-Bissau, in particular, alongside nostalgia for the past, there seems to be a strong ‘nostalgia for the future’ in Piot’s (2010) terms – a future that replaces untoward pasts, and that is partly embodied in widespread
Temporal Connections
101
migratory projects. During fieldwork in Bissau, in fact, I shared Piot’s struggle to understand what he called ‘the paradoxes of everyday life in contemporary West Africa’. Indeed, Guineans’ desire for displacement, partly resulting from political disappointments of recent decades, was reconciled with a proud experience of the land. As for the Togolese studied by Piot, the land was, for Guineans, lived ‘fully in their skin and infused with hope and dynamism’, and people filled the streets and the everyday with ‘hustle and laughter’ (ibid.: 20). To understand such paradoxes I draw on Moore’s (2011: 144–45) suggestion that hope has a temporal character that forces us away from a history of ideas which is too strongly attached to rationalities and ideologies, and orients us towards possible futures. *** By tackling memories and future aspirations of migration and return amongst Guineans involved in food-related practices and activities in Bissau and in Lisbon, we can better understand the need to reconceptualize home as dynamic processes ‘involving the acts of imagining, creating, unmaking, changing, losing and moving “homes”’ (Al-Ali and Koser 2002: 6). This chapter has explored the reconfiguration of memories and imaginaries that takes place with migration, adopting a phenomenological-oriented approach that understands them as temporal experiences, embodied in everyday materials like food, rather than abstract ideas that only exist in the past or in the future. Migrants write their presence in the new spaces they occupy while imagining possible futures and being simultaneously nostalgic for the past – in this way maintaining not only the link between proximate and distant homes (Blunt 2003; Carsten 2007), but also between past, present and future. Return projects also help to bridge these spatial and temporal distances in the making of Guinean lifeworlds. Returns can be imagined, provisional or repatriated and, like migration, they take place over space and time, developing their own histories and trajectories (Long and Oxfeld 2004). Return projects, like migration, reconcile past memories with new tensions, creating new transnational spaces of what Rouse (1991) named ‘pluri-local’ homes. Returns, and the performances they embody, also help to shape migration aspirations amongst those in GuineaBissau, who draw on a two-sided system of solidarity (Riccio 2008) whereby having a migrant relative or acquaintance can provide them with support and status but also competition, conflict and even djanfa imposed on the migrant – or prospective migrant – as was the case with Serifo’s forced return, Kadi’s inability to return, or the obstacles to Dembó’s attempts to migrate. In this process of collapsing space and time, food plays a key role. As Cook and Crang (1996: 131) have contended, foods embody materials and practices ‘inhabiting different times and spaces which, far from being neatly bounded, bleed into and mutually constitute each other’. It is therefore important to look not only at the materiality of food, but also of the practices
102
Food Connections
and spaces of food exchange and consumption in Lisbon and Bissau, which play a role in the process of keeping an active link between both locations. Unlike Counihan (1984) and Seremetakis (1994), whose respective work on Sardinian and Greek food indicate that the onset of modernity has resulted in a decline of customary foodways and therefore in the decline of food as a vehicle for social memory, I argue that, even with the need for adaptations, maintaining their foodways is a way for Guinean migrants to guard against loss. For Guineans at home, too, it is a vehicle with which one can reconcile desires and fears. Within the temporal and spatial link that it creates, food is an object of historical consciousness (Sutton 2001). Its related activities of sharing, consuming and exchanging are embodied in space and history, giving meaning to people’s imaginaries of the future and perceptions of the past, therefore making new ways of being (Moore 2011).
Notes 1. The roselle red peduncle fruit (ondjo) and a sauce made of lemon, onions, maggi stock cubes and chilli (konserva). 2. Bazin is an expensive dyed fabric of Malian origin. 3. I heard, however, similar stories of bodily transformation occurring in Guineans who had migrated as young children. A couple of return families reported having had to take their children to hospital in Bissau due to being seriously sick after the ingestion of kaldu di mankara, skalada or baguitche, ‘because their body wasn’t used to it’. 4. Filipino female migrants are an exception, and studies of their return visits can be found, for example, in Constable (2004). 5. Even more widely studied gendered subjectivities in transnational contexts, such as women’s requests of gifts and money from migrant men, have more often concerned other regional contexts (cf. Strathern 1988), as has the women’s act of reminding migrants of their moral duties through gifting of food (cf. Abbots 2011). 6. See Holtzman (2006b) for a review of scholarship on food and memory. 7. Portugal’s public service broadcasting in Africa. 8. A Creole term meaning ‘to be together’ or ‘to socialize’, which often includes the sharing of food and is, amongst migrants, an occasion to share memories from home. 9. The PAIGC formed the armed independence movement and remained in power during the one-party state period that followed independence until 1994, as well as from then on intermittently amidst strong political instability.
CHAPTER 4 TRANSNATIONAL EXCHANGE OF FOOD Gifts, Reciprocities and Trade
‘Velho!
[Old man] Do you know Jorge’s place around here?’ The taxi driver pulled the car over and asked for directions to Jorge’s agency, in one of Bissau’s peripheral neighbourhoods. The place was well known to many with whom I had spoken, who used it for sending homeland gifts (mainly of food) to their migrant kin, and for regularly collecting financial and material remittances that were sent to them. Although it was not immediately easy to find, the taxi driver eventually located the place, surrounded by a small market, a bar serving drinks and a local restaurant where I was to share meals with Neusa – Jorge’s sister-in-law – and her cousins and clients, in the months to come. The agency consisted of one large room where suitcases, cardboard boxes, plastic bags and other types of wrapped belongings were piled on shelves, on chairs and on the floor, either to be collected by clients in Bissau or to be sent by air to Lisbon (figure 4.1). Old car seats furnished the customers’ waiting area and the staff’s improvised office space near the window, where transactions were registered and packages organized. Jorge, despite giving his name to the agency and being in charge of the business, had been a migrant in Lisbon for nearly twenty years and was therefore on the other side of the transaction. In Bissau, it was his sister-in-law, Neusa, who took care of the agency with the help of her younger brothers and cousins. In their agency – one of many spread around the city – clients regularly came in to collect packages of clothes, shoes, jewellery, drinks (mainly wines and spirits), building materials and electronic consumer goods such as TV screens and mobile phones, sent from Lisbon by their migrant relatives. Not as immediately perceptible as those large packages but more significant for their quantity and regularity of arrival, were money and envelopes with documents linked to people’s migration plans. Necessary for visa applications,
104
Food Connections
Figure 4.1. Neusa’s agency in Bissau, 2010. © Maria Abranches.
Transnational Exchange of Food
105
certificates and records obtained in Portugal and sent to the applicant in Bissau materialized the kin’s help in the process. In the reverse direction, Guinean food predominated. On the day of my first visit, as well as on all other days with direct flights to Lisbon, Neusa’s clients came in with a variety of foodstuffs to be sent by air to their migrant relatives that night. Palm oil bottles, small packages of fresh vegetables, cashew nuts, groundnuts, fresh fruits and homemade konserva could be seen inside their bags, wrapped and labelled by Neusa’s helpers after being checked and weighed. The agency, like others that I visited on different occasions, was always busy on those days. On non-flight days, however, the reduction in the number of clients did not seem to lessen Neusa’s and her cousins’ restless movement, alongside constant phone calls and discussion of business-related issues, such as delays in payment or in the arrival of packages. It was not usually until midday, after a bustling morning, that they found the time to sit outside for breakfast – sometimes a rich dish of kuduru (bread) with goat meat in an onion sauce in order to provide their body with the energy required for such intense activity. On the destination side of the journey, Guinean migrants regularly dropped in to Jorge’s agency in Lisbon to ask if their food packages had arrived. A large front room with two desks and a back room where the material waiting to be sent or collected was stored served as his Lisbon office. Any delays in the regular arrival of food provisions were faced with suspicion by his clients, indicating that alongside solidarity and reciprocity, the act of receiving food from Guinea-Bissau was also, for migrants, a way to maintain a stable link with home and remind their kin of their duties towards them. Jorge’s clients shook their head in discontent when faced with failure regarding those obligations. ‘My wife must be cheating me’, I once heard a regular client say after insistently asking, ‘cashew nuts, groundnuts… Nothing? Has nothing been sent to me?’ The complexity involved in West African structures of kinship, which combine solidarity, obligations and demands, can be understood in a unique manner when investigated through the process of giving and reciprocating objects across borders, which agencies such as Neusa’s and Jorge’s facilitate. Yet the motivations that enable Guinean food to reach the migrant community in Lisbon go beyond processes of transnational giving and reciprocating, with the sending of food products for profit occupying a predominant role in its circulation. However, as others have noted (Appadurai 1986; Binsbergen 2005; Parry and Bloch 1989), an approach that opposes the spontaneous spirit of gifts to the calculated spirit of commodities is oversimplified. This chapter examines the relationships that are created from the need to make Guinean food circulate from Bissau to Lisbon as a gift, as much as from its exchange as a commodity, demonstrating that the role of food as one thing or the other is often entwined and not easy to distinguish. In the stories that
106
Food Connections
follow, I place the focus on transnational experiences of exchange that, in turn, help to maintain the connection between production and consumption.
Cross-Border Food Gifts and the Market Economy By the time I got to Neusa’s agency late one morning, she had already packed three large bags with foodstuffs to send on that night’s flight. They weighed twenty-three, eighteen and fifteen kilos. Her clients – the migrants’ relatives – looked at the piled bags from the waiting area, satisfied at having completed the tasks they were responsible for, although anxious about the risks that, in the remainder of the gifting process, could still inhibit a safe arrival in Lisbon (chapter 5). They also described a sense of tiredness – not only from the anxiety entailed in the process, but also from the physical demand of their tasks. As usual, they had started early in the morning. Their activity, they explained, consisted in early phone calls to order the fruits and vegetables directly to the bideiras from whom they regularly bought the produce, in order to ensure availability in the required quantity. A trip to Caracol market had followed, and from there to Neusa’s agency, where the produce would be weighed, packed, labelled and sent to Lisbon through carriers found at the airport. One client came in late with a small plastic bag of badjki, kandja and djagatu. The three parcels ready for shipment were already full, but one of Neusa’s cousins opened one, compressed it and, to the woman’s relief, managed to fit in her order. The least experienced could be faced with unexpected obstacles. A young woman had been informed that day by Neusa that the kuntchurbedja (pounded seafood) she needed to send to her sister could not be accepted, as it risked being withheld by customs authorities upon arrival. Visibly concerned, the woman was sitting in the waiting area, making several attempts to call the migrant sister who, however, was not answering the phone. ‘I will have to go to the airport and try to find a carrier myself’, she announced, her body tense from the perceived risk that ‘[her] sister would get upset’ and the consciousness of the effects that it could have on their relationship. Specific requests of food, as in this case, were often made by migrants. Two other young women also had an order to send to their sister in Lisbon, who was pregnant and ‘craving for homeland food’. Yet the two large plastic bags of foodstuffs they had given to Neusa’s cousin for weighing turned out to be too heavy, and the cost of shipping them was beyond their budget. Like the previous client, they called their sister in Lisbon and, when able to reach her, negotiated and discussed what should be prioritized. ‘Foli needs to be sent!’, the two sisters exclaimed, referring to the heavy fruit (Guinea gum vine) whose consumption is considered beneficial for pregnant women, as they apprehensively removed other produce from the bags until they reached the agreed two and a half kilos.
Transnational Exchange of Food
107
When talking about their role as food givers and their consequent responsibility for the corporeal and mental wellbeing of their migrant kin, Neusa’s clients were also aware that, in some cases, part of the food sent would be sold to other Guinean migrants in Lisbon. Income from those trading practices abroad would rarely reach them directly, but financial and material remittances were expected as a return (Abranches 2014a). This was one of the reasons for apprehension and concern when faced with obstacles with shipping, since failure in sending food could result in withdrawal of reciprocity. In some cases, they were able to keep part of the money sent by their migrant relatives for the purpose of buying food. At the destination side, in turn, some of the food that arrived with trading intention also ended up as gifts shared amongst fellow vendors, an experience that alleviated them from the tension and risks of their trade activities abroad (chapter 5). These practices confirm the unsuitability of the gift versus commodity dichotomy in the context of food and migration in West Africa. A critique of this dichotomy – and a more attentive look at the nuances that relate to both concepts – has predominated in anthropological studies of exchange since the early analyses of the Pacific region (e.g. Bourdieu 1977, 1990; Strathern 1985). Many now argue against an oversimplified view that opposes the spirit of reciprocity, sociability and spontaneity in which gifts are typically exchanged, to the profit-oriented, self-centred and calculated spirit that fires the circulation of commodities (Appadurai 1986; Binsbergen 2005; Parry and Bloch 1989). Within these critiques, Appadurai (1986) and Kopytoff (1986) focused not on the meanings of commodity, but on the processes and contexts through which a ‘thing’ becomes one, which can change in its life course. Migration, memories, aspirations and wellbeing needs – analysed in the previous chapters – are part of the processes and contexts in which Guinean food circulates, affecting its biography, from production to exchange and consumption, in different ways. There is nowadays widespread acceptance that commodities, seen as a phase in the biography of things, are not only a monopoly of modern, industrial economies, and that small-scale societies should not be romanticized (Appadurai 1986). There is also an acknowledgement that a gift-to-commodity sequence has not taken place everywhere, or in a linear manner, and that the spread of markets has not necessarily resulted in homogenization and cultural impoverishment or in the loss of gift-giving economies (Pottier 1999). Following Long (1996), Pottier brings forward the concept of ‘relocalization’, in the sense that changing local conditions are not dictated only by hegemonic capitalist interests, but rather ‘relocalized’ within local forms of organization. In Guinea-Bissau, the increased demand for homeland food generated by migration has reshaped the market and created local opportunities for profit-making. However, the rise of consumer demand, in this case, has transformed, rather than eliminated gifting practices.
108
Food Connections
As a phenomenological-oriented investigation of human perception and experience of food exchange reveals, people are not always separated from objects through commodification, since even with the introduction of levels of intermediary roles such as those performed by Neusa and Jorge, those at the destination end are still connected to food (the object) and to the land’s substance in which both originate. The fact that the origin of much of the food received at the other end was known to the final consumers suggests that physical separation does not always entail relational alienation in largely non-market societies. Hence, although these agencies result from a market opportunity that involves profit, it is also through them that the connection between food consumption and production is maintained, and that acts of giving, sharing, trading and reciprocating are connected, even when they are performed across borders. The practices they stage also indicate the complexity of the relationships that transnational food exchange generates. The trade of food, understood as a practice of exchange that involves profit, entails dilemmas of conciliating solidarity, obligations and personal interests. Agencies like Neusa’s and Jorge’s, for example, were a business opportunity generated by migration, but also by the need for an intermediary role in the transnational gifting system. Guinean migrants who own shops selling varied products aimed at an African clientele, as well as some of the more established migrant street sellers, also fulfil the intermediary function of making food and other goods circulate safely, by charging a small fee to those who want to use their services. These practices blur the boundaries between what is considered a shop or an agency, and a seller or ajensiador. Although they all share the need to make profit, they also help to fulfil the moral obligation to share homeland food with kinfolk. Such obligations, which are associated with the vital functions and unique qualities of food, help to continue the intimacy of existing exchange systems, where money is incorporated without necessarily creating rupture in social relationships (Abbots 2008; Evans 2001; Evers 1994).
The Value of Food in Transnational Exchange Within the commodity debate, emphasis on the exchange as source of economic value was brought forward in the work of Simmel (1978), contrasting with Marx’s central preoccupation with production. Here I am interested in the social value of food and its relationship with everyday experiences of exchange in the context of migration. I investigate the social value of Guinean food from a phenomenological perspective that sees it as the ability to connect people and their land by transforming sensations, memories and imaginaries embodied in food into lived experiences of those connections. Following Lambek (2008), I shift the focus of value from ‘having’ to ‘doing’ and ‘being’ and I argue that the social value of food, and its ability to ‘make’
Transnational Exchange of Food
109
and ‘connect’ people and places, is linked to actual acts and their ethics, which, in this case, include food production, preparation, consumption and – the focus of this chapter – exchange. Changes in the social value of food through practices of exchange, as well as how such practices transform people’s kinship relations, have been explored by scholars like Munn (1986), Uzendoski (2004) and Weiss (1996). Yet adding a transnational perspective to the inquiry, which has been rarer in the literature, can offer new insights into the social value of the materials exchanged. A multi-sited approach to the study of the social value of food, which accounts for both sides of the migration and the different roles involved in its circulation, has only recently started to be taken (MataCodesal and Abranches 2018), although previous key analytical contributions, such as those of Appadurai (1986) and Kopytoff (1986) had already highlighted the importance of tracing things’ shifting status in their circulation through different contexts. With migration, the adaptation of food is visible in practices of preparation and consumption. As discussed in chapter 2, produce unavailability may require, for example, endowing exogenous food with ‘original’ value (Abarca 2004) by preparing it with familiar techniques, which helps to maintain its social value and its identity as ‘Guinean food’. However, the continuity of the social value of food results not only from adaptations in preparation and consumption, but also in the relationships between people at different ends of the chain, depending on the supply networks at their disposition. Small-scale food supply, as Medina and Vázquez-Medina (2018) have explored in the case of Mexican migrants in the United States, is usually dependent on legal status and social circles, whereby only those who are able to move freely – who usually have citizenship or residence status, as well as some accumulated capital – are able to directly obtain the required foodstuffs. Guinean migrants who, for the reasons explored in the previous chapter, are unable to return to Guinea-Bissau, rely on personal networks of friends and relatives – facilitated by the work of agencies like Neusa’s and Jorge’s – for that acquisition. In line with what Medina and Vázquez-Medina (2018) found, such networks are highly valuable in enabling the circulation of produce from specific regions of origin, or from a familiar production site or distant kin bideira, therefore helping to bridge distances and maintain the value of the food received. It was common for Guineans in Lisbon to associate the absence of close relatives with an ‘almost physical pain’ that homeland food (even if adapted) helped to alleviate. ‘I miss my aunt’s bagri’, Samira once confessed with sorrow, as she ate the fish smoked by Carlos in his fumeiro. Others described a surge of sudden joy felt with the arrival of food sent directly from relatives. Fatu, for example, could hardly contain her excitement when cooking an evening dish of kaldu branku prepared with chilli received that morning from her mother, reproducing what Parveen (2017: 127) described as ‘[t]he belief that eating these foods together somehow will take you back to the “motherland”’.
110
Food Connections
‘Production knowledge’ and its impact on value, as emphasized by Appadurai (1986: 41), is as much about knowledge gaps between the production and consumption ends of the chain, as within processes of circulation and exchange. In the case of Neusa and Jorge, however, like other intermediaries in the circulation of food from Guinea-Bissau to Portugal, the contribution to reduce gaps between physically distant consumers and the original source is translated mainly in the overcoming of shipment difficulties, rather than knowledge discontinuities. Unlike in other contexts (cf. Appadurai 1986: 44), the intimate connection between all actors at the different stages of the exchange of Guinean food enables production knowledge – which combines technological and cosmological layers – and, consequently, social value, to be maintained.
Giving and Reciprocating across Borders Seen as part of people’s material world, food, like other worldly things, is embedded in a complex network of relations with people (Heidegger 1962), in which giving and reciprocating play an important role. Giving and reciprocating, as forms of exchange, have been the subject of a lengthy theoretical debate in anthropological literature, ever since the classic works of Malinowski (1922) and Mauss (1950). Scholarly work on the topic has tended, however, to predominate in the Pacific region (cf. Carrier 1992; Gregory 1982; Hagen 1999; Sahlins 1972; Weiner 1985), when compared to Africa. In the early 1990s, Piot (1991) drew attention to this gap in Africanist literature, and suggested that Bohannan’s (1955, 1959) theory of exchange amongst the Tiv of Nigeria, dating from the early 1950s, should be retrieved and its application tested in other African contexts. Bohannan argued that the Tiv divided exchangeable items into ranked categories or ‘spheres’, whereby items in one particular ranking were neutrally exchanged amongst themselves but not with those in different spheres. An inter-ranking exchange would be morally charged according to the value of the item and to whether the conversion was happening downwards or upwards. To Bohannan, however, such spheres were highly vulnerable to contact with European currencies, which were the eventual cause for the standardization of all items’ value and the collapse of the system. Piot (1991) picked up this ranking system and tried to apply it to the Kabre of northern Togo. He concluded that the Kabre exchange spheres exist not so much to group items in categories of exchange, but rather to order types of relationships. Within these relationships, Piot identified the role of the initiator and the respondent and explored the way in which these positions change in a second exchange (or reciprocity), hence helping to maintain relationships through the creation of debts. In Bohannan’s hierarchy, food and other subsistence products were positioned in the lowest ranking. This view was substantiated by authors like
Transnational Exchange of Food
111
Graeber (1996), according to whom wealth – embodied in objects of display and adornment – is the most important material of exchange, and Munn (1986), who suggested a hierarchy in types of goods where perishable and generic substances like food stand at the bottom, and imperishable valuables at the top. Yet for Munn, food might gain value through the act of giving. The fact that not only does the social, spiritual and material value of food change through practices of sharing, giving, reciprocating or exchanging, but also that such practices transform the way in which people create kinship relations, has been explored in the work of scholars like Uzendoski (2004) or Weiss (1996). Sutton (2001: 160) sees the perishability of food as its distinguishing characteristic, since unlike solid objects, it internalizes debt (author’s emphasis), calling for acts of remembrance and reciprocity. In Guinea-Bissau, the act of offering food is key in welcoming guests and is part of people’s everyday experience, either as donors or receivers. Food sharing gains even more relevance with migration, where homeland food is needed to nourish mind and body, and to maintain a connection with land and kin. Armindo, a frequent receiver of homeland food who lived and worked in close proximity to other Guineans in the suburbs of Lisbon, explained: In the neighbourhood where I live, if the neighbours know that I have received food from home, lunch has to be at my place… In my workplace [a construction site], at noon, we all go to the canteen with our lunches packed from home. Everyone starts looking at each other’s containers, to see who has kumida di tera [food from the land]. If I happen to have some, the others will say, ‘today we’re eating food from home!’ and we will all share it. Sali described the act of sharing food as unquestionable and unchangeable, and an intimate part of her experience as a restaurant owner in Lisbon: ‘many people don’t have money to pay, but I have to serve them food anyway. Food is something we cannot refuse to others. From the time when we lived in Guinea-Bissau, we were taught to share food. Everyone who comes to your house is welcomed with food. So this is what we do here as well’. Such descriptions may indicate the vital value of food, which is necessary for subsistence, but also the fact that food is not necessarily perceived as detached from people (Strathern 1985), since both share the same land of origin. Food is, in some ways, endowed with a soul, hence food gift-giving encapsulates and amplifies part of the essence of social relations, which are extended beyond the boundaries of the society to the whole cosmos (Godelier 1999: 105). When food giving is performed across borders, often through intermediaries like Neusa and Jorge, the principle whereby food, given its vital function and distinctiveness by symbolically representing home, hospitality and
112
Food Connections
assistance, is ‘more readily, or more necessarily, shared’ (Sahlins 1972: 215) does not disappear. The obligation of sharing food is as present in GuineaBissau as it is in Portugal and also between both countries. Although there is a gap in the literature on gift exchange in the context of migration, Cliggett (2005), for example, opposed, like Piot, Bohannan’s ranking of gifts, and suggested that for rural-urban migrants in Zambia, giving is less related to the economic value of the gift than it is to a way of investing in social relations over time. For Guineans, sending homeland food to distant relatives is a way of ensuring assistance and reminding them of home, contributing to alleviate the threats of loss and discomfort felt by migrants away from familiar land. The moral economy of food that Bryceson (1993) explored in the context of Tanzania can thus also be applied to Guinea-Bissau, and to migration from there to Europe, where access to homeland food and means of redistribution have to be guaranteed in order to maintain an important source of nourishment and wellbeing, in a context of bodily and spiritual unfamiliarity and uncertainty. However, as Cliggett suggests in the case of internal migration in Zambia, there is also an expectation of a future form of compensation in these acts. For the home-based kin, that return is materialized not only in the remittances received but also in support obtained for the accomplishment of their own migratory intentions. On the one hand, cross-border gifts of food might be, more than a means to reciprocate remittances, an intentional way of anticipating the arrival of money and other objects as a return (Abranches 2014a), in this way maintaining relationships through the continuous creation of debts (Piot 1991) or a mutual exchange of gifts (Malinowski 1922). On the other hand, just as migrants are compelled to remit as a way to remain connected to those at home and keep the path for return open (Cliggett 2005), their kin in the home country might use the process of giving to guarantee support in the accomplishment of their own migratory intentions. Dembó’s story, described in the previous chapter, shed some light on yet another kind of motivation, which anticipates an unspecified future reciprocity. Although the materials he sent to Seco from Bissau were, more than gifts, part of Seco’s requests, needed for his charm-making job, Dembó saw in the sending of those materials a way to guarantee future help in an imagined scenario. ‘I always send him exactly what he asks me to, because tomorrow I might be there and need a favour from Bissau’, he explained. ‘If I fail now, others will fail me later, and I don’t want that, because I know that I will get there [to Portugal] someday. I don’t know how, but I will’. In food giving and reciprocating processes, acts of secrecy and concealment of the material – a hidden exchange of things, as Bercovitch (1994) put it, where people act in a way that keeps others unaware of what has occurred – cannot be applied. This inapplicability contrasts with the transnational gifting of other objects and with the movement of people itself, which often involve secrecy in order to avoid gossiping or jealousy. Rosa,
Transnational Exchange of Food
113
after having been asked, without notice, to pick up her younger brother from the airport in Lisbon, complained, ‘I didn’t even know he was coming. You know Guineans… they never tell you. And he hardly brought me any food. And what he brought, I still had to share with my sister. In the end I only took home a handful of groundnuts, one gum vine, and one mango’. She commented with disappointment on the fact that when a relative arrives unannounced from Guinea-Bissau, often precisely to avoid exaggerated demands of gifts from home, an opportunity is missed to ask for homeland food, and what is brought is, in her view, usually insufficient. Unlike food, financial remittances involved, like the movement of people, more concealment. Despite examples of display like that performed by Serifo’s mother, described in the previous chapter, their arrival in Bissau was frequently hidden in order to avoid the gaze and envy of others, which could lead to insistent pleads for loans or, in the worst cases, to practices of djanfa. In Lisbon, however, the arrival of food was rarely hidden, and its sharing with others was an obligation that, as Rosa described, one could not escape.
