UNIVERISTY OF GHANA CENTRE FOR MIGRATION STUDIES TRANSNATIONAL CHILD RAISING ARRANGEMENTS: AN ETHNOGRAPHIC STUDY OF TRANSNATIONAL CAREGIVERS IN GHANA BY ERNESTINA KORLEKI DANKYI (10043191) THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON IN PARTIAL FULFILMENT OF THE REQUIREMNET FOR THE AWARD OF PHD MIGRATION STUDIES DEGREE JUNE 2014 University of Ghana http://ugspace.ug.edu.gh I DECLARATION I, Ernestina Korleki Dankyi, hereby declare that except for references to other people’s work which have been duly acknowledged, this thesis is the result of my independent research conducted at the Centre for Migration Studies, University of Ghana, Legon, under the joint supervision of Prof. Takyiwaa Manuh, Prof. Mariama Awumbila and Prof. Valentina Mazzucato. I also declare that as far as I know, this thesis has neither in part or in whole been published nor presented to any other institution for an academic award. ................................... ................................ Ernestina Korleki Dankyi Date Student ................................. Prof. Takyiwaa Manuh Date Principal Supervisor ................................... ................................. Prof. Mariama Awumbila Date Co-Supervisor ................................... .................................. Prof. Valentina Mazzucato Date Co- Supervisor University of Ghana http://ugspace.ug.edu.gh II DEDICATION This work is dedicated to Kwaku, Awurakua Boahemaa and Owuraku Mireku. I love you! University of Ghana http://ugspace.ug.edu.gh III ACKNOWLEDGEMENTS I have been richly loved on this journey and I would like to thank everybody who participated in this endeavour. I especially want to thank my supervisors, Prof. Takyiwaa Manuh, Prof. Mariama Awumbila and Prof. Valentina Mazzucato. You are supervisors par excellence! I also wish to thank the Centre for Migration Studies (University of Ghana) and the Department of Technology and Science and Society (Maastricht University) for jointly hosting me both in Ghana and the Netherlands during my PhD journey. My sincerest appreciation also goes to the Netherlands Organization for Scientific Research (NWO/WOTRO) for funding The TCRA project of which my PhD was a part. I am particularly grateful to my parents for being the primary caregivers of the children I left behind anytime I had to travel to the Netherlands or travel to Kumasi to collect data. Their support is immeasurable and I am forever grateful. My deepest appreciation also goes to my husband, Derrick Kwaku Dankyi, for the several lonely nights you endured and your kind support both in words and deeds. I love you! To my children, Awurakua and Owuraku, thanks for bearing with me. To Mrs. Boahemaa Ntim, thanks for being my biggest fan on this journey. To my family, both Tetteh and Dankyi especially my sister, Hannah and my grandparents. I love you. To Dr. Ernest Kwasi Adjei, thanks for a being a super host during my fieldwork in Kumasi. To Mr. Daniel Opoku-Agyemang, Mama Joyce and your family in Amsterdam, God bless you! To Patricia Mullenberg, formerly of Maastricht. Siyabonga! To Prof. Peter Quartey, we did it, Prof! To the Head, CMS, Dr. Delali Badasu. Akpe Kakaka! To Dr. Joseph Teye, Course coordinator, CMS, thanks for believing in me! To Dr. Akosua Keseboa Darkwah, the dedication page says it all. To the UG team - Mary, Francis, Shamsu-Deen, Kennedy and Dzifa and the UM team - Miranda, Karlijn, Djamila, Bilisuma, Victor, Marieke, Natalia, Kim, Patricia, Sabine. You people rock! I am particularly indebted to Miss Victoria Osei Bonsu and Margeret Meridith for editing this work. To the respondents of this study, caregivers, children and migrants who graciously provided data for this study, ‘Thank you’ is just not enough. My biggest and deepest gratitude goes to God Almighty and my Lord and Saviour Jesus Christ, who has loved me through each and every one of you. University of Ghana http://ugspace.ug.edu.gh IV ABSTRACT This study explores the experiences of caregivers in Transnational Child Raising Arrangements (TCRAs). It is part of a larger project which sought to examine how TCRAs affect life-chances of children who remain in the country of origin, their migrant parents and their caregivers in Ghana. It fills a lacuna in the literature on transnational parenting which have so far focused extensively on the migrant parents or the children they leave behind and the relationship between the migrants and their children. It also adds an African perspective to the discourse on transnational parenting as most of the existing studies are from Latin America and East Asia. The study particularly looks at the characteristics of the caregivers, their expectations and the challenges that they face within the context of child fosterage and is guided by a transnational approach. It is an ethnographic study that followed the lives of 15 caregivers in a matched sample (parent, child, and caregiver) for over two years and supplemented with one-off semi- structured interviews with non-matched 19 caregivers in the Greater Accra and Ashanti Regions. The study employs a conceptual framework that incorporates transnationalism, child fosterage norms and transnational care practices to make sense of the experiences of caregivers. The study found that kin constituted 88% of caregivers and most of the caregivers were women, largely affirming studies on transnational parenting which assert that women who migrate without their children fall on other women to help with child care. There was also the emerging phenomenon of migrants hiring the services of ‘private’ caregivers. The study also observed that more men were venturing into the traditional female sphere of care giving. Six out of the thirty-four caregivers were men, out of which four were the primary caregivers of the children. Secondly, caregivers in TCRAs have to deal with parents who have the ability to provide for the material needs of their children owing to their migrant status. Thanks to technological advancements, caregivers, parents and the children can ‘live here and there’ simultaneously, and this has implications for power relations in the TCRAs. The caregivers’ right to make major decisions concerning the children’s upbringing was almost non-existent, unlike in child fosterage as practiced within internal migration settings where the foster parents were entirely responsible for the upbringing of the child and thus the sole decision-makers of the child. The study establishes that the experiences of caregivers are largely shaped by a convergence of the norms of child fostering and transnational migration. The caregivers were caught between meeting society’s expectations of who a good caregiver is and new demands introduced through transnational migration. While traditional child fostering norms gave them greater control to raise the children, they were however restrained by parental involvement that is aided by the transnational status of the parents. The caregivers therefore had to navigate the demands that both child fosterage and transnational migration placed on them in order for their TCRAs to function properly. The study concludes among other things that caregivers are an active part of the transnational social field and their daily engagements both tangible and otherwise, aim at providing the physical and emotional needs of the children in their care as well as protecting the migrants from anxieties that may be engendered by the separation. In view of this, it is recommended for more attention to be paid to the perspectives of caregivers both in research and policy. The study further recommends that caregivers and migrants should endeavor to discuss the terms of the care arrangements – be they remittances, expectations of support from the migrants, or the duration of the care arrangement- to reduce the burdens that non- communication of these place on the caregivers. University of Ghana http://ugspace.ug.edu.gh V ABBREVIATIONS fCUBE - Free Compulsory Universal Basic Education GHS - Ghana Cedis GLSS - Ghana Living Standard Survey GSS - Ghana Statistical Service IOM - International Organization for Migration ISSER - Institute of Statistical Social and Economic Research JHS - Junior High School MOFEP- Ministry of Finance and Economic Planning NBSSI - National Board of Small Scale Industries NELM - New Economics of Labour Migration OECD - Organization for Economic Cooperation and Development PTA - Parent Teacher Association SHS - Senior High School SMS - Simultaneous Matched Sample SNA - Social Network Analysis TCRA - Transnational Child Raising Arrangements UNICEF- United Nations Children’s Fund University of Ghana http://ugspace.ug.edu.gh VI Table of Contents DECLARATION ....................................................................................................................... I DEDICATION ........................................................................................................................ II ABSTRACT............................................................................................................................ IV ABBREVIATIONS .................................................................................................................. V Table of Contents ................................................................................................................... VI List of Tables ............................................................................................................................ X List of figures ............................................................................................................................ X CHAPTER ONE: INTRODUCTION AND BACKGROUND TO THE STUDY ..................1 1.0 Introduction .......................................................................................................................1 1.1 The TCRA Project in Summary .........................................................................................4 1.2 Research Problem and Questions .......................................................................................7 1.3 Definition of a caregiver .................................................................................................. 10 1.4 Organization of the thesis................................................................................................. 10 CHAPTER TWO: LITERATURE REVIEW AND CONCEPTUAL FRAMEWORK ....... 13 2.0 Introduction ..................................................................................................................... 13 2.1 Transnationalism ............................................................................................................. 13 2.2 Transnational families ...................................................................................................... 21 2.3 Context in Transnational Parenting .................................................................................. 24 2. 4 Transnational Care Activities: Remittances and Communication ..................................... 27 2.4.2 Communication and Transnational Parenting ................................................................ 30 2.5 Summary and conceptual framework ............................................................................... 33 CHAPTER THREE: THE CONTEXT OF THE STUDY..................................................... 37 3.1 Overview of International Migration in Ghana ................................................................. 37 3.1.1 Historical overview ................................................................................................... 37 3.1.2 Estimates of Ghanaians abroad and destination .......................................................... 38 3.1.3 Ghanaian Migrants’ Remittances ............................................................................. 40 University of Ghana http://ugspace.ug.edu.gh VII 3.1.4 Transnational Activities of Ghanaians Abroad ........................................................... 42 3.2 Care and Care Giving in the Ghanaian Social Context...................................................... 44 3.3 Parenting in the Ghanaian Family .................................................................................... 48 3.3.1 Types of Child Fosterage ........................................................................................... 50 3.3.2 Child Fosterage and Social Change ........................................................................... 54 3.4 Summary ......................................................................................................................... 56 CHAPTER FOUR: RESEARCH METHODOLOGY AND STUDY AREAS ..................... 57 4.0 Introduction ..................................................................................................................... 57 4.1 Qualitative research ......................................................................................................... 57 4.2 Ethnographic Case Study ................................................................................................. 59 4.3 Selection of Cases -The Simultaneous Matched Sample (SMS) methodology .................. 62 4.3.1 How the SMS methodology was adopted for this study ............................................. 63 4.3.2 Challenges with the SMS........................................................................................... 68 4.4 Selection of Cases – Non-Matched Samples..................................................................... 69 4.5 Data Collection ................................................................................................................ 70 4.5.1 The Network Mapping Exercise - Name Generator Questionnaire ............................. 72 4.5.2 In-depth Interviews.................................................................................................... 74 4.5.3 Life history interviews with a focus on respondents upbringing. ................................ 75 4.5.4 Other Data Collection Activities ................................................................................ 76 4.6 Recording Data ................................................................................................................ 79 4.7 Data Analysis .................................................................................................................. 80 4.8 Ethical Considerations ..................................................................................................... 81 4.8.1Informed Consent, Confidentiality and Anonymity ..................................................... 81 4.8.2 Positionality and Representation ................................................................................ 84 4.9 Reflection: The Dilemma of an Ethnographer .................................................................. 87 4.10 Study Sites ..................................................................................................................... 91 4.10.1 Accra Metropolitan Area ........................................................................................ 91 4.10.2 Kumasi Metropolitan Area ..................................................................................... 94 CHAPTER FIVE: PROFILE OF THE CAREGIVERS ....................................................... 98 5.0 Introduction ..