Trading Food: Credit, Contracts and Relationships In line with food sharing obligations in and between the two countries, the moral duty to ‘lend’ food products amidst sellers and clients – which, although differing from the notion of ‘sharing’, would often not be paid back – was common in Bissau and in Lisbon. While there were frequent protests against the constant requests for ‘food loans’ from customers and other fellow vendors, the practice was maintained as a way to safeguard trustbased relationships based on debt and compensation. With the previously discussed social value of food and the close links between gift exchange and economic practice in mind, investigating loaning and other contractual practices in a transnational yet small-scale economic and social system helps to better understand the relationships that are generated and maintained through experiences of food exchange, when physical separation occurs. Following the gift and commodity debate, where both are seen as entangled in the biography of things, credit and debt have been addressed in anthropological literature as equally constitutive of exchange relationships, rather than mutually exclusive (Peebles 2010). Evers (1994) devised the notion of the ‘traders’ dilemma’, whereby traders have a moral obligation to share proceeds with kinsfolk and neighbours on the one hand, and the necessity to make profit, on the other. Likewise, Gudeman’s (2001) economic anthropology model identifies two realms of the economy that, despite being commonly separated in contemporary discourse, are, in fact, entangled in institutions and practices. The community realm designates on-the-ground social relationships and contextually defined values. Market, the second realm, consists of impersonal, global and anonymous short-term
114
Food Connections
exchanges. For Gudeman, as for most anthropologists nowadays, both have to be explored in order to better understand how local economies and local systems of value and exchange are incorporated in new ways into wider global market networks (Stone, Haugerud and Little 2000: 9), since even exchanges of the more abstract kind are only relatively independent of local cultural and social processes. The role of the socioeconomic and cultural aspects of trade has also been debated in the literature on migrant and ethnic businesses, which has centred on the interactive relationship between migrant traders’ involvement in co-ethnic and kinship networks – entailing a complex array of solidarity and obligation elements – and in the socioeconomic and politicoinstitutional environment of the host country. Kloosterman, van der Leun and Rath (1999) have named this dual framing ‘mixed embeddedness’, following Polanyi’s (1944) famous concept of ‘embeddedness’, which expresses the idea that human economy has always been embedded in society and that a self-regulating market economy could never be possible.1 More recently, Çağlar and Glick Schiller (2018) proposed the concept of ‘emplacement’ and explored the ways in which both migrant and non-migrant businesses live within multiscalar relationships of power. This approach helps to avoid methodological nationalism and prevents differentiation and categorization of migrant businesspeople as organized around cultural background and ethnic networks. However, it is important to note, as Guinean migrant food traders do, that the organization of their trade reproduces, in this case, some of the regulations and arrangements of exchange at home, given the intimate way in which the material exchanged – food – is connected with forms of social organization in Guinea-Bissau. It is also worth recognizing, as Çağlar and Glick Schiller (2018) also do, that a sense of disruption and disorientation, as well as experiences of discrimination in the job market, influence business choices. In the case of Guinea-Bissau, transnational food businesses were a response to a lack of other job possibilities (chapter 5), as well as an opportunity created by an increase in migration and consequent demand, which, in turn, results from migrants’ physical and spiritual wellbeing needs that Guinean food, with its familiar qualities and associated capacity to connect lifeworlds, helps to satisfy. As in food giving, relationships involved in Guinean food trade are regulated by local arrangements or contracts that help to cope with the unpredictability of daily life and to uphold community obligations. The term pati (which can be translated both as ‘to give’ or ‘to lend’) was often heard when transactions were performed in spaces like Rossio and Damaia, in Lisbon, and Caracol, in Bissau, as well as in several other arenas of food exchange. Many clients accumulated debts to the sellers with whom a relationship of exclusivity had been established, and sanctions were applied from both sides if those relationships failed to be kept. Sellers, however, only rarely demanded repayment.
Transnational Exchange of Food
115
Although loans were a constant subject of conversations and expressions of complaint, Guinean traders treated their own lack of demand for repayment as necessarily incorporated in their economic practice. In an attempt to explain the system of credit and debts, Dála once told me, ‘I don’t know. I have always acted this way, ever since I was a bideira in Bissau. I give, and then I feel ashamed to go after the customers and ask for the money back. Most of them eventually pay me back, but it can take a long time’. Entering the conversation, Boar – a baobab fruit ice-cream seller in Rossio – brought up the notion of solidarity, suggesting that the then recent economic crisis had reinforced the need for mutual help. However, not long after emphasizing solidarity, she good-humouredly said to a client who was asking for a loan, ‘n’ka tene sorbete di pati, so sorbete di kumpra!’ (I don’t have icecream to lend, only ice-cream to buy), and refused to serve him. Despite common protests and, occasionally, refusals to offer food on credit, the majority of transactions involved it as a strategy to maintain trust-based relationships between seller and customer, just like gifts of food were a means to keep relationships of reciprocity open. If credit was refused, the client could easily break the relationship and economically harm the trader, revealing that credit and debt both contribute to group solidarity in an entwined manner, and should not be seen separately as credit representing power and credit weakness (Peebles 2010: 226). As in the gifting process, the system of credit was more than a simple economic strategy, and the fact that Guinean creditors ‘could never say no’ was also a way to secure future compensation in case of need, since the return could come in the shape of a favour. For women in particular, selling on credit may also be, as in other West African countries, a means of increasing power and social outreach (Guérin 2006), putting them in control of the exchange and of the way in which debts are settled. Alongside selling on credit, other gendered customary systems of economic solidarity are continued with migration. Abota, for example, is the Guinean Creole word for the rotating savings system, mostly managed by women, in which each person gives a fixed amount of money every day, week or fortnight to the person who acts as manager of the group, and the distribution of the total saved amount rotates between each member. Although this is a longstanding practice in Guinea-Bissau, women food sellers in Lisbon increasingly engaged in the abota system after they felt the impact of the financial crisis that had hit migrants with particular intensity. Sali, who managed one abota group in Lisbon, collected the members’ share every day when she came down from the restaurant to sell her yams. ‘My yams bring us all together, as does the money we contribute to our group’, she explained. Notwithstanding necessary changes resulting from the migratory experience, relationships created through economic practice followed similar patterns of locally regulated contracts in Bissau, as they did in Lisbon. Sali described these practices as embodied in her way of life and maintained as a way of orienting her body in an otherwise disorientating lifeworld. Her
116
Food Connections
family name indicated a long past of succeeding generations of traders, which they ‘will never stop being’. She recalled an expression in Guinean Creole that she considered an accurate description of the feeling that, despite her migration, ‘[her] body belongs to the land’ and her food-related activities are embodied in who she is: ‘po, tudu tarda ki tarda na mar, i ka ta bida lagartu’ (no matter how long a stick remains in the sea, it will never become a lizard/ crocodile). In Bissau, women farmers had their fixed clients – the bideiras who, in Caracol, also had arrangements with the buyers that would send the produce to Lisbon. Sometimes the exchange was performed directly between farmer and the migrant trader’s kin; at other times through the use of more levels of intermediary action. Farmers, bideiras and clients occasionally quarrelled about breaks in their agreements, which could lead to changes in the people involved in the relationship. When I met Mila, for example, she had an agreement with Fanta and Safiatu, to whom she directly sold the crops that would then be sent to Lisbon. However, new contracts were established at the end of each rainy season, in October. ‘It depends on who asks me first’, she explained. The creation of these relationships had started, for Mila, years before, when the first traders began to send food to Lisbon. Samba, the first to inaugurate this kind of transnational trade with his brother Aliu in Lisbon, had been her client in Bissau. ‘I saw him and, after him, a growing number of people buying large quantities of some foodstuffs on specific days. Once Samba approached me and asked if I could bring him the produce once a week. He explained that he was sending it to Lisbon. We started working together, it went well, and then others asked me too’. Although not always directly correlated, ethnic and kinship relations influence the practices of selling on credit and establishing exclusivity contracts in both countries. In Lisbon, Guinean migrant traders addressed Guinean customers with the same ethnic origin as ‘relatives’ or ‘family’,2 and would more willingly sell on credit to them even if they had not met before. Such encounters would trigger memories of regional homelands and dishes, hence contributing not only to new relationships based on the exchange but also to bridge the distance between production and consumption across borders. Likewise, in Caracol, commercial bonds were often influenced by ethnic relations. Once, a young man who had only recently started buying foodstuffs to send to a relative in Portugal, and was therefore not yet familiar with the business, addressed Pona and asked for her help in choosing the produce. Pona, one of the Pepel bideiras selling in the market, held the piece of paper he had given her, where the products ordered from Lisbon were listed, and gave him advice on how to select the fruits and vegetables with the best chance of surviving the travel across borders. When he addressed her in Pepel, the woman exclaimed in surprise, ‘How did you know I was Pepel?’ The young man retorted, ‘I knew I was buying from a relative, otherwise I would not have asked for your help!’ Portes and Zhou (1992) defined
Transnational Exchange of Food
117
these practices amongst migrants as ‘bounded solidarity’ and ‘enforceable trust’. The first is created through migrants’ cultural distinctiveness, reinforcing their common origin, shared cultural heritage and mutual obligations amongst co-ethnic business owners, workers and customers. The latter reinforces the commonly accepted norms and guarantees the application of sanctions by the community in case of deviance (Zhou 2004). Yet social networks of relatives, co-nationals or co-ethnics do not necessarily result in frameworks of trust, especially when the transnational element is introduced. According to some male Fula traders in Lisbon, for example, wife-to-husband relationships might involve more risks, due to the tension resulting from women’s more limited access to individual economic gains in Bissau, and the fact that they could see the transnational food business as an opportunity for personal benefit. The business relationship between Teresa – who was on a return visit when I met her in Bissau – and her migrant husband offers one such example. Once, in Caracol, as she was telling me that the time had come for her to go back to Lisbon, a bideira sitting nearby shook her head and teased, laughing, ‘you’re not going back until you finish stealing your husband’s money’. Noticing my presence, Teresa felt the need to clarify: ‘She says that I don’t tell my husband how much I really spend here. If I spend 5,000 CFA francs on vegetables, I tell him that I’ve spent 10,000 CFA, as he’s the one sending the money’. Hesitantly, I asked if that was true. Teresa shrugged her shoulders and answered, ambiguously, ‘that’s what she says…’. Despite the concealment related to the use and distribution of money involved in transnational trade relationships, for the men who, in Lisbon, were dependent on their wives’ shipment of food from Bissau to supply their small businesses, the wives’ previous experience of migration was seen as playing an important role in avoiding deceit. An acknowledgement of the economic difficulties in migrants’ lives and consequent survival needs, rather than the image of wealth that Europe conveys to those who have never left Guinea-Bissau, was thought to be a reason for keeping to the agreement on the Bissauan side. As Fodé once said when I visited his shop: 60 per cent to 70 per cent of the families in Bissau depend on their migrant relatives. They think that we have everything, that we lack nothing. So, if I send them money to buy foodstuffs for me to sell here, they won’t understand my need, and they won’t do it. They will use the money for the household. It has to be someone who already knows how difficult life is here. Idrissa, a young man who was running the family’s small Guinean shop in Lisbon during his father’s absence, confirmed the complexity entailed in transnational family obligations and the key role played by gender, but also by age and lineage in the economic exchange. To him, the safest relationship was the one between male children and parents:
118
Food Connections
Deceit and betrayals also happen within families, but not between parents and children. Fathers and sons know that they work for the family, and no one can see their own family starving. But if it is an uncle, then he will have a separate house, and will also want to bring money to his house. If there’s not much money, it gets complicated. When conflicts occur, Guinean food vendors in Lisbon adopt changes in the provisioning process. In some cases, they terminate the relationships that had been previously established with senders in Bissau and start buying the produce from other migrant vendors who, by managing to receive larger quantities, also act as intermediaries. Alfa, a Fula food and mesinhu seller in Damaia, for example, described a series of deceits he had suffered, with an anguished expression. A sense of loss – of money, but also of trust and security – was manifest in his narrative of food trading, an activity that had started after similar dishonesty experienced in the construction sector, where migrant work had not been paid: I left the construction sector because they didn’t pay, and then the same happened with the food business… The first time that I went back to Guinea I showed my sister what she needed to do to send the products. But once I sent her 2000 Euros and she ‘ate’ it all, saying that she had lost it. The same thing happened again with 1000 Euros. I cut ties with her, and asked a friend to do it instead, but I also stopped trusting him because the produce he sent wasn’t fresh. He was buying cheap so that he could keep more money to himself. So I stopped it all, and started to buy from others here. It is not as profitable, but less risky. As Sali, who now made use of this system to stock her restaurant, put it, ‘I’ve had bad experiences. If the person there takes my money, then, what will I do?’
Introducing New Intermediaries in the Transnational Food Trade Amongst the most recent entrants in the Guinean transnational food trade are Chinese migrants. In Lisbon, part of the long-established Chinese entrepreneurial activity has recently been redirected towards an African clientele, with several Chinese-run shops being named ‘African Shops’, employing Guinean shopkeepers and offering the same food products that are sent by Guineans from Bissau. Some Guinean migrant traders have therefore become intermediaries in these transactions. Moreover, the complex web of intermediary levels with the involvement of Chinese migrants has created situations in which the produce is sold in bulk to them by Guineans who
Transnational Exchange of Food
119
receive it from Bissau, and bought by other Guinean vendors from Chinese shops in Lisbon before reaching the final client. Although most of the time they depend on kinship relations maintained between Portugal and Guinea-Bissau, Guinean food traders hence also extend such networks to wider economic relations (Abranches 2013b), illustrating the complex transnational networks that link societies and economies. Some Guinean migrants claimed to prefer negotiating with Chinese shop owners in order to avoid risking the kind of conflicts resulting from the community obligations described above. This demonstrates how local systems of credit are not just part of people’s inherent cooperativeness with each other (Binsbergen 1999; Evers 1994; Guérin 2006), but they also create tensions and perpetual debts, leading some traders to prefer ‘impersonal’ to ‘community’ forms of exchange. Guineans’ narratives about transactions with the Chinese in Lisbon often referred to the advantages of certainty of payment. If carried out between Guineans, the social obligation of facilitating community-based systems of credit and debt inhibited the assurance of an immediate payment for traders. Moreover, some saw the Chinese success in the business as resulting from stronger family support networks. Mário, a food trader in Lisbon who decided to sell exclusively to Chinese shops, once argued that ‘Guineans, on the contrary, cheat each other. If they have the money, they say they don’t, and they don’t pay’. Chinese traders in Lisbon, like Guinean food sellers, draw mostly on ethnic resources, which, according to Min and Bozorgmehr (2000), tend to generate patterns characterized by smaller businesses, serving mainly co-ethnics and low-income minority customers. However, they also rely on what Wong (1989) defined as ‘entrepreneurial familism’ – the style of economic organization that Guinean migrants praise for its emphasis on ingroup support, whereby the family is understood as a human and economic resource that helps them to achieve economic autonomy in host societies. This form of organization therefore allows them to create larger businesses that offer more varied products and serve a larger population of customers. Guinean consumers occasionally preferred to buy certain products in Chinese shops, whose selling system was considered more beneficial in the period of economic uncertainty they were experiencing. The main difference lay in the fact that Guinean food vendors sold small, pre-arranged packages of foodstuffs with fixed prices (a combination of a handful of kandja, djagatu and badjiki, for example, cost 2.50 Euros). These prices had been initially established by the first traders and thus far maintained by all, with no variations, as part of an agreed regulation of exchange mechanisms. Chinese sellers had, instead, created a more open system of exchange by selling the produce by weight, leaving the client with the choice of buying smaller quantities for a cheaper price, more similar to how it operates in Bissau. The small, pre-arranged packages of food were also seen as embodying an important aesthetic element of the exchange, despite the obstacles faced by
120
Food Connections
Figure 4.2. Foodstuffs for sale in Caracol market. Bissau, 2010. © Maria Abranches.
Figure 4.3. Small, pre-arranged packages of foodstuffs for sale in Damaia, Portugal, 2010. © Maria Abranches.
Transnational Exchange of Food
121
Figure 4.4. Filling and weighing containers with kaldu di mankara. Damaia, Portugal, 2010. © Maria Abranches.
122
Food Connections
Guinean migrants in reproducing open practices of product display in places like Rossio due to police control (chapter 5). When open display is possible, as is the case in Damaia, these packages are arranged in what vendors consider an adaptation in the practice of displaying small groups of similar vegetables in the markets of Bissau (figures 4.2 and 4.3). They are also part of the shared rules of the market, whereby similar arrangements between all traders allow for fair competition. Admir and others who, like him, own small shops, also express satisfaction with the aesthetic arrangements of their space and the products in it, which they consider superior to Chinese-run spaces, despite the precarious infrastructure of the improvised facilities where most of their shops were established. Smiling, Admir often indicated the shelves of carefully displayed products to his clients, as he told the story of their origin or journey. On one occasion, as he was opening the food packages sent by his wife from Bissau, which he had collected from the airport in the morning, he noticed the pre-prepared kaldu di mankara sent in a large plastic bag. ‘This must have been a last-minute thing’, he complained, perplexed, and added that ‘she used to send it in plastic containers’. He quickly visited the opposite homeware shop to buy a soup ladle and, using it, spooned the content of the bag into six clean containers weighing one kilo each when filled. He then stuck a label on each one and arranged them with care on the shelves, where they would be for sale at 7.50 Euros (figures 4.4 and 4.5). The labels, which had been made by a friend with the colours of the Guinean flag, showed the name of the product in Portuguese in order to attract a wider clientele.
Figure 4.5. Labelled 1kg containers of kaldu di mankara for sale. Damaia, Portugal, 2010. © Maria Abranches.
Transnational Exchange of Food
123
While Admir was concentrating on this activity, two elderly women arrived and asked to buy foli. Admir passed them a couple of chairs, put some foli in a bag and gave them an extra one for sharing, while continuing his task of filling in the containers with kaldu di mankara and praising the qualities of the sauce prepared by his wife in Bissau. The women, however, were more concerned with the medicinal benefits of the orange fruit in their hands (figure 4.6). They were old and needed it to remain healthy, they explained, while sucking its small seeds mixed with sugar, then passing it to me. During the rest of the afternoon, other clients came in and performed various exchanges with Admir, sometimes joining in the foli eating ritual, which some preferred to mix with gustu (meaning ‘taste’, it refers to maggi seasoning cubes) instead of sugar, borrowed from one of Admir’s shelves, in order to strengthen its sour taste. Due to the possibility of combining performances of exchange, consumption and sociality enabled by such spaces, as well as to the contractual agreements of exclusivity created amongst Guineans, the involvement of Chinese migrants in the Guinean food trade was not perceived as threatening by Guinean sellers. As Admir explained: Now the Chinese are selling Guinean food too, but that doesn’t bother me, because my clients already know me and they will come to me first. There are a lot more people selling now, but each of us has their own fixed clients. Only if I don’t have the product they’re looking for do they go to look somewhere else.
Figure 4.6. Crate of mango and foli (Guinea gum vine). Damaia, Portugal, 2010. © Maria Abranches.
124
Food Connections
The introduction of further and more impersonal intermediary roles between production and consumption in the case of trade across borders of course increases the knowledge gap between both (Appadurai 1986), yet it does not seem to change the social value of the food exchanged. As I discussed in chapter 2, food preparation techniques and commensality practices are able to compensate for the uncertainty of the food’s origin. Moreover, a shared knowledge of the process of circulation and exchange and its multiple levels, added to the economic benefits of buying at a cheaper price, also help to overcome the increased distance and minimize the disruption. This analysis elucidates what happens when local economies travel to a different socioeconomic space, where other economic actors and levels of intermediary trade are introduced with migration. It fills an important gap in studies of African economic anthropology and migration, which have been concerned either with the rise of the market economy in Africa and local economies’ ways of reshaping their organization, or with what happens at the destination side of the exchange, but which have rarely examined the process from both ends. The focus on food and the relationships it generates based on intimate experiences of credit, debt, risk, trust and morality in a West African economic exchange system, observed from both ends of the migration, reveals that distribution and consumption are not necessarily alienated from production. Even when the chain is physically separated by migration, it is also linked by unique relationships between people, food and the shared substance that connects them to the land.
The Fula Trader The fact that the Fula occupy most Guinean food selling spaces in Lisbon demands that one pay particular attention to their role in processes of exchange. Despite the Pepel and Mancanha origin of most Guinean fresh produce that arrives in Lisbon, Pepel and Mancanha migrant food traders encounter competition and fewer business opportunities in the migration context. One such example is that of Boar and Djamila. Having started by setting up a shop of Guinean products in a shopping centre not far from Rossio, the two Mancanha sisters soon had to close it down due to what they considered growing competition by the Fula. ‘And you know, a Muslim has to buy from a Muslim’, Djamila explained, referring to what she perceived as a higher number of Fula migrants amongst Guineans in Lisbon. ‘Now we’re fine, because what we receive from Bissau, we now sell to the Muslim traders, who then sell to their Muslim clients’, she continued, emphasizing their present role as intermediaries. In addition to that role, they had in the meantime engaged in baobab fruit ice-cream selling in Rossio, ‘because this is something that no one else sells. It became our speciality’.
Transnational Exchange of Food
125
There was not only the perception, shared by the two sisters and others, of a more recent arrival of Fula migrants, sometimes confounded with the view that the Fula were able to travel more frequently, but also a longer history of migration amongst the Fula. That history allowed some of them, like Aliu – one of the first Guinean food sellers in Lisbon – to move from street vending to slightly larger transnational businesses as ajensiador or shop owners. Yet Fula migrants, mainly men, described food selling, also in shops, as kusa di koitadesa (a poor person’s thing), associated with a small income and demanding physical and financial efforts. They saw it as a survival strategy that contributed to the household domestic economy, more than a source of social status as in the case of larger-scale businesses (chapter 5). Some also mentioned trade as the continuation of a skill acquired in Bissau, or before, passed on by their ancestors in the villages of origin or after migration to the city. Linked to this tradition was an occasional association with religion. In the view of Abdulai, an agency owner in Bissau, ‘it is in our origins. Mohamed, the prophet, was a trader, and this is why he asked Muslims to follow his steps, and they succeeded’. However, unlike the Senegalese Murid, for example, Guinean Fula food traders did not mobilize religious networks to expand their businesses. The use of religion and its resources to gain authority and social status, as well as new contacts in business (Bava 2003; Buggenhagen 2011; Evers 1994), was more often applied in the trade of other materials, such as the bazin garments that Adama and her friend sold across borders, as seen in the previous chapter. Clothes, shoes, spirits and wine, electronic equipment, building material and jewellery were some of the most regularly transported products from Lisbon to Bissau by agencies like Neusa’s, charged at ten Euros per kilo by Jorge in Lisbon. Although most of the time they were sent between families as gifts for their own use, when sent in larger quantities they indicated a type of business that generated higher income, providing traders with an image of success and higher social status. Similar activities that embody the image of a successful business and economic achievement have been found in other West African contexts, such as amongst Congolese traders between the Congo and Paris (MacGaffey and Bazenguissa-Ganga 2000). For Guinean Fula traders, however, these activities were more difficult to put into practice due to the higher initial investment needed. A Fula foam mattress seller that I met in Bissau offered an interesting view of Guineans’ participation in trading networks. As he explained, ‘everyone these days calls themselves a trader. In Guinea-Conakry there is an organized network of traders who travel frequently to China and Dubai, just like there is in other West African countries. Only Guinea-Bissau is still doing trade at a small scale’. This narrative suggests that class might also influence the scale of the business, as Min and Bozorgmehr (2000) observed amongst Korean and Iranian businesses in Los Angeles, since the largely impoverished population of Guinea-Bissau is seen, in this discourse, as unable to mobilize resources.3 Yet a small network
126
Food Connections
of importers and exporters of different products from Europe or Asia to Guinea-Bissau was also observable in Bissau. These, mostly Fula men, had usually been long-term migrants in Portugal in the past, or had Portuguese partners in the business, which again seems to suggest that, following Min and Bozorgmehr (2000), their larger businesses might in fact mobilize class more than ethnic resources. *** Ansumane, whose pride in feeding the children of his neighbourhood in Bissau was narrated in the previous chapter, self-identified as a successful transnational trader. As a Fula migrant who travelled regularly between Guinea-Bissau and Portugal, he was able to simultaneously sell food in Lisbon and other materials in Bissau. It was the materials sent from Lisbon to Bissau, however, that he considered the source of his status, rather than the opposite. When I visited his new house in Bissau, he proudly showed me the stand that he was building in the front yard, where he planned to sell the clothes and shoes brought from Lisbon in a large ship container. ‘I’m a lucky man. I don’t lack anything here’, he said, smiling. ‘I am building a big house, and we’ll have electricity and water soon’. After a pause, he continued, ‘I have everything I need in Lisbon too’. In Lisbon, where he had migrated in 1990 and now spent half of the year, he divided his activities between Guinean food and mesinhu selling in the street in Damaia and a small shop of other African-related products – mainly Malian music and films – he owned nearby. Later that day, as we sat on the porch of his new house in Bissau with some of his neighbours, I inadvertently threatened his performance of success when I asked how his business in Damaia was going, now that he was not there. He turned to me with an agitated look and called, ‘Disa kila! Disa kila!’ (Leave it! Leave it!). ‘That’s a secret’, he then continued, to my relief, with a calmer voice, whispering ‘my family here doesn’t know that I have businesses there’. Like a few others, Ansumane preferred to obtain the products he sold in Lisbon through intermediaries, rather than involving his relatives at home. The secrecy implied in Ansumane’s food selling in Lisbon was, I first thought, due to the lack of status it embodies when compared to the trade of more expensive, profitable products in the reverse direction. When we met later in Lisbon, however, he gave me a different explanation for his reaction of a few months earlier. ‘I need to make my family in Bissau believe that I am only living on benefits here, so that I can avoid more pidi, pidi [ask, ask] from them’, he then clarified. Revealing the complexity of such performances, Ansumane thus justified the need to conceal part of his achievements with the risks resulting from the home-based kin’s imagination of migrants’ wealth and the ensuing demands. Yet this incident also offers the possibility to reflect on the value of food. Despite the highly valuable social and material properties of Guinean food
Transnational Exchange of Food
127
for Guineans, resulting from the land’s shared substance and the unique relationships it entails through processes of exchange, food products and their trade are less significant economically. Indeed, although food is subject to the society’s sense of moral justice (Bryceson 1993) and therefore viewed as above market principles (Flynn 2005) where sharing obligations are in place, its trade is less likely to generate much profit, hence limiting the scope for remittances. Food selling was, for Ansumane, as for many Guinean food vendors in Lisbon, a livelihood strategy developed from the opportunity that an increased demand created, though one that also involved risks, both financial and relational. The next chapter will analyse such strategies in detail.
Notes 1. Others after Polanyi have borrowed the concept of ‘embeddedness’ when analysing the relationship between economic transactions and existing social relations (see, for example, Granovetter 1985). 2. See the discussion of ethnic and religious belonging in chapter 1. 3. Trying to categorize Guinean migrants according to a system of class differentiation is a difficult task. First, apart from the separation between the elite and the majority of the population, the weak capitalist organization of production in both rural and urban Guinea-Bissau does not leave room for a significant ‘classic’ bourgeoisie. Second, relatively independent smallholders comprise the majority of the rural population (Abreu 2011; Temudo 2008). Third, the way in which different ethnic groups are organized in more or less hierarchical or decentralized societies adds to the complexity involved in attempts to categorize systems of class.
CHAPTER 5 FOOD LIVELIHOODS AND NEW ECONOMIC SPACES A Critique of ‘Informality’
At Lisbon airport, Aminata described one of the passengers she was waiting for and asked me to stay alert. The description, which her mother had passed on to her over the phone from Bissau, was only brief. ‘He has a short-sleeve pullover on top of a white shirt’ was all that she knew about the carrier of her bag of Guinean foodstuffs – a passenger on the Bissau-Lisbon flight. The assurance of not missing the right carrier relied on an organized network of actors at the departure and arrival ends of the movement, most of whom received a small income for facilitating it. That morning, however, passengers from Bissau were taking longer than usual to arrive, although the flight had been announced for some time on the arrivals board as having landed. Aminata and the others on the receiving end started to express concern and apprehensively called their relatives in Bissau, searching for information that could indicate the reason for the delay. A couple of hours later, one of them announced that ‘none of [our] luggage boarded this time. Everything stayed in Bissau’. Some, sceptical, argued that the problem ought to have been related to the carriers failing to pick up their bags from the conveyer belt, as had occasionally been the case before. However, when the first passenger from Bissau finally came through the arrivals entrance, the initial suggestion was confirmed – the Portuguese airline TAP (Transportes Aéreos Portugueses) had sent a smaller aircraft that night, and therefore limited the amount of luggage allowed per passenger. Most of them were thus still queueing at the complaints desk inside, to find out how to get hold of their remaining belongings. Most likely, the luggage that had not been able to travel would be sent on the following flight within a couple of days, which was of no use to those expecting fresh vegetables and fruits intended for trade. Some blamed the Portuguese airline, assuming a shared understanding that the air company was aware
130
Food Connections
of and generally tolerated their activities. ‘Who will be responsible for our losses now? TAP does this only to upset us’, one of them said. While they agitatedly commented on their lost investments, more passengers eventually arrived with complaint forms that proved what had happened. Those who were waiting for their parcels of Guinean produce hurried towards the ones who had just arrived, trying to find out if their carriers were still inside. A number of them were lucky enough to receive a few small packages. Others called back their relatives in Bissau and asked them to return to the airport and try to recover at least a few of the unsent bags, whose fresh produce could still be sold in Bissau before going bad. Luggage limitation imposed by the airline was one of the risks faced by Guinean food traders. Another risk was that carriers would fail to collect their bags upon arrival, usually for fear of being caught by customs authorities. Sometimes, additionally, parcels were caught and held before departure, at the airport of Bissau, during a process known as luggage identification – a feared ritual performed between airport staff and passengers – which consisted of placing all the luggage on the floor of the area reserved for travellers, hence beyond the view of senders, and asking passengers to identify their own. For fear or suspicion of what they might contain, passengers often opted not to identify the bags they had been asked to carry by others, which generated particular investment loss for traders. Some senders and receivers in Bissau and in Lisbon also described not always having full knowledge of the reasons for the non-arrival of their parcels. Some therefore appealed to the spirits of the ancestors – the Iran – to minimize undetermined risks. As a women sender once explained: In the past, if I sent five parcels, only two would arrive in Lisbon. Three would disappear. I don’t know if they were stolen here, or upon arrival at the airport of Lisbon. So, I knew another fellow sender who was from Bijimita, a sacred town outside Bissau. They say it is sacred because if someone does something bad to you, you can go there and ask the Iran to punish that person. After seeing me repeatedly so upset, my friend told me: ‘I am from Bijimita. Next time someone steals your packages, I will take you there, and you can put djanfa on them. The Iran will know who they are’. I said ‘OK’. Then I had the idea of writing Bijimita on my bags to make people afraid of stealing. Since I started doing that, they have hardly been stolen. These activities and their risks – which are not only financial but also emotional, involving severe distress and weariness – are indicative of the complex system of relationships involved in what are usually labelled ‘informal economies’. This chapter offers an in-depth analysis of the relationship between spaces, materials and human actors that play a role in Guinean food trade livelihoods at the two ends of the migration. The analysis will draw
Food Livelihoods and New Economic Spaces
131
on a critical exploration of informality (Banks, Lombard and Mitlin 2020), with the stories described contributing to an understanding of not only how boundaries between formality and informality are permeable and intertwined, but also how they are transnational rather than localist (LourençoLindell 2004, Meagher 2013). The phenomenological-oriented perspective adopted in this book will allow these abstractions to be understood through people’s grounded everyday livelihood experiences in Guinea-Bissau and in Portugal and their role in the making of a Guinean transnational lifeworld.
Transnational Food Trade: A Livelihood Strategy Informality in contemporary Africa has tended to be examined either as an occasion for dynamic, vibrant and creative livelihood practices, or as an agent of poverty, collapsing economies, ethnic conflict and violent civil wars.1 In the last decades, however, there has been a growing concern with formal-informal linkages as well as divisions, and with the complexity of such relationships. The lack of jobs and declining real wages in urban Africa, alongside externally induced economic liberalization policies and a crisis of the global economy, has expanded the geographical scope of those relations, altered their forms and brought in new actors (Lourenço-Lindell 2004). Yet rapid informalization is not a uniquely African phenomenon. In Portugal, the growing labour demand in construction and public works between the late 1980s and the early 1990s favoured the informalization of the labour market in this sector and the turning of a blind-eye to the recruitment of irregular migrants, especially from Portuguese-speaking African countries (Peixoto 2002). In Lisbon, most Guinean men who arrived during this period and are now food traders and agency owners had their first work experience in the construction sector. Likewise, many long-term migrant women who now sell food started in the domestic sector. Given the gendered pattern of the first migration flows, men – who mainly composed those flows – have been able to accumulate more resources since their early migration experience and have therefore moved, more than women, from street selling to small shop ownership or to the role of ajensiador. From the late 1990s, new migration flows from Eastern Europe also targeted the informal market in Portugal. Paradoxically, this era of economic flexibility, which saw illegal paid work being tolerated at the very least, was accompanied by a political discourse promoting migration control (Peixoto 2002). The economic and financial crisis of 2007–2008, which was starting to have severe consequences at the time of this research, triggered further state reforms based on the idea of ‘flexibility’. These resulted in an aggravation of job insecurity for some groups (with migrants amongst the most affected), namely through longer working hours, less overtime pay, increase of shortterm contracts and flexibilization of dismissal rules (Esteves et al. 2017),
132
Food Connections
and in pushing migrants to other self-regulated activities such as the transnational trade of homeland food. Ironically, while associating such activities with marginal groups and seeing them as ‘predatory’, ‘parasitic’, ‘wasteful’ or ‘dubious’ for challenging the norms of an industrialized, capitalist economy, governmental policy simultaneously threatened more stable, salaried work through incentives to informality (Lopes 2006: 333). There is as of yet limited in-depth understanding of how being pushed to informal spheres of the economy is experienced and negotiated by migrants and their families. Research on Guinean migration and labour market integration has been scarce since Machado’s (2002) study of the 1990s,2 and little is known about Guineans’ specific strategies for coping with increasing economic uncertainty, especially after the economic crisis that started in 2007. This ethnography reveals that homeland food trade, as other informal selling activities, occupies a critical position in Guinean migrants’ diversified livelihood strategies in this context. Moreover, this chapter also covers the more general oversight from research on transnationalism and migration, migrant entrepreneurship and economic anthropology, on the everyday experience of these economic activities. In contrast to the more formal migrant business structures, which, run predominantly by Asian migrants, tend to receive more attention in the literature,3 this oversight might be related to the relatively small scale of homeland food selling activities amongst African migrants, their often secondary role, and their limitation in terms of economic contribution to household incomes. However, as I demonstrate here, these features do not downplay the importance of such businesses, both economically and socially, for those involved in them from both ends of the migration. For Guinean migrants, as for other African migrants, these activities draw partly from the experience of being excluded from opportunities in the regular labour market, which, in turn, follows previous exclusion in the country of origin, where opportunities for a valuable and stable occupation are significantly lacking. This double exclusion prompts them to create a new order through the construction of reliable social relations (MacGaffey and Bazenguissa-Ganga 2000). However, for those for whom the new created order is set around trade, the enduring sense of exclusion influences their attitude towards commerce. As MacGaffey and Bazenguissa-Ganga (2000) have also found in the case of transnational traders between the Congo and Paris, food trade was seen by most Guinean migrants as a secondary or temporary activity, from where they would move out as soon as an opportunity in the formal job market arose. For them, unlike for those involved at the Bissauan end of the chain, food trade was often not perceived as ‘work’. Admir, for example, whose story will be described below, claimed to have started this activity only through the insistence of his wife in Guinea-Bissau. For women in Guinea-Bissau – responsible for guaranteeing the household’s basic needs – engaging in food trading activities was, before the introduction of its transnational component, one of the few options that allowed
Food Livelihoods and New Economic Spaces
133
them to ensure those needs were satisfied. A household survey administered by Lourenço-Lindell (2004) in two neighbourhoods of Bissau in the 1990s revealed that trade was the main source of income and that retail trade of food items, including prepared food, drinks and charcoal was the most common trade activity (ibid.: 93). Even amongst civil servants – a socially valued profession since colonial times for many Africans (cf. MacGaffey and Bazenguissa-Ganga 2000) – a general problem of delayed salary payments in the country has led most women to develop a second activity for economic gain, often materialized in selling homemade meals and ice-cream or buying clothes in Senegal to resell in Bissau.4 Contrary to neo-liberal assumptions, however, such activities are a strategy for survival more than a way of providing a sufficient income for the working poor (Lourenço-Lindell 2004). Yet the transnational element of these economic activities – an opportunity generated by migration – seems to have partly changed the perception and experience of food trade for women at home. Indeed, except for the younger generations in Bissau, for whom participating in the transnational food trade might be an undesired family obligation (chapter 3), expanding their local food trading activities to a transnational level is often seen by women in the home country as an opportunity for accumulating some individual savings. This economic advantage for women at home is, as seen in the previous chapter, one of the reasons that make migrant male traders fear the risk of disloyalty in profit distribution in wife-to-husband business relationships. The diversification of livelihood strategies, widespread in Guinea-Bissau and amongst migrants, is a key coping strategy in the face of low and uncertain incomes on both sides of the migration. For many, involvement in trade started at a young age in rural Guinea-Bissau, as a way to contribute to the household with an income. Alfa, for example, used to go from Mansoa to Bissau – a distance of about 60 kilometres – to sell groundnuts and kola nuts in the city, at the age of twelve. Admir, who now owned a food shop near Alfa’s mesa, had a similar early experience in small-scale trade in Guinea-Bissau: My brothers still have cows in the village near Bafatá where I lived until I was eight years old, but I started selling in Bandim market in Bissau. I started by helping an older trader, by selling plastic bags, carrying his things from the storehouse… I was twenty years old then, and still a student. Then the old man started giving me some small things to sell, and I started little by little. Yet Admir, now unemployed from the construction sector – from where the largest part of his income normally came – dissociated his current selling activities from the notions of work and having a job. As he explained, ‘this was my wife’s idea, not mine, because I was here to work. I am a carpenter. I learned the skills here, and that’s my expertise. Trade is, for me, a
134
Food Connections
supplementary occupation. When I’m working, trade is only secondary’. Starting up a larger-scale trade activity, on the contrary, was a desired project, but it required exclusive dedication and considerable financial capital – which he did not have – for the initial investment. As he put it, ‘what pays for my meals at home is my job. I earn more in construction, but it’s not a regular job. Sometimes there’s work, sometimes there’s not. If I had the means, I’d dedicate myself only to large-scale trade’. His wife Lola, who had joined him in Portugal in the meantime, having left her sister with the responsibility of sending the food produce from Bissau, shared her husband’s desire. When we met in Bissau, she had told me, ‘if I find a way to send things in container ships, then I can send larger quantities and sell them at cheaper prices. At the moment, the prices of Guinean foodstuffs are too expensive for people’s purchasing power. You’ve seen it… we’re just selling to survive, not to save’. The impact of the financial crisis in Europe was generally felt by most of those involved in the transnational trade of Guinean food, regardless of the form and scale of their business. Alfa, who migrated to Portugal in 1988, recalled the first two decades of his stay as prosperous times, when ‘there was a lot of money, and many jobs’. His involvement in the food trade business had started during the 1990s and, as with many long-term traders – predominantly men – it had quickly progressed to involve large quantities of produce with which he supplied other Guinean vendors. Yet it had also declined more recently. Sali, too, based on her experience with the restaurant, once complained: In the past, I used to rent this place by myself. Now I had to ask for my sister’s and another friend’s help, so that I didn’t have to close it down. Back then, I used to make 200 Euros or more a day. Now I make 60 or 70 maximum. The Portugal of today is not the same as before. Nevertheless, she still considered the food business to be economically safer than other kinds of trade, justifying it with the idea that ‘si krizi ten o i ka ten, bariga ten ku kume’ (whether there is crisis or not, the stomach has to eat). Unlike Admir, Sali and others who arrived in the 1980s and 1990s, who had found work in the construction and service sectors and were in the meantime able to open small shops or restaurants, the more recently arrived migrants found in the occasional trade of food an alternative source of income during the initial period of migration. Meta, for example, had come to Portugal the year before we met, with a temporary visa to accompany her son who was having treatment in hospital. Her primary aim was to find a stable job with a work contract in order to apply for a stay permit, but she had spent the year selling homemade ondjo juice and konserva in Rossio, later combined with a temporary replacement job in the cleaning sector in a supermarket. Like other recently arrived migrants, Meta was not yet endowed with the necessary social networks and financial capital to have Guinean foodstuffs
Food Livelihoods and New Economic Spaces
135
sent from Bissau. Her involvement in the Guinean food trade thus consisted of buying the ingredients needed to prepare the ondjo juice and konserva sauce from Dála, with whom she had established a relationship of mutual help. With those ingredients, she prepared the juices and sauces at home and took them in small reused plastic bottles and containers to sell in Rossio. The products provided by Dála were occasionally bought by Meta on credit, following the group-regulated financial practices described in the previous chapter. In return, she helped Dála by selling what was left of the elderly woman’s produce later in the afternoon. The fact that Meta lived in Rossio facilitated her late stay and allowed Dála to return earlier to her more distant home, when her daughter or oldest granddaughter were not there to help. Buying the produce from other migrant traders, to resell or to prepare homemade food for selling, instead of having it sent from home, involved fewer risks, but also smaller quantities and a reduced business scale, hence a smaller income. It was a temporary income strategy put forward not only by the recently arrived migrants, but also by those who had experienced a betrayal of trust and seen their cross-border exchange relationships being terminated. Tensions and conflicts in the transnational frameworks of trust that are essential for the success of the food trade were analysed in the previous chapter. They occur at both ends but, given the mono-sited nature of most studies of migration, have tended to be associated more often with Africa, under the umbrella of ‘informality’. Although the size and central role of the informal sphere in African economic landscapes have partly resulted in a tendency to blur rather than clarify formal-informal distinctions and linkages, and to argue for normalization rather than for the need for critical analysis (Meagher 2013), a new focus on the complexity of such processes and definitions has also emerged.
Critical Informal Landscapes: The Airport Space In the context of urban Guinea-Bissau, Lourenço-Lindell (2004) has shown the value of combining different perspectives to better understand the expansion of informality, revealing how actors may wish ‘to have one foot in each sphere’ (ibid.: 85). While instances of economic activities’ ‘disengagement’ from the state may be driven by lack of governmental protection and signify the worsening of working conditions, they are often combined with forms of ‘engagement’ whereby political networking is necessary. Drawing on the view of informality as a site of critical analysis (Banks, Lombard and Mitlin 2020), the economic activities analysed in this book also highlight both the continuum and distinction between both spheres. Rather than being characterized by a lack of regulation, the informal sphere has a well-defined set of rules negotiated and enforced, not by the state alone, but by
136
Food Connections
diverse actors, including elites who operate in the two domains. An example of this is the weekly air cargo system that was set up as a regular economic activity by a small group of Guineans who had previously established networking ties with a business elite in the two countries, working alongside and often in cooperation with the ‘informal’ courier scheme. Allowing for risk reduction in the process of sending fresh produce across borders, it was used mostly by those who had shops or intermediary roles in Lisbon and therefore dealt with a larger quantity of food parcels to be sent and received. Tony led this activity from Bissau. With a partner in Lisbon – Teresa’s husband Aladje – his business worked as an intermediary between the Guinean community and TAP’s official air cargo services. Each Friday afternoon he received his clients’ packages, dealt with agricultural inspection procedures and took the packages to the airport customs area, to be dispatched to Lisbon on that evening’s flight. The storehouse that served as his office was bustling with activity on Fridays. Senders occupied most of the inside and outside space with their bags of fresh vegetables brought by young boys with carretas (trolleys) from Caracol market. The content of the bags was transferred into previously perforated cardboard boxes, weighed and then rearranged in the boxes in an attempt to reduce the final weight and cost, before they were finally sealed with tape and the name of the recipients and their contact number in Lisbon written on the front (figures 5.1 and 5.2). It was common for some of the content to be relocated from one client’s box to another in order to help balance out the weight and cost between them. This involved complex negotiations and several phone calls to Lisbon with explanations of where the produce was to be found upon arrival. These activities, which are at the centre of the transnational exchange of food, embody as much intensive corporeal practice as the previously explored rituals of food preparation and consumption. First, the way in which senders select and handle the material is viewed and experienced as intimately linked to particular ways of life (for example, the Fula carelessly force the foodstuffs into a bag, whereas the Mancanha gently transfer it, as seen in chapter 1). Second, their bodies bend, reach and lift in the process of selecting, packing, weighing and transporting the material. Finally, the agitated sounds of their negotiations and phone calls are part of the material surrounding of the transnational exchange that helps to keep the connections between the two lifeworlds. Tony started this activity in the early 2000s, after having worked for a Portuguese importer-exporter of wines and mangos, where he learned the necessary skills and made useful contacts with official dispatchers and TAP air cargo officials in Bissau. Yet the prices charged by what was known amongst Guineans as the ‘cargo system’ – which included agricultural inspection, customs and dispatch – were generally considered expensive by the majority of the traders I spoke to, who also complained of a price increase in a period of more deeply felt economic instability. In addition to the amount
Food Livelihoods and New Economic Spaces
137
Figure 5.1. Preparing food boxes to be sent to Lisbon. Tony’s office, Bissau, 2010. © Maria Abranches.
138
Food Connections
Figure 5.2. Food boxes ready to be sent to the airport. Tony’s office, Bissau, 2010. © Maria Abranches.
Food Livelihoods and New Economic Spaces
139
spent on food products and dispatch, expenses included paying others who, like the flight passengers who acted as carriers, also found a small means of livelihood in these activities. Carreta boys, for example, charged 300 CFA francs to transport the bags of food to Tony’s agency, not far from Caracol food market. Despite its risks, many food senders thus preferred the courier system for its cheaper costs or, for those who sent larger quantities, a combination of both. This meant that Tony and Aladje also experienced the effects of the financial crisis, with the volume of cargo received for dispatch dropping from around 1,000 kilos of parcels each week to only 500 kilos. Food trade related activities performed in the airport space illustrate other instances of the dependency of the informal sphere on the formal (Abranches 2013a). On my first field trip to the airport of Bissau I witnessed one of those occasions, as the practice of banning non-passengers from entering the airport space – an additional risk in such activities – was being imposed. Gathering outside the entrance door, those with packages to send directed pleas at members of staff who were implementing the ban, and speculation started to circulate, associating the interdiction with earlier quarrels between one sender and the airport staff, who were now using their power to retaliate. Alleging that ‘this is not fair’ and pointing out that they were ‘old clients’, Teresa, like others, asked, ‘please close your eyes, my son’. While some managed to approach familiar passengers and ask them to be the carriers of their bags before they crossed the entrance door, Teresa ended up taking a taxi back home with her parcels as, disappointed, she murmured ‘they were only harmless little leaves’. My introduction to the airport of Bissau was therefore a disappointing ‘working evening’ for my research participants, just like the morning described at the beginning of this chapter had been in Lisbon. When Aurélio, who owned a transnational ajensia in Bissau, had invited me to accompany him to the airport, he had suggested that I ‘see how they work’. The tasks of packing, labelling and driving to the airport late in the evening – often with a regular taxi driver who also benefited financially from these activities – and finding a trustworthy carrier, were described by all as ‘work’. On the one hand, the physically demanding nature of the tasks – mainly the late and long hours involved – made the experience laborious. On the other hand, they were part of a transnational economic activity that started in the urban smallholdings and storehouses around Bissau where the produce was harvested or stored, and that made up a large number of families’ livelihoods, both at home and abroad. At the airport of Bissau, regular complaints about a sense of fatigue, anxiety and risk involved in that particular stage of the circulation of food were part of most conversations. Migrants on temporary returns, who regularly helped in the task of sending foodstuffs to Lisbon, claimed to be perplexed by that rituality. Aua, who used to accompany her niece Yasin to
140
Food Connections
the places that materialize the food’s transnational journey – the market, Tony’s office and the airport – during her temporary return, described the airport ritual as the most prolonged and tense. Often, while the challenging task of finding carriers progressed in the hot inside area of the airport, she sat in the café outside with a soda, observing others who stood on the concrete footpath negotiating transactions or occupied improvised seats on the pavement, where the heat was more tolerable than inside. Men and women walked nearby selling cigarettes, groundnuts and canned drinks, and children moved vigorously near the entrance, selling small items like sweets and chewing gum, as well as adhesive tape and plastic bags that were frequently bought by Guinean senders to wrap up last-minute acquisitions to be sent to Portugal (Abranches 2013a). Aua frowned as she commented on this rituality, which differs from the ‘politics of airport function and design’ defined by Adey (2007: 516), whereby technological, architectural and retailing elements and functions are employed to arrange people in particular positions within a space of dwelling and spectatorship. She considered that assemblage of comfort and entertainment – with which she claimed to be more familiar – and the performance of sending and receiving practices in the airport of Lisbon, more pleasurable and acceptable. Acceptance, here, is therefore linked to familiarity, but also to the idea that certain aspects of formality are more desirable, suggesting that a preference for social relations and networks created outside formal institutional regulations can be juxtaposed with the desire to eliminate the adverse conditions of ‘informal’ work. Migrants on temporary returns were strategically positioned in the business, since they were more frequently acquainted with other migrant passengers and thus stood a better chance of successfully negotiating the transport. The majority of senders, however, had never been to Europe. The tasks they carried out at the airport of Bissau were contingent on closely interconnected social networks where each actor has a specific role in the movement of food – and other materials – from one country to the other, and rules have to be defined. Senders would alternate between trying to identify potential carriers themselves and, as Aminata’s parents often did, sitting down attentively, while an intermediary – who could be a member of the airport staff – helped them in that task, in exchange for a compensation, usually monetary. This compensation, in the case of carriers, was not always motivated by individual self-interest. Since most travellers who acted as carriers were paid in West African CFA francs, which were of no use once they arrived in Lisbon a few hours later, that money was usually given to their relatives before boarding, for use in Bissau. Such well-defined rules that link ‘formal’ and ‘informal’ actors and practices could also be sanctioned with performances of control and interdictions (such as banning non-passengers from entering the airport, as the story above illustrated), when an element of the group failed to fulfil part of the agreement. Practices of shipping are therefore complex and materialize a
Food Livelihoods and New Economic Spaces
141
way for all those involved to accumulate advantages – usually in the form of money or as an exchange of favours – and to maintain the system of relations that defines them.
The Market Space in Bissau and in Lisbon The phenomenological-oriented approach adopted in this study, which emphasizes the material experience of producing, consuming and exchanging food, and of the spaces it travels through, can help transform the abstraction of wider processes of transnationalism, globalization and trade into more precise and meaningful conceptual tools. Just as the airport space and the ritual practices of sending and receiving food that it stages introduce critical insights into the study of informal economic activities at a transnational level, the materiality of street markets like Caracol, in Bissau, and Rossio and Damaia, in Lisbon, reveals the way in which local food vendors are active participants in the formation of global economies and spaces (Grassi 2003). The anthropology of markets draws a distinction between physical marketplaces on the one hand, and the market principle on the other. Whereas the former is usually associated with periodic, dynamic spaces, the second is linked to commercial exchanges of a more abstract kind (Applbaum 2005). Yet, as I contend here, a phenomenological approach is useful to understand the entwinement of physical marketplace and market principle in a particular context, helping us to get a much more complete picture of the relationships involved and how they influence the everyday social and economic lives of the people concerned, even when they are geographically separated by migration. Markets as physical spaces are privileged locations to ethnographically study social interactions of different kinds. In Guinea-Bissau, as elsewhere in Africa, they have been fulfilling multiple social and economic functions since pre-colonial times. Contrary to western biased views of African markets as governed by disorganization and chaos as opposed to clearly normalized capitalist economic spaces, the way in which they are regulated follows welldefined criteria (Stoller 1996). These group regulations are then reproduced and adapted to new spaces of exchange with migration, as the previous chapter already revealed. In Bissau, the food market of Caracol, where most of the Guinean fresh food products that arrive in Lisbon originate, has expanded since 1988 from a spontaneous grouping of food sellers (Lourenço-Lindell 2004). According to the women bideiras who daily occupied the space, it was named after a Portuguese businessman who owned commercial establishments in the area. Just like Rossio, where the Guinean food selling area was constructed around the former Disabled War Veterans Association, now Palace of Independence
142
Food Connections
(figure 5.3), the sociohistorical construction of Caracol is hence also associated with the colonial past. A variety of services set around the area fulfil multiple functions for vendors, with restaurants and small bars set up in private houses forming an important part of the surroundings. The section of the market where the fresh vegetables and fruits to send to Portugal were exchanged was characterized by gendered spatial divisions. Whereas men improvised a few stalls with structures of wood and sold a variety of predominantly imported food products, such as canned tomatoes and beans, cooking oil and powder juice drinks, women used cardboard, plastic crates or tins of paint to sit down and arranged the fresh vegetables on the ground for display (figure 1.3). Each bideira occupied a fixed position – called a mesa – the luckiest sitting under one of the two mango trees that provide the only sources of shade in that important food provision space in the city. They paid 150 CFA francs per day to the city council for a mesa, and a further 150 CFA to Gina, the woman who managed the area. The market regulations were highly organized. Caracol, like all other markets in town, has its own association of sellers, whose president is often called in to resolve conflicts between sellers and clients, or lead campaigns to improve the conditions of the space. In Rossio and Damaia – two key areas of Guinean food exchange in Lisbon – the reproduction of spaces such as the Caracol market of Bissau is related not only to the nature of the economic exchange and to the material exchanged, but also to the social and material construction of the space where that exchange occurs (Abranches 2013b). As in Caracol, dwellers in these two spaces at the receiving end of the migration are dynamic performers who help to socially and materially construct them (cf. Sato 2007), including through a variety of nearby services that, like in Bissau, fulfil multiple functions (such as improvised Guinean restaurants, bars and a mosque). Reproducing common practices staged at Caracol, these services are used by Guinean migrant sellers and clients to share news and meals, practise their religion and other rituals, discuss politics, make plans and remember the past, as seen in chapter 3. Despite the absence of female associations based specifically on trade and market activities, women migrant sellers nevertheless organize themselves to resolve conflicts, as their kin do in Bissau. Differences in the spatial environment, however, create important distinctions between Rossio and Damaia, and influence the way in which trade is performed in each. For example, the fact that Damaia is an area of spontaneous squatter constructions makes it more deeply permeated by informality. The advantages are obvious – there is less control on street selling without a license, and some local shops have been established without the need to engage in the bureaucratic and costly official procedures for setting up and maintaining a business. The location of Rossio in Lisbon city centre, however, does not necessarily endow trade activities with more formal mechanisms, instead creating more constraints in overt performances of informal street
Figure 5.3. Rossio, Lisbon, with the Palace of Independence at the back, 2021. © Ana Estevens.
Food Livelihoods and New Economic Spaces 143
144
Food Connections
selling and the need for further financial capital for those wishing to set up a shop in the area. Another difference, seen in chapter 4, is in the practice of produce display with aesthetic concerns – also reproducing food markets in Bissau – which cannot be performed in similar ways in Rossio, given its more covert nature. With migration, gender and product specialization continued to shape the way in which the market space was constructed and used, rather than just abstractedly represented (Moore 1994). Not only were the materials that Guinean women and men sold often differentiated, but the physical positions they occupied could also be distinguished, especially in Rossio. Here, women arranged similar cardboard, plastic crates or tins of paint to sit down and display their products under the shade of a tree in the same area of the square every day, as they did in Caracol. The boxes were folded and left near the wall railing during the night, ready to be reused the next day.5 Men, more commonly, sold tobacco and kola nuts in a differentiated area, where they also performed their activity of ajensiador or sapateiro. Given the specific nature of mesinhu di tera and a perception of increased risk associated with the trade of the tree parts that compose them, they were more often sold in Damaia than in Rossio, or through individual requests to those who stored them at home. The increased risk of police intervention in the case of mesinhu selling in Rossio was, to the Guineans I spoke to, due to the unfamiliarity and therefore suspicion they generated amongst the Portuguese in that city centre location. Moreover, as described in chapter 1, the specialized knowledge required to deal with these materials entailed specificities not only in the practice of shipping them, but also in their commercialization, meaning that they yielded a smaller income when compared to other food-related products. Product specialization is particularly subject to group regulations (Domingues 2000). Dála, for example, used to sell baobab fruit ice-cream in Guinea-Bissau, but was unable to replicate that practice in Lisbon, due to Boar’s and Djamila’s anticipated specialism. As she explained, ‘I’d like to make ice-creams here as well, but Boar is already selling them’. She concluded, ‘I don’t do it so that she doesn’t get upset, and starts talking’, referring to the idea of gossiping that is, too, an important factor in regulating exchange behaviour. Just as Boar was known for her homemade ice-creams in Rossio, Sali was recognized for yam selling in the late afternoon, which provided an important occasion for food sharing and conviviality, also part of the social roles of African markets. She explained, ‘I used to prepare and sell African juices and croquettes, and sell them here in Rossio, and at the mosque. But then everyone started selling the same and I decided to try something new: boiled yams! There’s no one else selling those!’ In conjunction with these different configurations, the relationships created also differ in Rossio and Damaia. In the latter, most clients and
Food Livelihoods and New Economic Spaces
145
passers-by were Guineans or other Africans – Cape Verdeans in particular – who reside in large numbers in the neighbourhood. Regular Cape Verdean customers often spent as much time as Guinean clients with the street food sellers practising djumbai (socialization). One of them, Daniel, who lived in Damaia, confessed to preferring their company to other Cape Verdeans. Aminata teased him, saying ‘it’s because our food is better’. He bought a small bottle of palm oil and a pack of groundnuts – which we all shared – dragged a tin of paint closer to sit near us, and talked in Cape Verdean Creole, which shares similarities with Guinean Creole, about the land in Cape Verde. Aminata claimed that ‘it is too dry and nothing grows there’, comparing it to the fertile Guinean soil. In response, Daniel disapproved, good-humouredly, of the strange taste of some Guinean products. ‘This, for example’, he said, pointing to foli, ‘I can’t have it, not even with sugar. The taste is just too strong’, he concluded, frowning. These spaces therefore staged not just an actual exchange of food but also an exchange of ideas and imaginaries of food which, together, provide evidence of the importance of both experience and meaning in migrants’ lifeworlds. Cape Verdeans were also, as seen in chapter 2, the sellers of the maize and fish that comprise key products of Guinean commensality, hence interacting with Guineans from the position of buyers and sellers. During the foli eating occasion in Admir’s shop, described in the previous chapter, one of the regular fish seller women in Damaia stopped by. She sold Admir some fish, received a pack of cashew nuts in return and, like Daniel, commented with aversion on the sourness of the Guinean fruit that the elderly women were sharing. In Rossio, located in the historical city centre, different interactions resulted from the way in which Guineans and non-Guineans share that space. Being primarily a space of socialization for Guineans, hence narrow in scale as a ‘food market’, it did not follow the same city regulations as the African market explored by Stoller in New York City, for example, where organized tourist circuits bring ‘camera-packing Europeans to “shoot” the African market from a safe distance’ (1996: 779). Yet Guinean foodstuffs were occasionally also part of the attractions of this Lisbon downtown area for tourists, who I sometimes saw discretely pointing a camera as they walked by. Whereas such acts normally caused initial protests and ‘stop that!’ exclamations from Guineans, they were also amused by them once those tourists embarrassingly put their cameras away and left. To the foreign tourists and the Portuguese who approached Guinean food sellers, driven by curiosity, kola nuts seemed to be the produce which generated the greatest interest. Kola nut sellers were usually happy to let the curious passers-by try it, alerting them to its bitter taste. A Portuguese couple, for example, once asked Bacar the name of the products in the small bags that he and Dála, sitting nearby, were selling. Amused faces and laughs were exchanged as the couple grimaced when tasting the bitter fruit, and an expression of satisfaction was
146
Food Connections
visible in Bacar’s and Dála’s faces when the two clients decided to buy one bag of cashew nuts and one of veludo (velvet tamarind). ‘You can eat veludo like this’, Bacar explained, ‘it is very sweet. Or you can make juice with it’. The Guinean homeland was, on such occasions, brought to the knowledge of others, enabling the connection of lifeworlds to be extended in a way that gave Guineans not only a much needed sense of security and control but also a desired sense of recognition, of being agents and initiators of order in an ambivalent world (Jackson 2013: 14). However, such interactions were infrequent. Guineans’ economic activities were performed in a partially covert manner in Rossio, which led to a certain level of ‘invisibility’ for most non-Guineans, and therefore to the absence of a mediatized image of marginality or insecurity in the area. Nevertheless, discrimination also occurred. In spite of the friendliness with which Guineans addressed the curious passers-by and the satisfaction with which they explained the meanings and tastes of the products they sold, I was occasionally told stories about receiving looks of disdain, ‘especially from the elders who live in this area’, as Seco once told me. Then, pointing to a woman who was leaning against the opposite wall, he continued, ‘this old woman always looks at us like that, complaining that we obstruct the pavement’. These suspicious and discriminatory attitudes led to constraints to an open performance of exchange and display of products, and to the need to arrange the products in a way that allowed them to easily pack up if the police arrived, a necessity that produced a constant state of alert – the ‘somatic modes of attention’ (Csordas 1993) also experienced by Cidália and her husband in Monte Abraão (chapter 3). A recent increase in police intervention was, in fact, a common concern amongst Guineans in Rossio, and this targeted not only those ‘informal’ exchange activities, but also undocumented migrants. The more recently arrived migrants like Meta who, while struggling to regularize their situation, counted on selling homeland food as a livelihood strategy, particularly feared these interventions. According to Guinean food vendors in Rossio, however, the increase of state control experienced immediately after the start of the financial crisis of 2007–2008 had rarely led to serious consequences such as conviction or even deportation, although confiscation of goods was reported as happening occasionally. These effects were to be reinforced by the immigration act of 2012, when increased immigration control was imposed through more severe sanctions on undocumented migrants, justified by the need to adjust the national juridical framework to European policies on migration. While there is no evidence to suggest how this change in policy has effectively impacted upon migrant street traders, in 2019 new law revisions took a different direction, making regularization easier for those who had been making national insurance contributions for twelve months, without the need to prove legal entry in the country. The impact of such changes on more recently arrived migrant food traders and their activities is, too, yet to be seen.
Food Livelihoods and New Economic Spaces
147
In those years before these changes in law, the difficulties experienced by migrant street sellers prompted some of them to attempt to negotiate with the city council for more formal ways of running their economic activities, again suggesting that engagement with the state might also be preferred or necessary. Some women sellers aspired to open what they considered a big African market as in other European cities, something more along the lines of the African market studied by Stoller (1996) in New York City. For many, the Barbès market in Paris was the example to follow. It was frequently described as ‘the centre of Africa in Europe’, an economic space of African food variety, cheap prices and, above all, the opportunity to overtly sell and buy products on a larger scale in Europe. Yet schedule-related constraints of municipal spaces of exchange in Lisbon, which are limited to one or two days a week and to restricted opening times, would not allow the market space to fulfil its multiple social and economic functions for Guineans. The idea that ‘this is what we do in Guinea-Bissau, we can’t do it any other way’ was once used by Meta to describe the need for long hours spent in those spaces of exchange that are also spaces of ‘sitting’ and djumbai. Sitting, as Gaibazzi (2015) found in rural Gambia, was for Guineans an everyday experience that, entangled with the exchange of food, reproduced the homeland experience of being on and with the land. The example of the opportunities and risks that formal spaces of exchange pose for Guineans suggests that, as Meagher (2013) has pointed out, interaction with formal institutions is beneficial for some aspects of people’s activities, but not for others, and reinforces the need to understand the complexity of formal-informal linkages rather than their divisions. The hours spent in Rossio and Damaia, two spaces between which some vendors and clients travelled each day, were spent not only sharing news, food and politics, as in Caracol, but also helping newcomers with the process of finding income sources and obtaining visas. These spatial and material practices, reproduced and transformed with migration, help to make people’s lives simultaneously local and integrated into a larger system of relationships. With food placed at their centre, they enable a connection with the land that is essential to make sense of the newly created reality.
Bringing Closer a Distant Land: A Family’s Livelihood Story Wilson was the son of Teresa, introduced above in this chapter as one of the women senders from the airport of Bissau, an activity with which she engaged during a long return visit. The story of Wilson’s and his mother’s livelihood strategies in Lisbon illustrates the way in which temporalities are embodied in migrants’ livelihood experiences, and their link to land. As it will demonstrate, for Guinean food vendors who participate in a cross-border movement of food that needs swiftness in order to succeed, stillness is as much part of the experience as the demanding rapid dynamics of their livelihoods.
148
Food Connections
As seen in chapter 3, mobility has been given a central position in studies of migration, and tended to be associated with a modern landscape, where everything seems to be on the move on a larger scale and with a faster temporality. Authors like Massey (2005), Thrift (1996) and Urry (2000) have explored how experiences of mobility can be sensed through complex material and social practices that are performed in dynamic and relational places – a perspective that I have adopted when analysing transnational practices of producing, consuming and exchanging food. As Bissell and Fuller (2011) have contended, however, while the conceptual focus on mobility has allowed us to understand the consequences of time-space compression with globalization, it has also left a gap in our understanding of the capacity of stillness ‘to do things’ (ibid.: 5). Gaibazzi (2015), in his study of embodied permanence in a Gambian village, draws attention to how stillness and sedentary livelihoods – set around the land – are created amidst movement, rather than being separated from it. In similar ways, Guinean migrants who rush around receiving, separating, weighing, packing, transporting and selling the food produce in spaces such the airport and the street markets described above, also experience long hours of sitting and waiting, in which the land is recreated through smells, tastes and shared memories. The work of Teresa and her son Wilson materializes such experiences in particularly unique ways. In a context of economic uncertainty, mother and son have developed strategies that illustrate how livelihood needs and the need for proximity to the distant Guinean land can be juxtaposed. Teresa first went to Portugal in 1993 to accompany her son Wilson, who had been offered a short-term visa in order to receive treatment for a health condition. Upon arrival she worked as a cleaner in a café and in people’s houses, ‘from early morning to late in the evening’, under the precarious working conditions that characterized the sector and that period in Portugal. Yet her own health issues and the physically demanding nature of these jobs led her to quit and search for alternatives. As she passed Rossio every day on her way to hospital, she noticed the group of Guinean women who, by then, already gathered in significant numbers with small foodstuffs to sell. ‘Once, I decided to approach them’, she explained, ‘to see how they did it. And I thought that it could be a good way to make a living’. Partnering with her husband Aladje, who had in the meantime started his food trade ‘cargo’ business in connection with TAP’s official air cargo services, Teresa managed to receive enough food to sell to others in bulk, acquiring a role as broker. Gradually, however, she identified other geographical areas with a demand for Guinean food and began her own itinerant selling business in one of those neighbourhoods – in the periphery of Lisbon – where a significant number of people of African origin resided. After Teresa’s return to Bissau, more than a year before we met, she moved on to coordinate the Guinean end of the business (selecting the produce to send and organizing the dispatch once a week), while her son
Food Livelihoods and New Economic Spaces
149
Wilson inherited the distributing role in Lisbon. He replaced Teresa’s trolley with a van and extended the distribution to other estates of that peripheral district of Lisbon, which had in the meantime been converted into social housing, managing a list of orders, clients’ names and contact numbers. In those areas, where many Guineans lived without easy access to Guinean food shops or street markets, the arrival of Wilson’s van was eagerly awaited every Sunday. As his van drove in each week, there was usually someone who first noticed his presence from a window, a balcony or a courtyard nearby, and announced it to the neighbours. On those occasions, a small crowd gathered around the produce, which was held and smelled and, while the details of the exchange were negotiated with Wilson, the land where those food products originated was remembered and celebrated. These were instances of stillness and reconnection with the land, amidst the movement embodied in Wilson’s van and the freneticism of his business. One Sunday morning, a young Guinean woman, Anabela, was grilling fish outside a block of flats in one of the neighbourhoods. Aware of the arrival of Wilson’s van, she promptly called out a name to the window of a nearby first floor. As soon as the child who had been called came to the window she said, ‘go to my place and get my wallet quickly. Teresa’s son is here’. Then, turning to Wilson, she added, ‘I want pounded shrimp. I haven’t had it in more than two years!’ As usual, a friendly conversation, soon shared by other residents who had in the meantime come outside, developed along with the smell of Anabela’s grilled fish, mixed with the different smells of Guinean foodstuffs that emanated from Wilson’s van. Anabela reached into the inside of the van, held the packages of pounded shrimp and brought them to her nose, asking Wilson about the produce’s origin. Wilson was known to her, as to most residents, as ‘Teresa’s son’, or nephew, indicating a kinship relation created through shared ethnic or regional origin by many who, in those neighbourhoods, were also of Manjaco background. While Wilson was not familiar with the exact location of the lumu where the pounded shrimp had been bought, or where it had come from originally, the fact that it was from ‘Guinean land’ and embodied another Manjaco’s investment of time and careful selection gave Anabela the necessary trust and the sense of plenitude and wholeness also described by Sutton (2001) in the context of Kalymnos. As she put it, ‘if Teresa got it, it is good shrimp’. Still holding the package against her nose and referring to its smell, she continued, ‘this is Guinean land. It is where my body is going to return when I die’. Then, turning to Wilson, she added, ‘if I take two bags, will you offer me a third?’ Wilson shrugged his shoulders and nodded with a friendly expression of resignation. In response to the fragmentation that characterized Anabela’s life as migrant, separated from her land of origin and possibility of actual return, Wilson’s van therefore brought her a sense of ‘returning to the whole’ that provided immediate satisfaction (Sutton 2001: 82), which she described as indispensable in fulfilling the feeling of sodadi di tera (nostalgia for the land).
Figure 5.4. Inside Wilson’s van, near Lisbon, 2010. © Maria Abranches.
150 Food Connections
Food Livelihoods and New Economic Spaces
151
Yet the distant land – made present through the arrival of food that had grown in it – was, for the clients in those neighbourhoods, associated with more than nostalgia. As this book has shown, the travel of food brings the Guinean transnational lifeworld together by materializing memories as much as future plans and aspirations, corporeal and spiritual protection to its people, relationships of giving, reciprocating and exchanging, and livelihood strategies. For Wilson, the distribution of Guinean food was a way to keep a family business going and to receive some income, but also to maintain an active role in the community and feel closer to the land. As he explained, ‘it’s not just about selling and buying. I stay and talk to people for an hour or two, I hear about Guinea-Bissau from those who have just come from there, and it helps me keep connected with my homeland’. During his food distribution activity, Wilson held a piece of paper with the clients’ names, whose histories he knew well. He had their contact numbers and the list of products that some had previously ordered. As more people gathered around the van, he called those with orders to receive or debts to pay and, if unable to reach them on the phone, he asked the neighbours who had already come down to go up to get them or he rang their doorbell directly. On such occasions, he was sometimes held back by his parenti (relatives by shared ethnic belonging), eager to share stories – as was the case that day – about Manjaco land and the work that migrant associations were doing to build a community centre in a Manjaco town. While he was temporarily away, the animated conversations continued near the back of his van where the food for sale was held and discussed. ‘I want everything in this van’, Anabela’s sister said, provoking laughter amongst the others. As in other spaces of exchange, however, Anabela, her sister and the other clients half-jokingly complained about the cost of the products. Yet Wilson’s business followed the exchange regulations agreed amongst all Guinean food sellers in Lisbon. If he occasionally sold the small bags of kandja, djagatu and badjiki for 2 Euros instead of the usual 2.50 Euros, for example, the difference was written in his notebook, although the accumulated debt could take a long time to be paid back or, as seen in other exchange practices in the previous chapter, remain unpaid. Wilson, now in his thirties, migrated to Portugal when he was fifteen years old and had not returned since, yet he claimed to be more connected with the land than many who travelled frequently, through his involvement in the food trade. While talking about what he perceived as a large number of Guinean transnational families, at home and abroad, living exclusively from Guinean food commercialization in various ways, he gave the example of his own parents who, before migration, worked in the public and hospitality sector, with no links to commerce. With migration, their diversified participation in the trade of food included not only his weekly distribution and his father’s intermediary role with TAP cargo services, but also, for example, the activity of smoking fish ‘in Guinean style’ and selling it in bulk to Guinean
152
Food Connections
shops. Their story illustrates some of the ways in which business opportunities bring together economic needs with the need for familiar sensations and experiences of tastes and smells. *** This chapter has demonstrated how migrant food businesses, which are dependent on the role of others in Guinea-Bissau, are a response to structural factors at home and abroad, such as a lack of opportunities in the regular labour market that is experienced in Guinea-Bissau, as it is in Portugal. In addition, they are also a response to possibilities that arise from the close-knit group networks that Guineans, like other West Africans, maintain across borders, more than they are part of an intrinsic ‘business culture’ or a ‘cultural predisposition to business’ (cf. Kloosterman, van der Leun and Rath 1998, 1999; Kloosterman and Rath 2001; Light and Gold 2000; Portes, Guarnizo and Haller 2002; Zhou 2004). Finally, they are intimately linked to a corporeal need for the familiar sensations embedded in homeland food. Teresa’s experience of having moved from the public sector in GuineaBissau to employment in the domestic sector in Portugal, followed by trading food and now organizing it from the Bissauan end, highlights the advantage of adopting a perspective that goes beyond the one centring on the host country only, in the study of transnational economic activities. A multi-sited critical analysis of the informal economy reveals that people’s economic life in countries of destination may rely, as much as in societies of origin, on multiple survival activities that escape governmental control, in combination with alliances with the formal realm (Lourenço-Lindell 2004). Guinean food traders’ interactions with customs authorities and local councils in the airport and market spaces have exemplified this permeability, which brings advantages and precariousness at the same time. Despite the basic income and a desired connection with the land that this business may bring for Guinean migrants, as Wilson claimed, it is important not to confound such experiences with an increased empowerment and autonomy for small actors, as is often assumed in neo-liberal discourse. Equally, however, it is essential to bring forward the active rather than passive role these actors play in formal-informal linkages, as their home-based kin do when creating market-oriented smallholdings and spontaneous food markets in urban Guinea-Bissau – the spaces of production with which this book began, which materialize the beginning of a unique transnational, yet small-scale and intimate food chain.
Food Livelihoods and New Economic Spaces
153
Notes 1. Meagher (2007, 2013) provides an overview of this literature. 2. A few exceptions are the works of Carreiro and Sangreman (2011), Có (2004) and Quintino (2004). 3. Fonseca and Malheiros (2004) and a special issue of the journal Migrações, edited by Oliveira and Rath (2008), provide examples of studies of more formal ethnic businesses in Lisbon. 4. Fátima Camara de Barros, president of the Association of Women in Economic Activities in Bissau – AMAE (Associação das Mulheres de Actividade Económica), interview with the author, January 2010. 5. As figure 5.3 indicates, at the time this book was written the Covid-19 pandemic was affecting Guinean migrants’ practices of food exchange and socialization in Lisbon. The photo, taken in 2021, shows people wearing face masks, in a smaller gathering than was customary during the period in which the research was conducted, and the only food products present are hidden inside a bag. Police officers were often seen enforcing social distancing in the area (Ana Estevens, personal communication with the author).
CONCLUSION
Amongst the things of the world are people, animals, plants. All of them are living beings, endowed with spirit… They are all born, grow up, have children, get ill/old and die. —Teresa Montenegro, As Enxadas do Rei (my translation)
Farmers, bideiras, sellers, buyers, brokers, drivers, storehouse and agency owners, carreta (trolley) boys, flight passengers, airport staff, healers, charm makers and ordinary food consumers form a close-knit network of people in Guinea-Bissau and in Portugal involved in a transnational, yet small-scale and intimate movement of food. By tracing Guinean food from its production sites in West Africa to its final spaces of consumption in Europe, this book has revealed the effects of this movement on the economic and social lives of people at home and abroad and the role of food in maintaining connections between both ends of the migration. Although food and migration are two of the most fundamental matters in West African economic and social life, academic interest in the relationship between the two is only incipient. While there is some concern with this relationship in discourses of food security, anthropological research as of yet has remained shy of asking what changes in the connection between production, exchange and consumption of homeland food – and the relationships these activities generate – when physical distance is imposed by migration. This book had this question at its core, and used food meanings and practices embodied in the various phases of the chain, on two sides of a migratory landscape, to explore it. The multi-sited ethnography on which this study was centred has demonstrated that subsistence, wellbeing, pleasure, risk, home, religion, ancestral origin, memories, desires, imaginaries, solidarity, expectations, obligations and income are all part of the significance of food for Guineans. The embeddedness of these meanings in everyday experiences and relationships related to the circulation of homeland food connects Guineans at home and abroad through intersubjective realities, contributing
156
Food Connections
to make people’s lives simultaneously local and integrated into new spaces that are full of possibilities and constraints. While these processes have started to attract some academic interest from the perspective of food consumption in host societies, ethnographic accounts that consider the relationship between migration and food practices of farming and exchange in sending communities are remarkably rare. Even more so are studies that look at both sides of the journey, which have only recently (Mata-Codesal and Abranches 2018) given continuation to Cook and Harrison’s (2007) pioneer study of the intimate and embodied connections between Caribbean consumption practices in North London and farming in the Caribbean. Alongside this gap in the literature, existing social science research in this field and, in particular, anthropological studies, face the challenge of developing theoretical frameworks that account for such dynamic interplays and processes, where ‘here’ and ‘there’ are mutually dependent (Abbots 2016). In this book, I have attempted to respond to this challenge by adopting a phenomenological-oriented approach that links the meanings people attribute to food to their concrete experiences of production, exchange and consumption across borders. A phenomenological perspective was also used here as a theoretical tool with which to respond to the criticisms that studies of transnationalism have received for focusing on the connections rather than on the actual experiences of people’s lives (Appadurai 1996; Burawoy et al. 2000; Coleman and Collins 2006; Coleman and von Hellermann 2011; Gardner and Grillo 2002; Gupta and Ferguson 1997; Mitchell 1997). As Dunn (2010: 1) has put it, centring the analysis upon migrants rather than migration flows, and upon transnationals rather than transnationalism, enables a much more realistic picture to emerge. I therefore combined multi-sited ethnography with a grounded approach to transnationalism that does not lose sight of the wider processes of globalization but, instead, grounds these processes in useful analytical tools (Burawoy et al. 2000).
Phenomenology, Food and Migration Nutrition and health are key concerns within Food Studies, where phenomenological approaches have been applied to understand, for example, the experience of hunger or food preferences. Cognitive psychologists, in particular, have looked at the relationship between those experiences and the mental imagery in sensory modalities beyond sight and hearing, such as taste and smell (Tiggemann and Kemps 2005). From an anthropological perspective, an approach in which the subjective and objective aspects of reality are not divided by clear boundaries, but rather articulated in intersubjective realities (Jackson 2015), seemed like a particularly useful theoretical lens
Conclusion157
with which to explore the influence of materials like food and plants in people’s lives.1 Drawing on Ram and Houston (2015), I have centred the analysis of food production, exchange and consumption across borders on how Guineans perceive and experience the materiality of food (the substance that links people and food to a common soil or land) and the relationships that it generates, in order to understand how they sense and comprehend the world when separation is imposed by migration. In the case of Guinean food production, exchange and consumption, there is a clear interconnection between people’s consciousness of the land and its cosmology, materialized in the intimate sharing of a common substance, and their relationship with others, including when these relationships are based on economic interests. As Weiss (1996: 125) has found amongst the Haya communities of northwest Tanzania, ‘concrete objects, namely cooked foods, are considered as embodiments of social practices, whose specific form and properties (e.g., their gendered dimensions, their temporal unfolding, their centrality or periphery) can be recognized and interpreted through the experience of these foods’. Despite an early academic interest in food as a cornerstone of culture and social organization, the anthropology of food is in many ways a new field of studies, and has so far focused mostly on consumption. However, looking at the food cycle from production to exchange and consumption, rather than only at the latter, enables a more holistic understanding of how meanings and values of food are combined with its bodily proprieties and ensuing relationships. When the cycle crosses borders, the experiences engendered are complicated, yet even more central in people’s lifeworlds. Indeed, this approach has elucidated how the realms of the symbolic and the material of Guinean food generate and transform relationships between farmers, traders, consumers and other intermediaries in two countries of a migratory landscape, linking West Africa and Europe in particular ways. Just as phenomenological-oriented studies of food have so far remained limited in anthropology, so is the study of migration through a phenomenological lens also recent and contained. Although Buttimer (1976) produced an early attempt to adopt phenomenological approaches in the study of the connections between space and the human experience, only more recently has this suggestion been taken on. Willen (2007), for example, applied a ‘critical phenomenological approach’ to the study of migrant ‘illegality’, adding an understanding of this juridical status and sociopolitical condition that is concerned with a mode of being‐in‐the‐world. More recent research has focused on a phenomenology of ‘dailiness’ (Stones et al. 2019), emphasizing the interplay between the everyday experience of the lived body and imagination of migrants and the structural contexts wherein they move, also responding to a critique that phenomenology often receives for failing to address the historical conditions that shape social life (Mattingly 2019). Given that anthropological studies of food and migration are still incipient, a phenomenological examination of this relationship is practically
158
Food Connections
inexistent. What this study has found by combining imaginaries, memories and aspirations embedded in food and migration with the experience of production, exchange and consumption of food across borders is that although geographical distance affects people’s lifeworlds and adaptations have to be made, it does not necessarily change the value of food or increase alienation between both ends of the chain. Producers, traders and consumers, as well as other intermediaries in the two countries, remain connected through their experiences of food, which ‘provide a vital opening between self and world’ (Ram 2015: 31). The phenomenological-oriented approach I have adopted in this study is one of anthropological orientation, focused on experience and ‘intersubjectivity’ – i.e. on how human subjectivity is experienced as objective reality in people’s lifeworlds, and on how emotions, senses, or ‘felt immediacies of bodies in a number of sociocultural settings’ intersect with political, social, economic and discursive formations (Desjarlais and Throop 2011: 90). In the case of Guinean actors in the transnational circulation of food, this approach has elucidated their concerns, engagements, imaginaries and aspirations, without losing sight of the obstacles they face in maintaining a lifeworld set in two countries. It has also contributed a new perspective to economic anthropology, by looking at how livelihoods generated by the transnational trade of Guinean food are embedded in intimate relationships with others and with the material itself – the substance that constitutes food. To the community and impersonal realms of the economy (Gudeman 2001) we can therefore add a third realm – that of the materiality of the exchange. That the social value of the material is central in people’s economic exchange relations is not new in anthropology. What this study has added is that, in the case of food, community-level exchange forms are kept with the migration of traders and of the traded material. The continued juxtaposition of community and impersonal exchanges and the connection, rather than alienation, between consumption and production – even when the chain is geographically separated by migration – result in part from the distinctiveness of food: its substance, vital function and capacity for ‘creating’ natives (Sahlins 1972, 1985). Anthropology has been recognized as uniquely positioned for analysing the relationship between the body and exchange, seeing the latter as productive of social ties, allegiances, enmities and hostilities that are almost viscerally connected to the wellbeing of the individual body (Peebles 2010). In this sense, an approach that focuses on how social relations and modes of perception take bodily form in people’s life can tell us a great deal about how they experience the world. From a phenomenological perspective, understanding these intimate, corporeal meanings of food and the ways in which people are capable of reshaping them rather than suspending them with migration (Appadurai 1986), involves paying attention not just to meanings as such but to how people recreate their lifeworlds by experiencing those reshaped meanings as
Conclusion159
a new reality (Weiss 1996; West 2007). This approach also sheds some light on the status of remittances in studies of migration. As I have suggested elsewhere (Abranches 2014a), the social value of homeland food might outweigh the economic value of remittances, thereby challenging the commonly assumed position of the migrants’ kin as the only dependent within relationships of gift exchange, whereby migrants remit and their home-based kin reciprocate. Although these relationships are obviously embedded in the more general complexity of individuals’ lives and their confrontation with everyday choices, gifts of food and plant medicines from home are essential for migrants to stay connected and to lessen the corporeal and spiritual disruptions resulting from migration, as much as financial remittances are an important source of income for families in Guinea-Bissau (ibid.: 272). I therefore suggest a phenomenology of food and migration as a theoreticalmethodological field of studies, which includes a grounded approach to transnationalism and multi-sited ethnography to understand how people on the two sides of a migratory landscape make their lifeworlds through food-related everyday experiences, shaped by the meanings associated with that food.
Towards a Phenomenology of Food and Migration in Africa Something needs to be said here about applying to Africa a theoretical perspective that has a western philosophical tradition. Within the postcolonial debate, the self-critical position of anthropology and a concern with its relevance in the production of knowledge in and of Africa has gained increased prevalence (Olukoshi and Nyamnjoh 2011), as have considerations on how to avoid imposing boundaries and alien categories (Binsbergan 2011). One challenge in applying phenomenology to the study of food and migration in a West African context is, precisely, the need to account for context. This is partly resolved in its anthropological configuration, which tends to shy away from the more general, culture-free pronouncements of phenomenological philosophers, preferring instead to ground its findings within specific cultural and historical settings (Desjarlais and Throop 2011). The work of Africana phenomenology, with which anthropology shares this concern with context, is also worth acknowledging here. This subfield of Africana philosophy, which results from the engagement of thinkers from Africa and its diaspora (inspired by the work of WEB Dubois, Frantz Fanon and Lewis Gordon, amongst others), has added the impact of racialization of the self to the focus on consciousness – i.e. the impact of the external colonization of one lifeworld (the African) by another (the European) (Henry 2005). Race has also had a prominent role in food studies.2 Williams-Forson (2013), for example, has explored the meanings of chicken for African American people and how this food is used in performances of power and race, including in producing stereotypes and as a form of resistance to them.
160
Food Connections
Other authors have looked at the racialized encounter resulting from the circulation of cuisines and globalization of food (Bell and Valentine 1997; Cook and Crang 1996; Narayan 1995). While I have found that the circulation of Guinean food across borders is able to maintain an intimate link between production and consumption, rather than alienation resulting from a need for adaptation that might be racialized, this does not mean that racial as well as gendered power relations are not part of the process, and these are worth some consideration here. A material approach to race in the context of food studies emphasizes the emergence of race as a process associated with agricultural techniques, embodied knowledge and decisions about particular crops and plants, encounters in food markets and questions of access to land and labour (Slocum 2011). It is the combination of these practices – rearranged by colonialism and liberalization reforms, as the history of agricultural political economy in Guinea-Bissau demonstrates – and the materiality of food itself that separates and connects bodies, hence creating race. In fact, the cases described in this book illustrate everyday racialized power geometries – from the production sites where Pepel and Mancanha farmers compete for land entitlement, to the ethnic-based contracts established between sellers and buyers of food across borders, or the way in which preparing or eating food of Fula origin, or in a Fula way, creates a space of ‘Fulaness’. As discussed in chapter 1, the notion of ethnicity bears the influence of colonial impositions in pre-existent intimate relations between land, territory and social organization, and Guineans in both countries use their subjective experience of belonging (to a place of origin, a religion, or specific foods and food cultivation and preparation methods) to make their everyday lived experiences. Moreover, this book has also shown that racial violence is experienced in spaces of food exchange and consumption in Portugal, where police control or the suspicious and discriminatory attitudes of others generate particular coping strategies amongst Guineans. Memory practices that refer to the past as a preferred time, by removing the colonial oppression that characterized it, are one such strategy. Gender has received even more attention in food studies, including in feminist literature on Africa and in African and Black feminist approaches, which have offered a historical view of current gendered hierarchical relationships in food-related activities, and their colonial influence. African and Black feminist studies of food have been critical of feminist literature on Africa, arguing that African women’s experiences tend to be reduced to the category of ‘victim’ in western theories. The main critique is that, by being initially concerned with women’s income from crop production and commercialization, and more recently with their role in food provisioning and nutritional welfare of their families, this literature has tended towards ‘othering’ and ignored African women’s ecological knowledge and their role as participatory agents of change (Wane 2014). In her analysis of indigenous
Conclusion161
food processing practices amongst Kenyan rural women, Wane has paid attention to the everyday experience that makes indigenous knowledge and to how women ‘make sense of the construction and sustainment of the textual imaging of their experiences’ (2014: 9). Claiming indigeneity, in this case, is an embodiment of knowledge: a way to reclaim a connection between body and mind, by being aware of history, reflecting on the present and projecting the future. This is also in line with an examination of experience from a phenomenological perspective, which takes into account the way in which ‘our past experience is always retained in a present moment that is feeding forward to anticipate future horizons of experience’ (Desjarlais and Throop 2011: 88). Women are the primary agents involved in the production, exchange and preparation of the food that travels regularly from Bissau to Lisbon. In Bissau, these women have experienced an intensification of such activities, generated by the transnational trade, which offered the possibility of increased income earning. Yet, without wishing to fall into the victimization discourse of feminist literature on Africa, women’s income from crop production and exchange remains limited and mostly used in the household. In Lisbon, too, patterns of migration have reshaped the gendered construction of spaces of exchange, where women have come to replace men in the street selling business, and men have been able to set up shops or brokerage businesses that allow them to deal with a larger quantity of food and, therefore, higher income. In Bissau, women often described their role as aimed mostly at helping their migrant kin’s business. In turn, migrant men often claimed to have started these activities as a response to their female relatives’ insistence from Bissau, indicating that constraints and obligations are combined with strategies put into place to maximize gains from both sides. Within these gendered dynamics, my ethnographic research has shown that women can also have control over the decision about what to send and how much to spend or keep from the money they receive from their kin abroad. The risks associated with credit, debt, trust and the morality of food exchange are of a gendered nature, as in other West African economic systems (cf. Buggenhagen 2011; Guérin 2006). However, by focusing on the interconnection between all those involved, not just in the exchange but in all stages of the chain, this book was primarily concerned with gaining a holistic understanding of how production, exchange and consumption remain intimately linked through women’s and men’s perceptions and experiences of food, even with geographical distance. In the literature on West Africa, and Africa more generally, food has received particular attention from the perspective of food security and concerns with rural development, urban farming and health and nutrition, with some studies of development economics and nutrition, for example, as well as international organizations, paying attention to the impact of migration on such concerns. In anthropological studies of food and migration, however,
162
Food Connections
Africa has remained surprisingly absent when compared with other regions of the world, such as Latin America and Asia. Particularly lacking are studies that look at the transnational connections between production and consumption of African food, with exceptions such as Abranches (2014a, 2014b), Renne (2007) and Oliveira (2018).3 Helping to fill this gap in the literature, this book has been primarily a multi-sited ethnographic investigation of the social and economic lives of a close-knit network of people involved in the production, exchange and consumption of homeland food between two locations in West Africa and Europe. Through an exploration of their food-related activities across borders – from farming to selling, buying, giving, reciprocating, preparing and eating in spatial and material landscapes – it is possible to trace the social meanings and values of food, as well as their adaptations, and better understand the relation between these and the everyday experiences that make people’s lifeworlds. Whereas this seems predominantly an anthropological concern, a phenomenological-oriented approach to food and migration in West Africa can also help to explain issues that meet the interests and concerns of disciplines and study areas like health and nutrition and development economics. Indeed, the proliferation and operationalization of urban farming, food markets and trade – which are concerns shared by many researchers on West Africa – do not operate outside a widespread migration and the new demands and challenges that come with the experience of being a migrant away from familiar food. Relatedly, international migration amongst West Africans is not removed from an intimate connection with the land of origin and, as a result, with the food that grows in it and associated food choices and practices. In fact, ignoring how farming and produce exchange are embedded in people’s perceptions and experiences of the land is one of the reasons for the gap between local and international discourses on food insecurity (Pottier 1999), and often makes scientific discourse on nutrient balance oblivious to what really matters to the people concerned (Pottier 2016). Within health and nutrition studies, some authors have already argued for the need to take into account the perceived medicinal value of certain foods for migrants when designing health care strategies, and to ensure that they maintain traditional eating habits by facilitating the availability of familiar food (Burns 2004). This is especially important given that some health conditions, such as diabetes and cardiovascular disease, tend to be more common amongst migrants due to, amongst other factors, dietary change (Burns 2004; Misra and Ganda 2007; Pieroni et al. 2007). A holistic investigation of such processes, perceptions and experiences, such as the one attempted in this book, can shed further light on how health-related or food security consequences of the (un)availability of familiar food are entwined with consequences at the level of relationships and livelihoods, also essential in connecting people’s economic and social lives in the context of migration.
Conclusion163
The circulation of Guinean food from Guinea-Bissau to Portugal, initiated with Aliu and his brother in the early 1990s – the story with which this book started – is crucial in the making of a Guinean lifeworld for both migrants and their families, even if it is a small-scale form of trade made of intimate connections between people in the two countries. In the case of Guinea-Bissau, formal exports have little significance in the global marketplace, with the exception of unprocessed cashew nuts, which, representing the country’s major export, caters predominantly to the Indian market. In 1993, a report by the US Agency for International Development identified a series of obstacles that justified the weak position of Guinean produce in the Portuguese market. Amongst those obstacles, poor packaging, labelling, refrigeration and transportation, and a lack of skilled personnel and management, were considered, by Portuguese importers, more critical than the actual quality of the produce, which was assessed as superior. The report concluded that there was a high level of tolerance by the authorities on both sides to let the produce be shipped, even if in what were considered far from ideal conditions. Although this was seen as a ‘temporary advantage of informality’, it was expected to change over time, following an increase of control patterns in Portugal (Miller 1993). So far, things have not changed. The courier system is still the main channel used by Guineans to move their food from the origin land to the migration setting. Taking part of its tchon with it, food offers bodily and spiritual protection, influences historical subjectivities, spatial and material memories and future aspirations, and impacts upon relationships of exchange and livelihoods. The connections it enables between both countries are embedded in the land, shared by people, animals, food and plants in what Teresa Montenegro (2009) – whose words opened this chapter – has named ‘the things of the world’. Rooted in the materiality and sociality of that common land, these connections, as well as their adaptations with migration, can be better understood through a view that does not separate experience of the world from its cognition. They also need to be understood not just as local food practices travelling to new spaces, but as an intrinsic part of the transformation of space – and, as a consequence, of the lives of its inhabitants – that is shaped by people and their food.
Notes 1. See Desjarlais and Throop (2011) for a review of phenomenological approaches in anthropology. 2. See Slocum (2011) for a review of this literature. 3. Although not recent, Shipton’s (1990) review of the anthropological literature on food security in Africa showed that migration was mainly only mentioned as a coping strategy to deal with famine and a solution to food shortages.
REFERENCES
Abarca, Meredith E. 2004. ‘Authentic or Not, It’s Original’, Food and Foodways 12(1): 1–25. Abbots, Emma-Jayne. 2008. ‘“It’s the Modern Way of Life”: Food, Mobility and Time Politics among Newly-Wealthy Campesinos in Highland Ecuador’. PhD dissertation. London: Goldsmiths, University of London. . 2011. ‘“It Doesn’t Taste as Good from the Pet Shop”: Guinea Pig Consumption and the Performance of Transnational Kin and Class Relations in Highland Ecuador and New York City’, Food, Culture & Society 14(2): 205–23. . 2016. ‘Approaches to Food and Migration: Rootedness, Being and Belonging’, in Jakob A. Klein and James L. Watson (eds), The Handbook of Food and Anthropology. London: Bloomsbury, pp. 115–32. Abranches, Maria. 2013a. ‘When People Stay and Things Make Their Way: Airports, Mobilities and Materialities of a Transnational Landscape’, Mobilities 8(4): 506–27. . 2013b. ‘Transnational Informal Spaces Connecting Guinea-Bissau and Portugal: Food, Markets and Relationships’, Urban Anthropology and Studies of Cultural Systems and World Economic Development 42(3,4): 333–76. . 2014a. ‘Remitting Wealth, Reciprocating Health? The “Travel” of the Land from Guinea-Bissau to Portugal’, American Ethnologist 41(2): 261–75. . 2014b. ‘Perceptions of the “Country” through the Migration of “City”Grown Crops: Guinean Food in Bissau and in Lisbon’, in Nuno Domingos, José Manuel Sobral and Harry G. West (eds), Food between the Country and the City: Ethnographies of a Changing Global Foodscape. London: Bloomsbury, pp. 127–42. Abreu, Alexandre. 2011. ‘Migration and Development in Contemporary GuineaBissau: A Political Economy Approach’, PhD dissertation. London: School of Oriental and African Studies, University of London. Adey, Peter. 2007. ‘“May I Have your Attention”’: Airport Geographies of Spectatorship, Position, and (Im)mobility, Environment and Planning D: Society and Space 25(3): 515–36. Aguilar, Renato, Huco Monteiro and Aquino Duarte. 2001. Bandim – Subsídios para uma Política de Apoio ao Pequeno Negócio. Bissau: Instituto Nacional de Estudos e Pesquisa. Ahmed, Iftikhar (ed.). 1985. Technology and Rural Women: Conceptual and Empirical Issues. London: George Allen & Unwin.
166References
Ahmed, Sara. 1999. ‘Home and Away: Narratives of Migration and Estrangement’, International Journal of Cultural Studies 2(3): 329–47. Al-Ali, Nadje and Khalid Koser. 2002. ‘Transnationalism, International Migration and Home’, in Nadje Al-Ali and Khalid Koser (eds), New Approaches to Migration? Transnational Communities and the Transformation of Home. London: Routledge, pp. 1–14. Alvarez, Robert R. and George A. Collier. 1994. ‘The Long Haul in Mexican Trucking: Traversing the Borderlands of the North and the South’, American Ethnologist 21(3): 606–27. Amselle, Jean-Loup and Elikia M’Bokolo (eds). 1985. Au Coeur de l’Ethnie: Ethnies, Tribalisme et État en Afrique. Paris: La Découverte. Anderson, Warwick, Ricardo Roque and Ricardo Ventura Santos. 2019. ‘Introduction’, in Warwick Anderson, Ricardo Roque and Ricardo Ventura Santos (eds), Luso-Tropicalism and its Discontents: The Making and Unmaking of Racial Exceptionalism. New York: Berghahn Books, pp. 1–20. Appadurai, Arjun. 1986. ‘Introduction: Commodities and the Politics of Value’, in Arjun Appadurai (ed.), The Social Life of Things: Commodities in Cultural Perspective. Cambridge: Cambridge University Press, pp. 3–63. . 1988. ‘How to Make a National Cuisine: Cookbooks in Contemporary India’, Comparative Studies in Society and History 30(1): 3–24. . 1996. Modernity at Large: Cultural Dimensions of Globalization. Minneapolis: University of Minnesota Press. Applbaum, Kalman. 2005. ‘The Anthropology of Markets’, in James G. Carrier (ed.), A Handbook of Economic Anthropology. Cheltenham: Edward Elgar Publishing, pp. 275–89. Arenas, Fernando. 2015. ’Migrations and the Rise of African Lisbon: Time-Space of Portuguese (Post)coloniality’, Postcolonial Studies 18(4): 353–66. Bâ, A. Hampaté. 1981. ‘The Living Tradition’, in J. Ki-Zerbo (ed.), General History of Africa I: Methodology and African Prehistory. Oxford: Heinemann (UNESCO), pp. 166–203. Banks, Nicola, Melanie Lombard and Diana Mitlin. 2020. ‘Urban Informality as a Site of Critical Analysis’, Journal of Development Studies 56(2): 223–38. Barth, Fredrik (ed.). 1969. Ethnic Groups and Boundaries: The Social Organization of Cultural Difference. Boston: Little, Brown and Company. Basu, Paul and Simon Coleman. 2008. ‘Introduction: Migrant Worlds, Material Cultures’, Mobilities 3(3): 313–30. Bava, Sophie. 2003. ‘De la “Baraka aux Affaires”: Ethos Économico-Religieux et Transnationalité chez les Migrants Sénégalais Mourides’, Revue Européenne des Migrations Internationales 19(2): 69–84. Bell, David and Gill Valentine. 2007. Consuming Geographies: We Are Where We Eat. London: Routledge. Bell, Vikki. 1999. ‘Performativity and Belonging’, Theory, Culture & Society 16(2): 1–10. Ben-Ze’ev, Efrat. 2004. ‘The Politics of Taste and Smell: Palestinian Rites of Return’, in Marianne Elizabeth Lien and Brigitte Nerlich (eds), The Politics of Food. Oxford: Berg, pp. 141–60. Bercovitch, Eytan. 1994. ‘The Agent in the Gift: Hidden Exchange in Inner New Guinea’, Cultural Anthropology 9(4): 498–536.
References167
Bielenin-Lenczowska, Karolina. 2018. ‘Spaghetti with Ajvar: An Ethnography of Migration, Gender, Learning and Change’, in Diana Mata-Codesal and Maria Abranches (eds), Food Parcels in International Migration: Intimate Connections. London: Palgrave Macmillan, pp. 117–39. Binsbergen, Wim van. 1988. ‘The Land as Body: An Essay on the Interpretation of Ritual among the Manjaks of Guinea-Bissau’, Medical Anthropology Quarterly 2(4): 386–401. . 1999. ‘Mary’s Room: A Case Study on Becoming a Consumer in Francistown, Botswana’, in Richard Fardon, Wim van Binsbergen and Rijk van Dijk (eds), Modernity on a Shoestring: Dimensions of Globalization, Consumption and Development in Africa and beyond. Leiden: Eidos, pp. 179–206. . 2005. ‘Commodification: Things, Agency and Identities: Introduction’, in Wim van Binsbergen and Peter Geschiere (eds), Commodification: Things, Agency and Identities (The Social Life of Things Revisited). Münster: Lit Verlag, pp. 9–51. . 2011. ‘Existential Dilemmas of a North Atlantic Anthropologist in the Production of Relevant Africanist Knowledge’, in René Devisch and Francis B. Nyamnjoh (eds), The Postcolonial Turn: Re-Imagining Anthropology and Africa. Oxford: African Books Collective, pp. 117–42. Bissell, David and Gillian Fuller. 2011. ‘Stillness Unbound’, in David Bissell and Gillian Fuller (eds), Stillness in a Mobile World. London: Routledge, pp. 1–17. Bissell, William Cunningham. 2005. ‘Engaging Colonial Nostalgia’, Cultural Anthropology 20(2): 215–48. Blunt, Alison. 2003. ‘Collective Memory and Productive Nostalgia: Anglo-Indian Homemaking at McCluskieganj’, Environment and Planning D: Society and Space 21(6): 717–38. . 2005. ‘Cultural Geographies of Home’, Progress in Human Geography 29(4): 505–15. Bohannan, Paul. 1955. ‘Some Principles of Exchange and Investment among the Tiv’, American Anthropologist 57(1): 60–70. . 1959. ‘The Impact of Money on an African Subsistence Economy’, The Journal of Economic History 19(4): 491–503. . 1967. ‘Africa’s Land’, in George Dalton (ed.), Tribal and Peasant Economies: Readings in Economic Anthropology. Austin: University of Texas Press, pp. 51–60. Bordonaro, Lorenzo I. 2009. ‘“Culture Stops Development!”: Bijagó Youth and the Appropriation of Developmentalist Discourse in Guinea-Bissau’, African Studies Review 52(2): 69–92. Boserup, Ester. 1970. Woman’s Role in Economic Development. London: George Allen & Unwin. Bourdieu, Pierre. 1977. Outline of a Theory of Practice, trans. R. Nice. Cambridge: Cambridge University Press. . 1984. Distinction: A Social Critique of the Judgement of Taste, trans. R. Nice. Cambridge, MA: Harvard University Press. . 1990. The Logic of Practice, trans. R. Nice. Cambridge: Polity Press. Bourdieu, Pierre and Loïc J.D. Wacquant. 1992. An Invitation to Reflexive Sociology. Cambridge: Polity Press. Boym, Svetlana. 2001. The Future of Nostalgia. New York: Basic Books. Brah, Avtar. 1996. Cartographies of Diaspora: Contesting Identities. London: Routledge.
168References
Brain, James L. 1976. ‘Less than Second-Class: Women in Rural Settlement Schemes in Tanzania’, in Nancy J. Hafkin and Edna G. Bay (eds), Women in Africa. Stanford, CA: Stanford University Press, pp. 265–84. Brown, Linda Keller and Kay Mussell (eds). 1984. Ethnic and Regional Foodways in the United States: The Performance of Group Identity. Knoxville: The University of Tennessee Press. Bryceson, Deborah. 1993. Liberalizing Tanzania’s Food Trade: The Public and Private Faces of Urban Marketing Policy, 1939-88. London: James Currey. Buggenhagen, Beth. 2011. ‘Are Births just “Women’s Business”? Gift Exchange, Value, and Global Volatility in Muslim Senegal’, American Ethnologist 38(4): 714–32. Burawoy, Michael et al. 2000. Global Ethnography: Forces, Connections, and Imaginations in a Postmodern World. Berkeley: University of California Press. Burkill, H.M. 1985. The Useful Plants of West Tropical Africa, vol. 1-6. London: Kew Publishing. Burns, Cate. 2004. ‘Effect of Migration on Food Habits of Somali Women Living as Refugees in Australia’, Ecology of Food and Nutrition 43(3): 213–29. Burrell, Kathy. 2008. ‘Managing, Learning and Sending: The Material Lives and Journeys of Polish Women in Britain’, Journal of Material Culture 13(1): 63–83. Butler, Judith. 1988. ‘Performative Acts and Gender Constitution: An Essay in Phenomenology and Feminist Theory’, Theatre Journal 40(4): 519–31. Buttimer, Anne. 1976. ‘Grasping the Dynamism of Lifeworld’, Annals of the Association of American Geographers 66(2): 277–92. Çağlar, Ayşe and Nina Glick Schiller. 2018. Migrants & City-Making: Dispossession, Displacement, and Urban Regeneration. Durham, NC: Duke University Press. Carreira, António. 1960. ‘Região dos Manjacos e dos Brames (Alguns Aspectos da sua Economia)’, Boletim Cultural da Guiné Portuguesa XV(60): 735–84. . 1964. ‘A Etnomínia dos Povos de entre o Gâmbia e o Estuário do Geba’, Boletim Cultural da Guiné Portuguesa XIX(75): 233–75. . 1966. ‘Aspectos Históricos da Evolução do Islamismo na Guiné Portuguesa (Achegas para o seu Estudo)’, Boletim Cultural da Guiné Portuguesa XXI(84): 405–55. Carreira, António and Artur Martins Meireles. 1959. ‘Notas sobre os Movimentos Migratórios da População Natural da Guiné Portuguesa’, Boletim Cultural da Guiné Portuguesa XIV(53): 7–19. Carreiro, Maria João and Carlos Sangreman. 2011. Arquitectos de um Espaço Transnacional Lusófono: A Diaspora Guineense em Portugal. Porto: Edição Fundação Portugal. Carrier, James G. 1992. ‘The Gift in Theory and Practice in Melanesia: A Note on the Centrality of Gift Exchange’, Ethnology 31(2): 185–93. Carsten, Janet. 2007. ‘Introduction: Ghosts of Memory’, in Janet Carsten (ed.), Ghosts of Memory: Essays on Remembrance and Relatedness. Malden, MA: Blackwell Publishing, pp. 1–35. Carvalho, Clara. 2002. ‘Ambiguous Representations: Power and Mimesis in Colonial Guinea’, Etnográfica 6(1): 93–111. Chabal, Patrick, Ulf Engel and Leo de Haan. 2007. ‘Introduction’, in Patrick Chabal, Ulf Engel and Leo de Haan (eds), African Alternatives. Leiden: Brill, pp. 1–7. Chanock, Martin. 1985. Law, Custom and Social Order: The Colonial Experience in Malawi and Zambia. Cambridge: Cambridge University Press.
References169
. 1991. ‘Paradigms, Policies and Property: A Review of the Customary Law of Land Tenure’, in Kristin Mann and Richard Roberts (eds), Law in Colonial Africa. Portsmouth, NH: Heinemann, pp. 61–84. Chapman, Tony. 2001. ‘There’s No Place Like Home’, Theory, Culture & Society 18(6): 135–46. Ciekawy, Diane and Peter Geschiere. 1998. ‘Containing Witchcraft: Conflicting Scenarios in Postcolonial Africa’, African Studies Review 41(3): 1–14. Cliggett, Lisa. 2005. ‘Remitting the Gift: Zambian Mobility and Anthropological Insights for Migration Studies’, Population, Space and Place 11(1): 35–48. Có, João Ribeiro. 2004. ‘As Associações das Comunidades Migrantes em Portugal e a sua Participação no Desenvolvimento do País de Origem: o Caso Guineense’, SOCIUS Working Paper 12/2004. Lisbon: Centro de Investigação em Sociologia Económica e das Organizações. Cole, Jennifer. 2001. Forget Colonialism? Sacrifice and the Art of Memory in Madagascar. Berkeley: University of California Press. Coleman, Simon and Peter Collins. 2006. ‘Introduction: “Being… Where?” Performing Fields on Shifting Grounds’, in Simon Coleman and Peter Collins (eds), Locating the Field: Space, Place and Context in Anthropology. Oxford: Berg, pp. 1–21. Coleman, Simon and Pauline von Hellermann. 2011. Multi-Sited Ethnography: Problems and Possibilities in the Translocation of Research Methods. New York: Routledge. Connerton, Paul. 1989. How Societies Remember. Cambridge: Cambridge University Press. Constable, Nicole. 2004. ‘Changing Filipina Identities and Ambivalent Returns’, in Lynellyn D. Long and Ellen Oxfeld (eds), Coming Home? Refugees, Migrants, and Those Who Stayed Behind. Philadelphia: University of Pennsylvania Press, pp. 104–24. Cook, Ian and Philip Crang. 1996. ‘The World on a Plate: Culinary Culture, Displacement and Geographical Knowledges, Journal of Material Culture 1(2): 131–53. Cook, Ian and Michelle Harrison. 2007. ‘Follow the Thing: “West Indian Hot Pepper Sauce”’, Space and Culture 10(1): 40–63. Counihan, Carole M. 1984. ‘Bread as World: Food Habits and Social Relations in Modernizing Sardinia’, Anthropological Quarterly 57(2): 47–59. . 2009. A Tortilla is Like Life: Food and Culture in the San Luis Valley of Colorado. Austin: University of Texas Press. Crush, Jonathan. 2013. ‘Linking Food Security, Migration and Development’, International Migration 51(5): 61–75. Csordas, Thomas J. 1993. ‘Somatic Modes of Attention’, Cultural Anthropology 8(2): 135–56. Dalakoglou, Dimitris. 2010. ‘Migrating-Remitting-“Building”-Dwelling: HouseMaking as “Proxy” Presence in Postsocialist Albania’, Journal of the Royal Anthropological Institute 16(4): 761–77. D’Alisera, Joann. 2001. ‘I Love Islam: Popular Religious Commodities, Sites of Inscription, and Transnational Sierra Leonean Identity’, Journal of Material Culture 6(1): 91–110. Davidson, Basil. 1966. A History of West Africa to the Nineteenth Century. New York: Anchor Books.
170References
Davidson, Joanna. 2003. ‘Native Birth: Identity and Territory in Postcolonial GuineaBissau, West Africa’, European Journal of Cultural Studies 6(1): 37–54. De Boeck, Filip. 1999. ‘“Dogs Breaking their Leash”: Globalization and Shifting Gender Categories in the Diamond Traffic between Angola and DRCongo (19841997)’, in Danielle de Lame and Chantal Zabus (eds), Changements au Féminin en Afrique Noire: Anthropologie et Littérature, vol. 1. Paris: L’Harmattan, pp. 87–114. De Bruijn, Mirjam, Rijk van Dijk and Jan-Barth Gewald. 2007. ‘Social and Historical Trajectories of Agency in Africa’, in Patrick Chabal, Ulf Engel and Leo de Haan (eds), African Alternatives. Leiden: Brill, pp. 9–20. Desjarlais, Robert and C. Jason Throop. 2011. ‘Phenomenological Approaches in Anthropology’, Annual Review of Anthropology 40: 87–102. Devisch, Renaat, Filip De Boeck and Danielle Jonckers (eds). 1995. Alimentations, Traditions et Développements en Afrique Intertropicale. Paris: l’Harmattan. Dias, Eduardo Costa. 1999. ‘Protestantismo e Proselitismo na Guiné-Bissau: Reflexões sobre o Insucesso do Proselitismo no Oio e na Província Leste’, Lusotopie 6: 309–18. Diouf, Mamadou. 2000. ‘The Senegalese Murid Trade Diaspora and the Making of a Vernacular Cosmopolitanism’, trans. Steven Rendall, Public Culture 12(3): 679–702. Domingues, Maria Manuela. 2000. ‘Estratégias Femininas entre as Bideiras de Bissau’. PhD dissertation. Lisbon: Universidade Nova de Lisboa. Douglas, Mary and Baron Isherwood. 1979. The World of Goods: Towards an Anthropology of Consumption. New York: Routledge. Drucker-Brown, Susan. 1995. ‘The Court and the Kola Nut: Wooing and Witnessing in Northern Ghana’, Journal of the Royal Anthropological Institute 1(1): 129–43. Dudley, Sandra H. 2010. Materialising Exile: Material Culture and Embodied Experience among Karenni Refugees in Thailand. New York: Berghahn Books. Dunn, Kevin. 2010. ‘Guest Editorial. Embodied Transnationalism: Bodies in Transnational Spaces’, Population, Space and Place 16(1): 1–9. Edensor, Tim. 2002. National Identity, Popular Culture and Everyday Life. Oxford: Berg. Embaló, Birgit. 2012. ‘Civil–Military Relations and Political Order in Guinea-Bissau’, Journal of Modern African Studies 50(2): 253–81. Eriksen, Thomas Hylland. 2010. Ethnicity and Nationalism: Anthropological Perspectives. London: Pluto Press. Escobar, Arturo. 1995. Encountering Development: The Making and Unmaking of the Third World. Princeton, NJ: Princeton University Press. Esteves, Alina et al. 2017. Condições de Vida e Inserção Laboral dos Imigrantes em Portugal: Efeitos da Crise de 2007-2008. Lisbon: Observatório das Migrações. Evans, Mike. 2001. Persistence of the Gift: Tongan Tradition in Transnational Context. Waterloo, ON: Wilfrid Laurier University Press. Evans-Pritchard, Edward E. 1937. Witchcraft, Oracles and Magic among the Azande. Oxford: Clarendon Press. Evers, Hans-Dieter. 1994. ‘The Traders’ Dilemma: A Theory of the Social Transformation of Markets and Society’, in Hans-Dieter Evers and Heiko Schrader (eds), The Moral Economy of Trade: Ethnicity and Developing Markets. London: Routledge, pp. 7–14. Fairhead, James and Melissa Leach. 1994. ‘Représentations Culturelles Africaines et Gestion de l’Environnement’, Politique Africaine 53: 11–24.
References171
. 1996. Misreading the African Landscape: Society and Ecology in a ForestSavanna Mosaic. Cambridge: Cambridge University Press. Falzon, Mark-Anthony (ed.). 2009. Multi-sited Ethnography: Theory, Praxis and Locality in Contemporary Research. Farnham: Ashgate. Fardon, Richard. 2008. ‘Cosmopolitan Nations, National Cosmopolitans’, in Pnina Werbner (ed.), Anthropology and the New Cosmopolitanism: Rooted, Feminist and Vernacular Perspectives. Oxford: Berg, pp. 233–59. Ferguson, James. 1999. Expectations of Modernity: Myths and Meanings of Urban Life on the Zambian Copperbelt. Berkeley: University of California Press. . 2002. ‘Of Mimicry and Membership: Africans and the “New World Society”’, Cultural Anthropology 17(4): 551–69. . 2013. ‘How to Do Things with Land: A Distributive Perspective on Rural Livelihoods in Southern Africa’, Journal of Agrarian Change 13: 166–74. Fikes, Kesha. 2008. ‘Diasporic Governmentality: on the Gendered Limits of Migrant Wage-Labour in Portugal’, Feminist Review 90: 48–67. . 2009. Managing African Portugal: The Citizen-Migrant Distinction. Durham, NC: Duke University Press. Flynn, Karen Coen. 2005. Food, Culture, and Survival in an African City. New York: Palgrave Macmillan. Fonkoué, Ramon A. 2019. Nation without Narration: History, Memory and Identity in Postcolonial Cameroon. New York: Cambria Press. Fonseca, Domingos. 1997. Os Mancanha. Bissau: Ku Si Mon Editora. Fonseca, Maria Lucinda and Jorge Malheiros. 2004. ‘Immigration and Globalisation from Below: The Case of Ethnic Restaurants in Lisbon’, Finisterra XXXIV(77): 129–56. Forrest, Joshua. 1992. Guinea-Bissau. Power: Conflict and Renewal in a West African Nation. Oxford: Westview Press. . 2002. ‘Guinea-Bissau’, in Patrick Chabal (ed.), A History of Postcolonial Lusophone Africa. London: Hurst & Company, pp. 236–63. Fortier, Anne-Marie. 1999. ‘Re-membering Places and the Performance of Belonging(s)’, Theory, Culture & Society 16(2): 41–64. . 2000. Migrant Belongings: Memory, Space, Identity. Oxford: Berg. Frazão-Moreira, Amélia. 2009. Plantas e ‘Pecadores’. Percepções da Natureza em África. Lisbon: Livros Horizonte. . 2016. ‘The Symbolic Efficacy of Medicinal Plants: Practices, Knowledge, and Religious Beliefs amongst the Nalu Healers of Guinea-Bissau’, Journal of Ethnobiology and Ethnomedicine 12(24): 1–15. Friedman, Jonathan. 2002. ‘From Roots to Routes’, Anthropological Theory 2(1): 21–36. Froment, Alain et al. (eds). 1996. Bien Manger et Bien Vivre: Anthropologie Alimentaire et développement en Afrique Intertropicale: Du Biologique au Social. Paris: L’Harmattan. Gable, Eric. 1998. ‘Appropriate Bodies: Self through the Other in Manjaco and Portuguese Representation, 1946-1973’, Visual Anthropology Review 14(1): 3–18. . 2003. ‘Manjaco Rulers after a Revolution’, Africa 73(1): 88–112. . 2006. ‘The Funeral and Modernity in Manjaco’, Cultural Anthropology 21(3): 385–415.
172References
Gaibazzi, Paolo. 2015. Bush Bound: Young Men and Rural Permanence in Migrant West Africa. New York: Berghahn Books. Galli, Rosemary E. 1990. ‘Liberalisation is not Enough: Structural Adjustment and Peasants in Guinea-Bissau’, Review of African Political Economy 49: 52–68. . 1995. ‘Capitalist Agriculture and the Colonial State in Portuguese Guinea, 1926-1974’, African Economic History 23: 51–78. Galli, Rosemary E. and Jocelyn Jones. 1987. Guinea-Bissau: Politics, Economics and Society. London: Frances Pinter. Gardner, Katy. 1993. ‘Desh-bidesh: Sylheti Images of Home and Away’, Man 28(1): 1–15. . 2002. Age, Narrative and Migration: The Life Course and Life Histories of Bengali Elders in London. Oxford: Berg. . 2008. ‘Keeping Connected: Security, Place, and Social Capital in a “Londoni” Village in Sylhet’, Journal of the Royal Anthropological Institute 14(3): 477–95. Gardner, Katy and Ralph Grillo. 2002. ‘Transnational Households and Ritual: An Overview’, Global Networks 2(3): 179–90. GEE. 2020. Resident Foreign Population in Portugal – Guinea-Bissau. Lisbon: GEE (Gabinete de Estratégia e Estudos). Retrieved 15 January 2021 from https:// www.gee.gov.pt/pt/lista-publicacoes/estatisticas-de-imigrantes-em-portugal-por-nacionalidade/paises/Guin%C3%A9-Bissau/3845-populacao-estrangeira-comestatuto-legal-de-residente-em-portugal-guine-bissau/file Gemmeke, Amber. 2018. ‘West African Plants and Prayers in the Netherlands: Nourishment through Visible and Invisible Substances’, in Diana Mata-Codesal and Maria Abranches (eds), Food Parcels in International Migration: Intimate Connections. London: Palgrave Macmillan, pp. 143–67. Geshiere, Peter. 1997. The Modernity of Witchcraft: Politics and the Occult in Postcolonial Africa. Charlottesville: The University Press of Virginia. Glick Schiller, Nina, Linda Basch and Cristina Blanc-Szanton (eds). 1992. ‘Towards a Transnational Perspective on Migration: Race, Class, Ethnicity, and Nationalism Reconsidered’, Annals of the New York Academy of Sciences 645. Godelier, Maurice. 1999. The Enigma of the Gift, trans. Nora Scott. Cambridge: Polity Press. Gomes, Elsa Teixeira et al. 2003. Plantas Medicinais da Guiné-Bissau: Manual Prático. Bissau: Acção para o Desenvolvimento (AD). Goody, Jack. 1982. Cooking, Cuisine and Class: A Study in Comparative Sociology. Cambridge: Cambridge University Press. Gottlieb, Alma. 1982. ‘Sex, Fertility and Menstruation among the Beng of the Ivory Coast: A Symbolic Analysis’, Africa 52(4): 34–47. Graeber, David. 1996. ‘Beads and Money: Notes toward a Theory of Wealth and Power’, American Ethnologist 23(1): 4–24. Granovetter, Mark. 1985. ‘Economic Action and Social Structure: The Problem of Embeddedness’, American Journal of Sociology 91(3): 481–510. Grassi, Marzia. 2003. Rabidantes: Comércio Espontâneo Transnacional em Cabo Verde. Lisbon: Imprensa de Ciências Sociais. Gregory, C.A. 1982. Gifts and Commodities. London: Academic Press. Grillo, Ralph. 1997. ‘Discourses of Development: The View from Anthropology’, in R.D. Grillo and R.L. Stirrat (eds), Discourses of Development: Anthropological Perspectives. Oxford: Berg, pp. 1–33.
References173
Grillo, Ralph and Valentina Mazzucato (eds). 2008. ‘AfricaEurope: Transnational Linkages, Multi-Sited Lives’ (Special Issue), Journal of Ethnic and Migration Studies 34(2). Gudeman, Stephen. 2001. The Anthropology of Economy. Malden, MA: Blackwell Publishing. Guérin, Isabelle. 2006. ‘Women and Money: Lessons from Senegal’, Development and Change 37(3): 549–70. Gupta, Akhil and James Ferguson. 1992. ‘Beyond “Culture”: Space, Identity, and the Politics of Difference’, Cultural Anthropology 7(1): 6–23. Gupta, Akhil and James Ferguson. 1997. ‘Discipline and Practice: “The Field” as Site, Method, and Location in Anthropology’, in Akhil Gupta and James Ferguson (eds), Anthropological Locations: Boundaries and Grounds of a Field Science. Berkeley: University of California Press, pp. 1–46. Guyer, Jane I. 1981. ‘Household and Community in African Studies’, African Studies Review 24(2/3): 87–137. Hage, Ghassan. 1997. ‘At Home in the Entrails of the West: Multiculturalism, Ethnic Food and Migrant Home-Building’, in Helen Grace et al. (eds), Home/World: Space, Community and Marginality in Sydney’s West. Sydney: Pluto Press Australia, pp. 99–153. Hagen, James M. 1999. ‘The Good behind the Gift: Morality and Exchange among the Maneo of Eastern Indonesia’, Journal of the Royal Anthropological Institute 5(3): 361–76. Hannerz, Ulf. 1996. Transnational Connections: Culture, People, Places. London: Routledge. . 2003. ‘Being There… and There… and There! Reflections on Multi-Site Ethnography’, Ethnography 4(2): 201–16. Hannoum, Abdelmajid. 2019. ‘Memory at the Surface: Colonial Forgetting in Postcolonial France’. Interventions 21(3): 367–91. Havik, Philip J. 1995. ‘Relações de Género e Comércio: Estratégias Inovadoras de Mulheres na Guiné-Bissau’, Soronda: Revista de Estudos Guineenses 19: 25–36. Heidegger, Martin. 1962. Being and Time, trans. J. Macquarrie and E. Robinson. Oxford: Basil Blackwell. Heimer, Franz-Wilhelm. 2002. ‘Fronteiras e Identidades Sociais em África’, in António Custódio Gonçalves (ed.), Multiculturalismo, Poderes e Etnicidades na África Subsariana. Porto: Faculdade de Letras e Centro de Estudos Africanos da Universidade do Porto, pp. 23–27. Henry, Paget. 2005. ‘Africana Phenomenology: its Philosophical Implications’, CLR James Journal 11(1): 79–112. Holtzman, Jon. 2006a. ’The World is Dead and Cooking’s Killed It: Food and the Gender of Memory in Samburu, Northern Kenya’, Food and Foodways 14(34): 175–200. . 2006b. ‘Food and Memory’, Annual Review of Anthropology 35: 361–78. . 2007. ’Eating Time: Capitalist History and Pastoralist History among Samburu Herders in Northern Kenya’, Journal of Eastern African Studies 1(3): 436–48. . 2009. Uncertain Tastes: Memory, Ambivalence, and the Politics of Eating in Samburu, Northern Kenya. Berkeley: University of California Press.
174References
Howes, David. 1996. ‘Introduction: Commodities and Cultural Borders’, in David Howes (ed.), Cross-cultural Consumption: Global Markets, Local Realities. London: Routledge, pp. 1–16. Idahosa, Pablo Luke Ehioze. 2002. ‘Going to the People: Amílcar Cabral’s Materialist Theory and Practice of Culture and Ethnicity’, Lusotopie 9(2): 29–58. Imbali, Faustino. 1992. ‘Um Olhar sobre o Sistema Alimentar Balanta: O caso das Tabancas de Mato Farroba e Cantone’, Soronda: Revista de Estudos Guineenses 14: 3–27. INE. 2011. Statistical Yearbook of Portugal 2010. Lisbon: INE (Instituto Nacional de Estatística de Portugal). Retrieved 15 January from http://www.ine.pt/xportal/ xmain?xpid=INE&xpgid=ine_publicacoes&PUBLICACOESpub_boui=13381539 0&PUBLICACOEStema=5414321&PUBLICACOESmodo=2 . 2013. Republica da Guiné-Bissau - Recenseamento Geral da Populaçao e Habitaçao 2009. Bissau: INE (Instituto Nacional de Estatística da Guiné-Bissau). Retrieved 15 January 2021 from http://www.stat-guinebissau.com/nada41/index. php/catalog/10 Jackson, Jean E. 2019. Managed Multiculturalism: Indigeneity and the Struggle for Rights in Colombia. Stanford, CA: Stanford University Press. Jackson, Michael. 1989. Paths towards a Clearing: Radical Empiricism and Ethnographic Enquiry. Bloomington, IN: Indiana University Press. . 2013. Lifeworlds: Essays in Existential Anthropology. Chicago: The University of Chicago Press. . 2015. ‘Afterword’, in Kalpana Ram and Christopher Houston (eds), Phenomenology in Anthropology: A Sense of Perspective. Bloomington, IN: Indiana University Press, pp. 293–303. Jackson, Peter, Philip Crang and Claire Dwyer. 2004. ‘Introduction: The Spaces of Transnationality’, in Peter Jackson, Philip Crang and Claire Dwyer (eds), Transnational Spaces. London: Routledge, pp. 1–23. Janowski, Monica. 2007. ‘Introduction: Feeding the Right Food: The Flow of Life and the Construction of Kinship in Southeast Asia, in Monica Janowski and Fiona Kerlogue (eds), Kinship and Food in Southeast Asia. Copenhagen: NIAS Press, pp. 1–23. Jao, Mamadú. 2003. ‘Origem Étnica e Migração entre os Mancanha da Guiné-Bissau’, Soronda: Revista de Estudos Guineenses 6(NS): 107–20. Je̜ drej, M.C. and Rosalind Shaw. 1992. ‘Introduction: Dreaming, Religion and Society in Africa’, in M.C. Je̜ drej and Rosalind Shaw (eds), Dreaming, Religion and Society in Africa. Leiden: Brill, pp. 1–20. Johnson, Michelle. 2006. ‘“The Proof is on my Palm”: Debating Ethnicity, Islam and Ritual in a New African Diaspora’, Journal of Religion in Africa 36(1): 50–77. . 2009. ‘Death and the Left Hand: Islam, Gender, and “Proper” Mandinga Funerary Custom in Guinea-Bissau and Portugal’, African Studies Review 52(2): 93–117. Kapferer, Bruce. 1997. The Feast of the Sorcerer: Practices of Consciousness and Power. Chicago: The University of Chicago Press. . 2002. ‘Introduction. Outside all Reason – Magic, Sorcery and Epistemology in Anthropology’, Social Analysis: The International Journal of Anthropology 46(3): 1–30. Karner, Christian. 2007. Ethnicity and Everyday Life. London: Routledge.
References175
Keane, Webb. 2005. ‘Signs are not the Garb of Meaning: on the Social Analysis of Material Things’, in Daniel Miller (ed.), Materiality. Durham, NC: Duke University Press, pp. 182–205. Kearney, M. 1995. ‘The Local and the Global: The Anthropology of Globalization and Transnationalism’, Annual Review of Anthropology 24: 547–65. Kershen, Anne J. (ed.). 2002. Food in the Migrant Experience. Aldershot: Ashgate. Kloosterman, Robert, Joanne van der Leun and Jan Rath. 1998. ‘Across the Border: Immigrants’ Economic Opportunities, Social Capital and Informal Business Activities’, Journal of Ethnic and Migration Studies 24(2): 249–68. . 1999. ‘Mixed Embeddedness: (In)formal Economic Activities and Immigrant Businesses in the Netherlands’, International Journal of Urban and Regional Research 23(2): 252–66. Kloosterman, Robert and Jan Rath. 2001. ‘Immigrant Entrepreneurs in Advanced Economies: Mixed Embeddedness further Explored’, Journal of Ethnic and Migration Studies 27(2): 189–201. Kopytoff, Igor. 1986. ‘The Cultural Biography of Things: Commoditization as Process’, in Arjun Appadurai (ed.), The Social Life of Things: Commodities in Cultural Perspective. Cambridge: Cambridge University Press, pp. 64–91. Kothari, Uma. 2008.‘Global Peddlers and Local Networks: Migrant Cosmopolitanisms’, Environment and Planning D: Society and Space 26(3): 500–16. Lambek, Michael. 2008. ‘Value and Virtue’, Anthropological Theory 8(2): 133–57. Lambek, Michael and Paul Antze. 1996. ‘Introduction: Forecasting Memory’, in Michael Lambek and Paul Antze (eds), Tense Past: Cultural Essays in Trauma and Memory. New York: Routledge, pp. xi–xxxviii. Law, Lisa. 2001. ‘Home Cooking: Filipino Women and Geographies of the Senses in Hong Kong’, Cultural Geographies 8(3): 264–83. Lee, Sandra Soo-Jin. 2000. ‘Dys-appearing Tongues and Bodily Memories: The Aging of First-Generation Resident Koreans in Japan’, Ethos 28(2): 198–223. Lentz, Carola. 2007. ‘Land and the Politics of Belonging in Africa’, in Patrick Chabal, Ulf Engel and Leo de Hann (eds), African Alternatives. Leiden: Brill, pp. 37–58. Lepri, Jean Pierre. 1986. ‘Contribuição para a Análise Sociológica da Guiné-Bissau Actual’, Soronda: Revista de Estudos Guineenses 1: 143–68. Levitt, Peggy. 1998. ‘Social Remittances: Migration Driven Local-Level Forms of Cultural Diffusion’, International Migration Review 32(4): 926–48. Levitt, Peggy and Mary C. Waters (eds). 2002. The Changing Face of Home: The Transnational Lives of the Second Generation. New York: Russell Sage Foundation. Levitt, Peggy, Josh DeWind and Steven Vertovec. 2003. ‘International Perspectives on International Migration: An Introduction’, International Migration Review 37(3): 565–75. Light, Ivan and Steven J. Gold. 2000. Ethnic Economies. San Diego: Academic Press. Linnekin, Jocelyn. 1991. ‘Cultural Invention and the Dilemma of Authenticity’, American Anthropologist 93(2): 446–49. Lobban, Richard and Peter Karibe Mendy. 1997. Historical Dictionary of the Republic of Guinea-Bissau. Lanham, MD: Scarecrow Press. Long, Lynellyn D. and Ellen Oxfeld. 2004. ‘Introduction: An Ethnography of Return’, in Lynellyn D. Long and Ellen Oxfeld (eds), Coming Home? Refugees, Migrants, and Those who Stayed Behind. Philadelphia: University of Pennsylvania Press, pp. 1–15.
176References
Long, Norman. 1996. ‘Globalization and Localization: New Challenges to Rural Research’, in Henrietta L. Moore (ed.), The Future of Anthropological Knowledge. London: Routledge, pp. 37–59. Long, Norman and Magdalena Villarreal. 1998. ‘Small Product, Big Issues: Value Contestations and Cultural Identities in Cross-Border Commodity Networks’, Development and Change 29(4): 725–50. Longhurst, Robyn. 1997. ‘(Dis)embodied Geographies’, Progress in Human Geography 21(4): 486–501. Lopes, Carlos. 1982. Etnia, Estado e Relações de Poder na Guiné-Bissau. Lisbon: Edições 70. . 1987. A Transição Histórica na Guiné-Bissau: do Movimento de Libertação Nacional ao Estado. Bissau: Instituto Nacional de Estudos e Pesquisa. . 2005. ‘O Kaabu e os seus Vizinhos: uma Leitura Espacial e Histórica Explicativa de Conflitos’, Afro-Ásia 32: 9–28. Lopes, Daniel Seabra. 2006. ‘Mercados Encobertos: os Ciganos de Lisboa e a Venda Ambulante’, Etnográfica 10(2): 319–35. Loude, Jean-Yves. 2003. Lisbonne dans la Ville Noire. Arles: Actes Sud. Lourenço-Lindell, Ilda. 1995. ‘The Informal Food Economy in a Peripheral Urban District: The Case of Bandim District, Bissau’, Habitat International 19(2): 195–208. . 2004. ‘Trade and the Politics of Informalization in Bissau, Guinea-Bissau’, in Karen Tranberg Hansen and Mariken Vaa (eds), Reconsidering Informality: Perspectives from Urban Africa. Uppsala: Nordiska Afrikainstitutet, pp. 84–98. MacGaffey, Janet and Remy Bazenguissa-Ganga. 2000. Congo-Paris: Transnational Traders on the Margins of the Law. Bloomington, IN: Indiana University Press. Machado, Fernando Luís. 2002. Contrastes e Continuidades: Migração, Etnicidade e Integração dos Guineenses em Portugal. Lisbon: Celta Editora. Malinowski, Bronislaw. 1922. Argonauts of the Western Pacific: An Account of Native Enterprise and Adventure in the Archipelagoes of Melanesian New Guinea. Reprint, London: Routledge & Kegan Paul, 1983. Marcus, George E. 1995. ‘Ethnography in/of the World System: The Emergence of Multi-Sited Ethnography’, Annual Review of Anthropology 24: 95–117. Massey, Doreen. 2005. For Space. London: Sage. Mata-Codesal, Diana and Maria Abranches (eds). 2018. Food Parcels in International Migration: Intimate Connections. London: Palgrave Macmillan. Mattingly, Cheryl. 2019. ‘Defrosting Concepts, Destabilizing Doxa: Critical Phenomenology and the Perplexing Particular’, Anthropological Theory 19(4): 415–39. Mauss, Marcel. 1950. The Gift: Forms and Functions of Exchange in Archaic Societies, trans. Ian Cunnison. Reprint, London: Routledge, 1990. Mavhunga, Clapperton Chakanetsa. 2014. Transient Workspaces: Technologies of Everyday Innovation in Zimbabwe. Cambridge, MA: MIT Press. Mbembe, Achille. 1985. Les Jeunes et l’Ordre Politique en Afrique Noire. Paris: L’Harmattan. Meagher, Kate. 2007. ‘Introduction: Special Issue on Informal Institutions and Development in Africa’, Africa Spectrum 42(3): 405–18. . 2013. ‘Unlocking the Informal Economy: A Literature Review on Linkages between Formal and Informal Economies in Developing Countries’, Cambridge, MA: Wiego Working Paper 27.
References177
Medina, F. Xavier and José A. Vázquez-Medina. 2018. ‘From Ingredient to Dish: The Role of Supply in the Culinary Practices of Mexican Migrants in the United States’, in Diana Mata-Codesal and Maria Abranches (eds), Food Parcels in International Migration: Intimate Connections. London: Palgrave Macmillan, pp. 191–210. Mendy, Peter Karibe. 1990. ‘A Economia Colonial da Guiné-Bissau: “Nacionalização” e Exploração, 1915-1959’, Soronda: Revista de Estudos Guineenses 9: 23–51. . 2003. ‘Portugal’s Civilizing Mission in Colonial Guinea-Bissau: Rhetoric and Reality’, The International Journal of African Historical Studies 36(1): 35–58. MEPIR. 2011. Segundo Documento de Estratégia Nacional de Redução da Pobreza (DENARP). Bissau: Ministério da Economia, do Plano e Integração Regional (MEPIR). Retrieved 15 January 2021 from http://www.imf.org/external/lang/ Portuguese/pubs/ft/scr/2011/cr11353p.pdf Merleau-Ponty, Maurice. 1945. Phenomenology of Perception, trans. Colin Smith. Reprint, London: Routledge, 1962. Miller, Daniel. 1994. Modernity – an Ethnographic Approach: Dualism and Mass Consumption in Trinidad. Oxford: Berg. (ed.). 2001. Home Possessions: Material Culture behind Closed Doors. Oxford: Berg. . 2008. The Comfort of Things. Cambridge: Polity Press. . 2012. Consumption and Its Consequences. Cambridge: Polity Press. Miller, John. 1993. ‘Portugal como Mercado para Produtos Frescos Agrícolas da Guiné-Bissau: Relatório Final’, Washington: USAID. Mills, David. 1999. ‘“Progress” as Discursive Spectacle: But What Comes after Development?’, in Richard Fardon, Wim van Binsbergen and Rijk van Dijk (eds), Modernity on a Shoestring: Dimensions of Globalization, Consumption and Development in Africa and Beyond. Leiden: Eidos, pp. 91–116. Min, Pyong Gap and Mehdi Bozorgmehr. 2000. ‘Immigrant Entrepreneurship and Business Patterns: A Comparison of Koreans and Iranians in Los Angeles’, International Migration Review 34(3): 707–38. Mintz, Sydney W. 1985. Sweetness and Power: The Place of Sugar in Modern History. New York: Penguin. Mintz, Sidney W. and Christine M. Du Bois. 2002. ‘The Anthropology of Food and Eating’, Annual Review of Anthropology 31: 99–119. Misra Anoop and Om P. Ganda. 2007. ‘Migration and Its Impact on Adiposity and Type 2 Diabetes’, Nutrition 23(9): 696–708. Mitchell, Jon P. 2001. ‘Introduction’, in Jon P. Mitchell and Paul Clough (eds), Powers of Good and Evil: Social Transformation and Popular Belief. New York: Berghahn Books, pp. 1–16. Mitchell, Katharyne. 1997. ‘Transnational Discourse: Bringing Geography back in’, Antipode 29(2): 101–14. Mommersteeg, Geert. 1988. ‘“He Has Smitten Her to the Heart with Love”: The Fabrication of an Islamic Love-Amulet in West Africa’, Anthropos 83: 501–10. Montenegro, Teresa. 2009. As Enxadas do Rei. Bissau: Ku Si Mon Editora. Moore, Henrietta L. 1988. Feminism and Anthropology. Minneapolis: University of Minnesota Press. . 1994. A Passion for Difference: Essays in Anthropology and Gender. Bloomington, IN: Indiana University Press. . 2011. Still Life: Hopes, Desires and Satisfactions. Cambridge: Polity Press.
178References
Moore, Henrietta L. and Todd Sanders. 2001. ‘Magical Interpretations and Material Realities: An Introduction’, in Henrietta L. Moore and Todd Sanders (eds), Magical Interpretations, Material Realities: Modernity, Witchcraft and the Occult in Postcolonial Africa. London: Routledge, pp. 1–27. MOPCU-DGHU. 2005. Plano Geral Urbanístico de Bissau. Bissau: MOPCU-DGHU (Ministério das Obras Públicas, Construção e Urbanismo, Direcção Geral de Habitação e Urbanismo). Mueller, Martha. 1977. ‘Women and Men, Power and Powerlessness in Lesotho’, Signs 3(1): 154–66. Munn, Nancy D. 1986. The Fame of Gawa: A Symbolic Study of Value Transformation in Massim (Papua New Guinea) Society. Durham, NC: Duke University Press. . 1992. ‘The Cultural Anthropology of Time: A Critical Essay’, Annual Review of Anthropology 21: 93–123. Murrey, Amber. 2017. ‘Decolonising the Imagined Geographies of “Witchcraft”’, Third World Thematics: A TWQ Journal 2(2-3): 157–79. Nakamura, Carolyn and Peter Pels. 2014. ‘Using “Magic” to Think from the Material: Tracing Distributed Agency, Revelation, and Concealment at Çatalhöyük’, in Ian Hodder (ed.), Religion at Work in a Neolithic Society: Vital Matters. Cambridge: Cambridge University Press, pp. 187–224. Narayan, Uma. 1995. ’Eating Cultures: Incorporation, Identity and Indian Food’, Social Identities 1(1): 63–86. Nash, Catherine. 2000. ‘Performativity in Practice: Some Recent Work in Cultural Geography’, Progress in Human Geography 24(4): 653–64. Niewswand, Boris. 2011. Theorising Transnational Migration: The Status Paradox of Migration. New York: Routledge. Nora, Pierre. 1984. Les Lieux de Mémoire, vol. 1: La République. Paris: Gallimard. Oliveira, Catarina Reis and Jan Rath (eds). 2008. ‘Empreendedorismo Imigrante’ (Special Issue), Migrações 3. Oliveira, Catarina Reis and Natália Gomes. 2019. Indicadores de Integração de Imigrantes: Relatório Estatístico Anual 2019. Lisbon: Observatório das Migrações. Oliveira, Tiago Silveiro. 2018. ‘Inkumenda di Téra: The Informal Circulation of Cabo Verdean Food Products’, in Diana Mata-Codesal and Maria Abranches (eds), Food Parcels in International Migration: Intimate Connections. London: Palgrave Macmillan, pp. 169–90. Olukoshi, Adebayo and Francis Nyamnjoh. 2011. ‘The Postcolonial Turn: An Introduction’, in René Devisch and Francis B. Nyamnjoh (eds), The Postcolonial Turn: Re-Imagining Anthropology and Africa. Oxford: African Books Collective, pp. 1–27. Osseo-Asa, Fran. 2005. Food Culture in Sub-Saharan Africa. Westport, CT: Greenwood Press. Parkin, David. 1999. ‘Mementoes as Transitional Objects in Human Displacement’, Journal of Material Culture 4(3): 303–20. Parry, Jonathan and Maurice Bloch. 1989. ‘Introduction: Money and the Morality of Exchange’, in Jonathan Parry and Maurice Bloch (eds), Money and the Morality of Exchange. Cambridge: Cambridge University Press, pp. 1–32. Parveen, Razia. 2017. Recipes and Songs: An Analysis of Cultural Practices from South Asia. London: Palgrave Macmillan.
References179
Peebles, Gustav. 2010. ‘The Anthropology of Credit and Debt’, Annual Review of Anthropology 39: 225–40. Peixoto, João. 2002. ‘Strong Market, Weak State: The Case of Recent Foreign Immigration in Portugal’, Journal of Ethnic and Migration Studies 28(3): 483–97. Pélissier, René. 2001a. História da Guiné: Portugueses e Africanos na Senegâmbia (1841-1936), vol. 1. Lisbon: Editorial Estampa. . 2001b. História da Guiné: Portugueses e Africanos na Senegâmbia (18411936), vol. 2. Lisbon: Editorial Estampa. Pels, Peter. 1998. ‘The Magic of Africa: Reflections on a Western Commonplace’, African Studies Review 41(3): 193–209. Petridou, Elia. 2001. ‘The Taste of Home’, in Daniel Miller (ed.), Home Possessions: Material Culture behind Closed Doors. Oxford: Berg, pp. 87–104. Pieroni, Andrea et al. 2007. ‘Medicinal Perceptions of Vegetables Traditionally Consumed by South-Asian Migrants Living in Bradford, Northern England’, Journal of Ethnopharmacology 113(1): 100–10. Piot, Charles. 1991. ‘Of Persons and Things: Some Reflections on African Spheres of Exchange’, Man 26(3): 405–24. . 1999. Remotely Global: Village Modernity in West Africa. Chicago: The University of Chicago Press. . 2010. Nostalgia for the Future: West Africa after the Cold War. Chicago: The University of Chicago Press. PNUD. 2006. Rapport National sur le Développement Humain en Guinée-Bissau 2006. Retrieved 15 January 2021 from http://hdr.undp.org/sites/default/files/rndhgw.pdf. Polanyi, Karl. 1944. The Great Transformation: The Political and Economic Origins of our Time. Boston, MA: Beacon Press. Portes, Alejandro, Luis Eduardo Guarnizo and William J. Haller. 2002. ‘Transnational Entrepreneurs: An Alternative Form of Immigrant Economic Adaptation’, American Sociological Review 67(2): 278–98. Portes, Alejandro and Min Zhou. 1992. ‘Gaining the Upper Hand: Economic Mobility among Immigrant and Domestic Minorities’, Ethnic and Racial Studies 15(4): 491– 522. Pottier, Johan. 1999. Anthropology of Food: The Social Dynamics of Food Security. Cambridge: Polity Press. . 2016. ‘Observer, Critic, Activist: Anthropological Encounters with Food Insecurity’, in Jakob A. Klein and James L. Watson (eds), The Handbook of Food and Anthropology. London: Bloomsbury, pp. 151–72. Quintino, Maria Celeste R. 2004. Migrações e Etnicidade em Terrenos Portugueses. Guineenses: Estratégias de Invenção de uma Comunidade. Lisbon: Instituto Superior de Ciências Sociais e Políticas da Universidade Técnica de Lisboa. Ram, Kalpana. 2015. ‘Moods and Method: Heidegger and Merleau-Ponty on Emotion and Understanding’, in Kalpana Ram and Christopher Houston (eds), Phenomenology in Anthropology: A Sense of Perspective. Bloomington, IN: Indiana University Press, pp. 29–49. Ram, Kalpana and Christopher Houston. 2015. ‘Introduction: Phenomenology’s Methodological Invitation’, in Kalpana Ram and Christopher Houston (eds), Phenomenology in Anthropology: A Sense of Perspective. Bloomington, IN: Indiana University Press, pp. 1–26.
180References
Ranger, Terence. 1983. ‘The Invention of Tradition in Colonial Africa’, in Eric Hobsbawm and Terence Ranger (eds), The Invention of Tradition. Cambridge: Cambridge University Press, pp. 211–62. Rapport, Nigel and Andrew Dawson. 1998. ‘Home and Movement: A Polemic’, in Nigel Rapport and Andrew Dawson (eds), Migrants of Identity: Perceptions of Home in a World of Movement. Oxford: Berg, pp. 19–38. Ray, Krishnendu. 2004. The Migrant’s Table: Meals and Memories in Bengali-American Households. Philadelphia: Temple University Press. Renne, Elisha P. 2007. ‘Mass Production Food Traditions for West Africans Abroad’, American Anthropologist 109(4): 616–25. Riccio, Bruno. 2008. ‘West African Transnationalisms Compared: Ghanaians and Senegalese in Italy’, Journal of Ethnic and Migration Studies 34(2): 217–34. Richards, Audrey I. 1939. Land, Labour and Diet in Northern Rhodesia: An Economic Study of the Bemba Tribe. London: Oxford University Press. Robertson, George et al. 1994. ‘As the World Turns: Introduction’, in George Robertson et al. (eds), Travellers’ Tales: Narratives of Home and Displacement. London: Routledge, pp. 1–6. Rose, Gillian. 1999. ‘Performing Space’, in Doreen Massey, John Allen and Philip Sarre (eds), Human Geography Today. Cambridge: Polity Press, pp. 247–59. Rouse, Roger. 1991. ‘Mexican Migration and the Social Space of Postmodernism’, Diaspora 1(1): 8–24. Sahlins, Marshall. 1972. Stone Age Economics. London: Routledge. . 1985. Islands of History. Chicago: The University of Chicago Press. . 1999. ‘Two or Three Things that I Know about Culture’, Journal of the Royal Anthropological Institute 5(3): 399–421. Salih, Ruba. 2003. Gender in Transnationalism: Home, Longing and Belonging among Moroccan Migrant Women. London: Routledge. Sanders, Todd and Harry G. West. 2003. ‘Power Revealed and Concealed in the New World Order’, in Harry G. West and Todd Sanders (eds), Transparency and Conspiracy: Ethnographies of Suspicion in the New World Order. Durham, NC: Duke University Press, pp. 1–37. Saraiva, Clara. 2008. ‘Transnational Migrants and Transnational Spirits: An African Religion in Lisbon’, Journal of Ethnic and Migration Studies 34(2): 253–69. Sarró, Ramon. 2010. ‘Map and Territory: The Politics of Place and Autochthony among Baga Sitem (and their Neighbours)’, in Jacqueline Knörr and Wilson Trajano Filho (eds), The Powerful Presence of the Past: Integration and Conflict along the Upper Guinea Coast. Leiden: Brill, pp. 231–52. Sato, Leny. 2007. ‘Processos Cotidianos de Organização do Trabalho na Feira Livre’, Psicologia & Sociedade 19: 95–102. Seremetakis, C. Nadia. 1994. ‘The Memory of the Senses: Part 1’, in C. Nadia Seremetakis (ed.), The Senses Still: Perception and Memory as Material Culture in Modernity. Boulder: Westview Press, pp. 1–18. Shaw, Rosalind. 2002. Memories of the Slave Trade: Ritual and the Historical Imagination in Sierra Leone. Chicago: The University of Chicago Press. Sheller, Mimi and John Urry. 2006. ‘The New Mobilities Paradigm’, Environment and Planning A 38(2): 207–26. Shipton, Parker. 1990. ‘African Famines and Food Security: Anthropological Perspectives’, Annual Review of Anthropology 19: 353–94.
References181
Shipton, Parker and Mitzi Goheen. 1992. ‘Introduction. Understanding African LandHolding: Power, Wealth, and Meaning’, Africa 62(3): 307–25. Silva, Artur Augusto. 1953. ‘Considerações sobre os Direitos de Família e Propriedade entre os Fulas da Guiné Portuguesa e suas Recentes Transformações’, Boletim Cultural da Guiné Portuguesa VIII(31): 405–15. Simmel, Georg. 1978. The Philosophy of Money, trans. Tom Bottomore and David Frisby. London: Routledge and Kegan Paul. Slocum, Rachel. 2011. ‘Race in the Study of Food’, Progress in Human Geography 35(3): 303–27. Smith, Michael Peter and Luis Eduardo Guarnizo (eds). 1998. Transnationalism from Below. New Brunswick, NJ: Transaction Publishers. Smith, Robert Courtney. 2006. Mexican New York: Transnational Lives of New Immigrants. Berkeley: University of California Press. Spooner, Brian. 1986. ‘Weavers and Dealers: The Authenticity of an Oriental Carpet’, in Arjun Appadurai (ed.), The Social Life of Things: Commodities in Cultural Perspective. Cambridge: Cambridge University Press, pp. 195–235. Stewart, Pamela J. and Andrew Strathern. 2003. ‘Introduction’, in Pamela J. Stewart and Andrew Strathern (eds), Landscape, Memory and History: Anthropological Perspectives. London: Pluto Press, pp. 1–15. Stoller, Paul. 1995. Embodying Colonial Memories: Spirit Possession, Power and the Hauka in West Africa. New York: Routledge. . 1996. ‘Spaces, Places, and Fields: The Politics of West African Trading in New York City’s Informal Economy’, American Anthropologist 98(4): 776–88. . 2009. The Power of the Between: An Anthropological Odyssey. Chicago: The University of Chicago Press. Stone, M. Priscilla, Angelique Haugerud and Peter D. Little. 2000. ‘Commodities and Globalization: Anthropological Perspectives’, in Angelique Haugerud, M. Priscilla Stone and Peter D. Little (eds), Commodities and Globalization: Anthropological Perspectives. Lanham: Rowman & Littlefield, pp. 1–29. Stones, Rob et al. 2019. ‘One World Is Not Enough: The Structured Phenomenology of Lifestyle Migrants in East Asia’, British Journal of Sociology 70(1): 44–69. Strathern, Marilyn. 1985. ‘Kinship and Economy: Constitutive Orders of a Provisional Kind’, American Ethnologist 12(2): 191–209. . 1988. The Gender of the Gift: Problems with Women and Problems with Society in Melanesia. Berkeley: University of California Press. Sudarkasa, Niara. 1986. ‘“The Status of Women” in Indigenous African Societies’, Feminist Studies 12(1): 91–103. Sutton, David E. 2001. Remembrance of Repasts: An Anthropology of Food and Memory. Oxford: Berg. Teixeira, Maria. 2008. ‘Sorcellerie et Contre-Sorcellerie: un Réajustement Permanent au Monde. Les Manjak de Guinée-Bissau et du Sénégal’, Cahiers d’Études Africaines XLVIII(1-2), 189–90: 59–79. Teixeira da Mota, Avelino. 1948. ‘Classificação e Evolução da Casa e Povoamento Indígena’, in Avelino Teixeira da Mota and Mário G. Ventim Neves (eds), A Habitação Indígena na Guiné Portuguesa. Bissau: Centro de Estudos da Guiné Portuguesa, pp. 9–136. . 1951. ‘Notas sobre o Povoamento e a Agricultura Indígena na Guiné Portuguesa’, Boletim Cultural da Guiné Portuguesa VI(23): 657–80.
182References
Temudo, Marina P. 2008. ‘From “People’s Struggle” to “this War of Today”: Entanglements of Peace and Conflict in Guinea-Bissau’, Africa 78(2): 245–63. . 2009. ‘A Narrativa da Degradação Ambiental no Sul da Guiné-Bissau: uma Desconstrução Etnográfica’, Etnográfica 13(2): 237–64. . 2011. ‘Planting Knowledge, Harvesting Agro-Biodiversity: A Case Study of Southern Guinea-Bissau Rice Farming’, Human Ecology 39: 309–21. Temudo, Marina P. and Manuel B. Abrantes. 2013. ‘Changing Policies, Shifting Livelihoods: The Fate of Agriculture in Guinea-Bissau’, Journal of Agrarian Change 13(4): 571–89. Thrift, Nigel. 1996. Spatial Formations. London: Sage. Tiggemann, Marika and Eva Kemps. 2005. ‘The Phenomenology of Food Cravings: The Role of Mental Imagery’, Appetite 45(3): 305–13. Tilley, Christopher. 2006. ‘Introduction: Identity, Place, Landscape and Heritage’, Journal of Material Culture 11(1-2): 7–32. Tolia-Kelly, Divya. 2004. ‘Locating Processes of Identification: Studying the Precipitates of Re-memory through Artefacts in the British Asian Home’, Transactions of the Institute of British Geographers 29(3): 314–29. . 2010. Landscape, Race and Memory: Material Ecologies of Citizenship. Farnham: Ashgate. Tuomainen, Helena Margaret. 2009. ‘Ethnic Identity, (Post)Colonialism and Foodways’, Food, Culture & Society 12(4): 525–54. Urry, John. 2000. Sociology Beyond Societies: Mobilities for the Twenty-first Century. London: Routledge. Uzendoski, Michael A. 2004. ‘Manioc Beer and Meat: Value, Reproduction and Cosmic Substance among the Napo Runa of the Ecuadorian Amazon’, Journal of the Royal Anthropological Institute 10(4): 883–902. Vertovec, Steven. 2009. Transnationalism. London: Routledge. Vigh, Henrik. 2006. Navigating Terrains of War: Youth and Soldiering in GuineaBissau. New York: Berghahn Books. Waldinger, Roger and David Fitzgerald. 2004. ‘Transnationalism in Question’, American Journal of Sociology 109(5): 1177–95. Wallace, Dale. 2015. ‘Rethinking Religion, Magic and Witchcraft in South Africa: From Colonial Coherence to Postcolonial Conundrum’, Journal for the Study of Religion 28(1): 23–51. Wane, Njoki Nathani. 2014. Indigenous African Knowledge Production: FoodProcessing Practices among Kenyan Rural Women. Toronto: University of Toronto Press. Watson, James L. 1983. Tairora Culture: Contingency and Pragmatism. Seattle: University of Washington Press. Watson, James L. and Jakob A. Klein. 2016. ‘Introduction: Anthropology, Food and Modern Life’, in Jakob A. Klein and James L. Watson (eds), The Handbook of Food and Anthropology. London: Bloomsbury, pp. 1–27. Weiner, Annette B. 1985. ‘Inalienable Wealth’, American Ethnologist 12(2): 210–27. Weiss, Brad. 1996. The Making and Unmaking of the Haya Lived World: Consumption, Commoditization, and Everyday Practice. Durham, NC: Duke University Press. Werbner, Richard. 1998. Memory and the Postcolony. African Anthropology and the Critique of Power. London: Zed Books.
References183
. 2002. ‘Introduction – Postcolonial Subjectivities: The Personal, the Political and the Moral’, in Richard Werbner (ed.), Postcolonial Subjectivities in Africa. London: Zed Books, pp. 1–21. West, Harry G. 2005. Kupilikula: Governance and the Invisible Realm in Mozambique. Chicago: The University of Chicago Press. . 2007. Ethnographic Sorcery. Chicago: The University of Chicago Press. . (ed.). 2011. ‘Food and Migration’ (Special Issue), Food, Culture & Society 14(2). Whitehead, Ann. 1981. ‘“I’m Hungry, Mum”: The Politics of Domestic Budgeting’, in Kate Young, Carol Wolkowitz and Roslyn McCullagh (eds), Of Marriage and the Market. London: Routledge, pp. 88–111. Wilding, Raelene. 2007. ‘Transnational Ethnographies and Anthropological Imaginings of Migrancy’, Journal of Ethnic and Migration Studies 33(2): 331–48. Wilk, Richard R. 1999. ‘“Real Belizean Food”: Building Local Identity in the Transnational Caribbean’, American Anthropologist 101(2): 244–55. Willen, Sarah S. 2007. ‘Toward a Critical Phenomenology of “Illegality”: State Power, Criminalization, and Abjectivity among Undocumented Migrant Workers in Tel Aviv, Israel’, International Migration 45(3): 8–38. Williams-Forson, Psyche. 2013. ‘More Than Just the “Big Piece of Chicken”: The Power of Race, Class, and Food in American Consciousness’, in Carole Counihan and Penny Van Esterik (eds), Food and Culture: A Reader. New York: Routledge, pp. 107–18. . 2014. ‘“I Haven’t Eaten if I Don’t Have My Soup and Fufu”: Cultural Preservation through Food and Foodways among Ghanaian Migrants in the United States’, Africa Today 61(1): 69–87. Wimmer, Andreas. 2008. ‘The Making and Unmaking of Ethnic Boundaries: A Multilevel Process Theory’, American Journal of Sociology 113(4): 970–1022. Wong, Siu-Lun. 1989. Emigrant Entrepreneurs: Shanghai Industrialists in Hong Kong. Hong Kong: Oxford University Press. World Bank. 2021. Guinea-Bissau. Retrieved 15 January 2021 from https://data. worldbank.org/country/guinea-bissau Wright, Marcia. 1983. ‘Technology, Marriage and Women’s Work in the History of Maize-Growers in Mazabuka, Zambia: A Reconnaissance’, Journal of Southern African Studies 10(1): 71–85. Zhou, Min. 2004. ‘Revisiting Ethnic Entrepreneurship: Convergencies, Controversies, and Conceptual Advancements’, International Migration Review 38(3): 1040–74.
INDEX A Abarca, Meredith E., 18, 55 abota (rotating savings system), 115 Abrantes, Manuel B., 31 agriculture, 29–30, 33–34. See also farming; harvest; production airport (of Bissau and Lisbon), 4, 16–17, 20, 35, 81, 86–87, 106, 113, 122, 129– 30, 135–36, 138 (figure 5.2), 139–41, 147–48, 152, 155 ajensia (transnational agency), 14, 19, 46, 103, 104 (figure 4.1), 105–6, 108, 139 ajensiador (transnational agency owner), 27, 79, 108, 125, 131, 144, 155 alienation, between production and consumption, 19, 108, 158, 160 amulets, 16, 18, 44–45, 55, 69–70, 82. See also charms; mesinhu animal products, 16, 18, 44–45, 52, 55, 68–69, 91 Appadurai, Arjun, 54, 95, 107, 109–10 authenticity, 7, 18, 26, 55–56, 60 B badjiki (roselle leaves), 2, 23, 31, 35, 38, 42, 53, 87, 119, 151 Bafatá, 1, 10, 30–31, 44, 46, 133 bagri (Guinean sea catfish), 56, 60, 109 baguitche (okra and eggplant), 35, 52, 102n3 Balanta, 10, 22n13, 29, 98 Tempu di (time of), 29 Bambadinca, 46, 69 banana leaves, 52, 61 bandida (criminal, immoral), 89 Bazenguissa-Ganga, Remy, 132 bazin (fabric), 77, 102n2, 125
belonging, ethnic, 10, 27–28, 30, 43, 48, 57–58, 64, 151. See also familia (parenti) bentana (Guinean tilapia), 56, 60 Biafada, 46, 82 bideiras (women food sellers), 36, 37 (figure 1.3), 38–40, 57, 77, 79, 81, 91, 99, 106, 109, 115–17, 141–42, 155 bifanas (pork sandwiches), 59, 96 birth, 23, 61 Bissau geographical division of, 31–32, 49n6 neighborhoods, history of, 16, 93, 95 swamp areas, 34 urban elite of, 66, 78 body weakening of, 51–53, 62–63, 67, 74, 81–82, 85, 92–93 Bohannan, Paul, 25, 110 Bolama Island, 27, 43, 48, 96–97 Bourdieu, Pierre, 5, 21n7 brokers, 4, 148, 155. See also intermediaries Bula, 26–27, 29, 42, 49n3 buyers, 17, 34, 42–43, 116, 145, 155, 160. See also clients; costumers C Çağlar, Ayşe, 114 cancer, 52, 74 cardiovascular disease, 54, 162 cargo (system, business), 136, 148 carriers, 4, 20, 80–81, 106, 129–30, 139–40 cashew nuts, 33, 39, 49n8, 86, 105, 145–46, 163 charms, 69–71, 74, 112, 155. See also amulets; mesinhu
186Index
chicken, smoked (smoking), 95–96 clients, 40, 42–43, 59, 71–73, 91, 93, 96, 98–99, 103, 105–8, 113–16, 118–19, 122–24, 136, 139, 142, 144–47, 149, 151. See also buyers; costumers clothes as trade, 77–78, 126, 133 Có, 29 colonialism, 9, 19, 24, 26–27, 32–33, 47, 98–100, 160 commensality, 60, 62, 124, 145 commodities, 19, 65, 105, 107 commodity debate, 108, 113 computers, study of, 82–83, 85 consciousness historical, 19, 30, 73, 93, 102 of the colonial past (of colonization), 97, 100, 159 construction sector, 2, 64, 118, 131, 133–34 consumers, 17, 20, 43, 54, 56, 59, 108, 110, 119, 155, 157–58 consumption across borders, 3, 28, 56, 108, 116, 124, 156–58, 162 and food (in anthropology), 1, 52–53, 157 ritual, 64, 136 spaces of food, 3, 15, 17, 48, 57, 59, 79, 102, 123, 155, 160 Cook, Ian, 56, 156 cooking, 55, 59–60, 71, 109. See also food preparation cosmology, 18, 23–24, 65, 68, 92, 157 cosmopolitanism, 19, 58, 85–86, 88–89, 91 coup d’état, 22n15, 99 courier system, 20, 136, 139, 163 cowry shells, 45, 66, 70 credit, 19, 113, 115–16, 119, 124, 135, 161 crisis, financial (2007–2008), 14, 20, 100, 115, 131–32, 134, 139 Criston (Christian), 28 crops, 3, 23–24, 32–33, 38, 57, 62, 73, 79, 116, 160–61 cuisine ethnic, 30 Indian, 58 Mexican, 54 national, 30, 60
Palestinian, 58 regional, 30 cultivation methods (techniques), 29–30, 160 customers, 19, 27, 103, 113, 115–17, 119. See also buyers; clients Cape Verdean, 145 customs (airport), 46, 69, 106, 130, 136, 152 D Damaia, 15, 20, 22n20, 58 (figure 2.1), 61, 63–64, 79, 88, 93, 94 (figure 3.1), 95–96, 114, 120 (figure 4.3), 122, 126, 144–45, 147 as food market, 20, 141–42 Davidson, Joanna, 26 death, 23, 27, 52, 61, 83, 93 debt, 19, 110–11, 113–15, 119, 124, 151, 161 development agricultural, 32, 47, 90 colonially inherited idea of, 89 economics, 21, 161–62 idioms of, 19, 34 imaginaries of, 80, 84 paradigm, 33 as a project, 82–85 rural, 161 diabetes, 53–54, 60, 162 Disabled War Veterans Association, 2, 14, 141. See also Palace of Independence discrimination, 69, 97, 100, 114, 146 distribution, 3, 20, 28, 38, 43, 124, 149, 151.. See also exchange djagatu (African eggplant), 23, 31, 35, 54, 106, 119, 151 djanfa, 18–19, 45, 55, 64–66, 71, 83, 86–89, 91–92, 101, 113, 130. See also sorcery; witchcraft djila (itinerant traders), 62 djumbai (socialization), 96, 145, 147 E eating, 48, 53, 59–60, 63, 71, 78–79, 89–90, 96, 109, 111, 123, 145, 160, 162 economic anthropology, 20, 113, 132, 158
Index187
African, 19, 124 elite, urban, 22n17, 31–32, 66, 78 ethnic groups, 10, 13–14, 22n13, 26–27, 30–31, 38, 43, 59, 80, 95–96, 116–17, 119, 127n3 ethnicity, 10, 17–18, 26, 28–30, 43, 48, 57, 160 Evers, Hans-Dieter, 113 exchange aesthetic element of, 119, 122 materiality of, 39, 77, 111, 158 morality of, 161 spaces of, 14–15, 17, 20, 21n3, 57, 59, 86, 91–93, 95, 102, 141–42, 145–47, 151, 153n5, 160–61 transnational, 19, 31, 38, 108–9, 113, 124, 136 See also distribution experience ‘anthropology of’, 4 lived, 3, 8, 18, 24, 28, 48, 60, 83, 89, 108, 160 F Fairhead, James, 25 familia (parenti), 27, 64, 96, 151. See also belonging, ethnic famine, 1, 32, 163n3 Farim, 27, 30 farmers, 4, 27, 29, 31–33, 43, 47, 116, 155, 157, 160 women, 17, 33–34, 36, 38, 47, 116 See also producers farming, 3, 24, 30–31, 80, 156, 162 commercial, 33–34, 36 urban, 161–62 See also agriculture; harvest; production faroba (néré), 39 fasting, 18, 61–63 feminist literature on Africa, 33, 36, 160–61 on food (African and Black), 160 Ferguson, James, 25, 78, 88 fertility, 23, 25, 61 Festas de Lisboa (annual city festivities), 59
fish, 30, 35, 42–43, 48, 56–57, 60, 75n5, 145, 149 Balanta (selebsonh or pis Balanta), 43 dried, 16, 42–43, 56, 87 smoked (smoking), 16, 42, 56, 58, 73, 93, 109, 151 foli (Guinea gum vine), 39, 87, 106, 113, 123, 145 fogon (stove), 38 fonio, 57, 60, 75n6 Fonkoué, Ramon, 84, 97 food adaptation, 18–19, 54–57, 60, 62–63, 73–74, 90, 102, 109, 122, 160, 162 demand of, 2, 4, 20, 35–36, 39–40, 47, 107, 114, 127, 148, 162 encounters, 7, 54–56 European (di branku), 52–53, 59–60, 73–74, 74n1 habits, 44, 52–53, 60, 67, 162 homeland, 2, 4, 7, 20, 36, 53–54, 61, 78, 86, 89, 106–9, 111–13, 132, 146, 152, 155, 159, 162 industrialized, 59, 73–74 markets, 3, 16, 20, 34–36, 47, 81, 93, 144–45, 152, 160, 162 materiality of, 3, 18, 54–55, 64, 66–67, 73, 101, 126–27, 157, 160 meanings of, 3–4, 7, 9, 17–18, 21, 48, 61, 66, 73, 85, 146, 155–59, 162 organic, 74, 90 packages (parcels), 17, 39, 43, 63, 81, 86, 103, 105–6, 119, 120 (figure 4.3), 122, 130, 136, 139, 149 preparation, 3, 10, 17–20, 21n8, 43–44, 48, 54, 57–58, 60–64, 73, 93, 100, 109, 124, 136, 160–61. See also cooking security, 1, 36, 85, 155, 161–62, 163n3 sharing, 3, 20, 56–59, 61, 73, 92–93, 102, 102n8, 108, 111–13, 123, 127, 144–45, 147 social value of, 19, 35, 78, 108–10, 113, 124, 158–59 touch of, 30, 63, 93, 100. See also senses unfamiliar, 54, 80 unknown, 45 foodstuffs. See food
188Index
foodways, 7, 21n8, 39, 48, 55–56, 85, 102 forgetting, 97–98 colonial, 97 fumeiro (smokehouse), 56, 58, 109 funerals, 27, 61, 67 Fula, 10, 12–14, 22n12, 27, 30–31, 38, 42–44, 47, 57–59, 61–64, 73, 91–92, 96, 98, 100, 124–26, 136, 160 diet (dish), 57, 62 as foreigners, 12, 30, 47, 64 Futa-, 12, 27 ‘Fulaness’, 30, 57, 64, 160 Futa-Djallon, 12, 27, 30
healers, 44–47, 52, 68–74, 155. See also kuranderu; muru healing, 18, 43–44, 46, 64, 66, 71–74, 82, 92, 95 health, 18, 23, 40, 45, 51–54, 59–60, 62, 67, 74, 82, 87, 92, 98, 148, 161–62 risks, 53, 82 studies, 4, 21, 54, 156, 162 treatment, 51, 148 Heimer, Franz-Wilhelm, 26 Holtzman, Jon, 60, 102n6 Houston, Christopher, 4 Howes, David, 54
G Gable, Eric, 12–13, 78, 92 Gabu, 10, 27, 30, 44, 60, 65 Gaibazzi, Paolo, 22n9, 82, 147–48 galinha di matu (Guinea fowl), 71 Gambia, 10, 25–26, 72, 79, 82, 147 Geba Christians, 13, 89, 97 River, 13, 42 gender, in food studies, 160–61 gifts, 16–17, 19, 39, 46, 56, 67, 88, 102n5, 103, 105–107, 111–13, 115, 125, 159 Glick Schiller, Nina, 114 globalization, 9, 141, 148, 156 of food, 160 gossip, 19, 86, 88–89, 91, 93, 112, 144. See also rumour groundnuts, 10, 32, 35, 39, 105, 113, 133, 140, 145 Gudeman, Stephen, 113–14 Guinea-Bissau history of, 2, 9–13, 32–33 migration from, 3, 13–14 Guinea-Conakry, 9, 12, 21n2, 24–25, 77, 125 Gupta, Akhil, 25
I illness, 4, 23, 51–52, 54, 62, 74 imaginaries, 17, 19, 38, 74, 78–80, 83–84, 90, 101–2, 108, 145, 155, 158 informality, 131–32, 135, 142, 163 informalization, 131 intermediaries, 17, 36, 38, 42–44, 58, 110– 11, 118, 124, 126, 140, 157–58 intermediary roles, 16, 19, 27, 39–40, 42, 57, 108, 116, 118, 124, 136, 151 See also brokers intersubjective mode of being, 83 Iran, 24, 29, 46, 64, 83, 130. See also spirits (of the ancestors)
H Hannoum, Abdelmajid, 97–99 Harrison, Michelle, 56, 156 harvest, 10, 17, 29, 31, 34, 36, 39–40, 47–48, 139. See also agriculture; farming; production
J Jao, Mamadú, 26–27 jealousy, 19, 45, 65, 86–87, 112 Johnson, Michelle, 28 K Kaabu, 26–27 kabas (calabash), 61, 100 kabasera (baobab fruit), 39, 53, 63, 94 (figure 3.1) ice-cream, 96, 115, 124, 144 kaldu branku (white sauce), 35, 90, 109 kaldu di mankara (groundnut sauce), 35, 63, 102n3, 121 (figure 4.4), 122–23 kaldu di tcheben (palm oil sauce), 35, 39, 55 kandja (okra), 35, 38, 42, 53, 106, 119, 151 Kapferer, Bruce, 65, 67 koitadesa (unfortunate, poor), 84–85, 125
Index189
konserva (lemon juice, chilli and onions), 53, 77, 102n1, 105, 134–35 Kopytoff, Igor, 107, 109 kuduru (bread), 105 kuranderu, 45. See also healers; muru L lalu (pounded baobab leaves), 53 land African, 17, 23–25 as material and spiritual, 16, 23–24, 27, 46, 54–55, 65–68, 70, 74, 92, 157, 163 meanings of, 17, 24, 47 nostalgia for (sodadi di tera), 149 politics of, 17 as protection (protection of), 18, 24, 65 See also soil; tchon; tera Largo São Domingos, 2, 21n3. See also Rossio Leach, Melissa, 25 liberalization, 31–33, 131, 160 lifestyle, 44, 82, 89–90 ‘café’, 88–89 lifeworld, 2–3, 7–9, 15–16, 18–20, 21n4, 30–31, 47, 53, 55, 61–65, 73–74, 78, 83, 86, 101, 114–15, 131, 136, 145–46, 151, 157–59, 162–63 linkages, formal-informal, 20, 131, 135, 147, 152 Lisbon Guinean migrants in, 10, 14, 39, 42–43, 56, 59, 61, 89, 91, 105, 107, 109, 124, 131 liti durmidu (curdled milk), 63–64 loans, 25, 113, 115 lokatero (transporter), 42 Lopes, Carlos, 26 Lourenço-Lindell, Ilda, 133, 135 luck (in economic activities), 42 lumu (weekly markets), 40, 42, 149 lupini beans, 60–61 ‘Lusotropicalism’, 97, 100 M MacGaffey, Janet, 132 Machado, Fernando Luís, 14, 21n5, 132 maize (corn), 56–57, 60, 62–64, 145 grilled 57–58, 73, 93
malagueta (chilli), 35, 52–53, 59, 109 Mancanha, 10, 14, 23–24, 26–27, 29, 31, 38–39, 43, 49n3, 124, 136, 160 Mandinga, 10, 12, 14, 22n12, 27–28, 31, 44–47, 61, 63–64, 69, 72–73, 77, 96 ‘Mandinganess’, 30, 64, 72 mandiple (spondias mombin), 95 mango, 39–40, 41 (figure 1.4), 48, 87, 94 (figure 3.1), 113, 123 (figure 4.6), 136 tree, 40, 142 Manjaco, 10, 13–14, 27, 31, 88, 92, 97, 149, 151 Marcus, George, 8, 22n11 market Bandim, 15–16, 82, 88, 95, 133 Caracol, 16, 20, 37 (figure 1.3), 38, 40, 42, 80, 91, 99, 106, 114, 116, 120 (figure 4.2), 136, 139, 141–42 gendered spatial division of, 142, 144 labour, 7, 20, 131–32, 152 Monte Abraão, 95–96, 146 principle, 20, 127, 141 space, 144, 147, 152 marriage, 61, 64, 91, 93, 100 material culture, 3, 6, 52 Meagher, Kate, 20, 147, 153n1 meat, 35, 52, 55, 59 ‘European’, 59 goat, 52, 105 Guinean cakes, 59 medicines, 15–16, 23, 43–45, 47, 52, 71, 82, 92, 159. See also mesinhu Medina, F. Xavier, 109 memories, 2–3, 6, 8, 17, 19–20, 25, 29, 57, 74, 78–79, 84, 93, 95–102, 102n6, 102n8, 107–8, 116, 148, 151, 155, 158, 160 embodied, 93 performative, 95 spatial and material approach to, 19, 92, 163 mesa (stall, selling place), 40, 49n10, 95, 98–99, 133, 142 mesinhu, 15–16, 44–47, 52, 69, 71, 82, 86, 91–92, 144. See also amulets; charms; medicines migration African, 1, 3
190Index
aspirations, 19, 31, 38, 79, 82–83, 101 control, 131, 146 policies, 7, 146 projects, 9, 38, 78, 80–81, 83, 85, 88, 100–1 West African, 7 millet, 57, 60, 62–64, 93 Mindará neighbourhood (Bissau), 95 Mintz, Sydney W., 7 mobility, 8, 19, 77–79, 148 social, 87 modernity, 19, 54, 65, 78, 85–86, 89, 102 moni (millet, sugar and water), 18, 62–64, 93 Moore, Henrietta L., 5–6, 33, 86, 101 moransa (residential area), 38 multi-sited ethnography, 1, 7–9, 17, 22nn10–11, 109, 152, 155–56, 159, 162 Murrey, Amber, 65–66 muru, 45–46, 71–72. See also healers; kuranderu Musulmano (Muslim), 28 N Nakamura, Carolyn, 18, 66 Nalu, 42 networks, 19, 78, 109, 114, 119, 125, 152 religious, 125 social, 117, 134, 140 transnational, 44, 57, 119 nostalgia, 95, 100, 149, 151 colonial, 97–98, 100 nutrition, 1, 21, 54, 156, 161–62 O obligations, 19, 23, 38, 61, 84–85, 105, 108, 113–14, 117, 119, 127, 155, 161 ondjo (roselle fruit), 77, 102n1, 134–35 orta, 23–24, 36, 40. See also smallholdings P packing, 80–81, 136, 139, 148 PAIGC (African Party for the Liberation of Guinea and Cape Verde), 13, 22n16, 32, 98–99, 102n9 Palace of Independence, 22n19, 141, 143 (figure 5.3). See also Disabled War Veterans Association palm kernels, 39, 42
palm oil, 16, 35, 39, 42, 55, 61, 87, 94 (figure 3.1), 105, 145 past, colonial, 7, 10, 26, 65, 80, 84, 95, 97–98, 100, 142, 160 pati (give, lend), 114–15 Pels, Peter, 18, 66 Pepel, 10, 29, 31, 49n6, 93, 95, 116, 124, 160 performativity, 6, 48 pharmacy (mesinhu selling place), 44, 46, 82 phenomenology, 4–5, 8, 156–57, 159 ‘of food and migration’, 20, 159 pidi, pidi (ask, ask), 88, 126 Pilum neighbourhood (Bissau), 95 Piot, Charles, 5, 65, 83, 85, 100–1, 110 plants, 2–3, 5, 14, 18, 23, 28, 65, 74, 80, 92–93, 155, 157, 160, 163 medicinal, 44–47, 49n11, 56, 95. See also medicines, mesinhu police (control, intervention), 63, 122, 144, 146, 153n5, 160 Portes, Alejandro, 116 Portugal migrants in (migration to), 2–3, 10, 13–14, 19, 22n13, 51, 60, 80, 126, 131, 134, 148, 151 postcoloniality, Portuguese, 96 Pottier, Johan, 36, 85, 107 prasa (urban centre), 32, 78, 89 producers, 17, 38, 42–43, 79, 158. See also farmers product specialization (in food businesses), 42, 144 production experience of, 3, 15, 141, 156, 158 spaces of, 3, 15, 18, 24, 29, 34, 47–48, 57, 152 See also agriculture; farming; harvest Q Qur’an, 46, 72 Qur’anic inscriptions (texts, verse), 46, 70, 72–73 R race, 27, 97 in food studies, 159–60 Ram, Kalpana, 4
Index191
Ramadan, 18, 61–64, 93 realities, intersubjective, 155–56 reciprocity, 17, 19, 105, 107, 110–12, 115 religion, 4, 12, 28, 30, 67–68, 73–74, 125, 142, 155, 160 remittances, 14, 27, 86, 91–92, 103, 107, 112–13, 127, 159 return, 78–79, 81, 86–92, 101, 102nn3–4, 109, 112, 117, 139–40, 147–49 aspirations, 19, 86, 101 migrants, 53, 78, 82, 87, 90 projects, 9, 78–79, 83, 86–87, 101 rice, 29–30, 32–36, 49n7, 52, 60–61, 68, 90, 99 Richards, Audrey I., 23, 42 ritual, 18, 39, 61–62, 64, 67, 71, 74, 123, 130, 136 140–42 Rossio, 2, 14–16, 20, 21n3, 59, 61–63, 68, 70, 77–79, 93, 96, 114–15, 122, 124, 134–35, 141–48 rumour, 65, 88. See also gossip S Sahlins, Marshall, 48, 55 sanctions, 24, 88, 91–92, 114, 117, 146 sapateiro (shoemaker), 69–71, 144 Sarró, Ramon, 24–25 seafood, 16, 42, 56, 87, 106 sellers, 2, 4, 16–17, 91, 95, 98, 108, 113–15, 123, 125, 141–42, 145, 151, 155, 160 Cape Verdean women, 57, 145 Chinese, 119 street, 93, 108, 145–47 (see also street selling) women, 17, 34, 40, 47, 62, 99, 115, 142, 147 See also traders, vendors Senegal, 9–10, 12–13, 25, 31, 62, 77, 133 senses, 4, 6–7, 65, 93, 158 shipping, 46, 82, 106–7, 140, 144 shops, Chinese, 60, 118–19, 122 simola (payment, offering), 66–68, 71, 90 skalada (dried and salted fish), 42–43, 102n3 skrita (writing), 70–71, 75n8 smallholdings, 16, 23, 152 urban, 17, 34–35, 47, 139 See also orta
smell, 6, 20, 39, 56–58, 78, 93, 100, 148–49, 152, 156. See also senses socialization, spaces of, 14, 19, 21n3, 59, 73, 93, 95, 145 soil, 10, 17, 23, 30, 47–48, 60, 71, 73, 145, 157. See also land; tchon; tera solidarity, 61, 82, 85, 88, 100–1, 105, 108, 114–15, 117, 155 female (women), 36, 99 sorcery, 45, 65, 67. See also djanfa; witchcraft sorghum, 57 space materiality of, 20, 86, 96, 101–2, 141 ‘spirit of politics’, 23–24, 28, 54 spirits (of the ancestors), 3, 18, 24, 28, 46, 65–67, 74, 85, 88, 130. See also Iran Stoller, Paul, 93, 145, 147 storehouses, 40, 42, 70, 136, 139, 155 Strathern, Marilyn, 53, 73 street selling, 125–26, 131, 142, 144, 161. See also sellers, street Structural Adjustment Programme (SAP), 33–35, 47, 99 sukulbembe (West African pepper), 35 Susu, 42, 88 Sutton, David E., 79, 111, 149 T TAP (Transportes Aéreos Portugueses), 129–30, 136, 151 taste, 6, 20, 39, 56–59, 61, 123, 145, 148, 152, 156 familiar, 53, 78, 93 Guinean, 35, 39, 57 See also senses tchon, 23, 27, 30, 43, 49n6, 70–72, 97, 163. See also land; soil; tera technology, 33, 84–86 Temudo, Marina P., 23, 28, 31, 49n7 tera, 23, 39, 44, 52, 144, 149 fidju di (children of), 30, 47, 73 kumida di (food of), 73, 111 See also land; soil; tchon tobacco, 14, 61, 144 trade materiality of, 3 morality of, 19
192Index
risks associated with, 80, 106–7, 117, 127, 130, 135, 139, 144, 147, 161 transnational, 3, 17, 20, 21n5, 38, 40, 42, 88, 116–18, 124, 132–34, 158, 161 traders, 1, 10, 12, 19–20, 38, 40, 42–43, 47, 57, 62, 71, 78, 85–86, 93, 113–19, 122, 124–26, 130–136, 146, 152, 157–58. See also sellers; vendors transnationalism, 5–9, 132, 141, 156 grounded approach to, 8, 156, 159 phenomenological-oriented approach to, 17 tree bark, 44, 94 (figure 3.1) roots, 44, 46, 94 (figure 3.1), 95 shade, 40, 77, 95, 142, 144 trust, 19, 43, 80, 113, 115, 117–18, 124, 135, 149, 161
vendors, 15, 36, 47, 56–57, 63–64, 107, 113, 118–19, 122, 134, 141–42, 146–47. See also sellers; traders
V Vázquez-Medina, Jose A., 109 veludo (velvet tamarind), 39, 63, 94 (figure 3.1), 146
Y yams, 59, 77–79, 115, 144
W Wane, Njoki Nathani, 33, 160–61 war civil, 13, 22n15, 131 liberation (independence), 13, 22n14, 26, 32, 98 veterans, 14 women participation in, 98 Weiss, Brad, 5, 65, 157 wellbeing, 18, 47, 73–74, 84, 98, 107, 112, 114, 155, 158 Williams-Forson, Psyche, 74, 159 witchcraft, 18, 55, 65–67, 75n7. See also djanfa; sorcery
Z Zhou, Min, 116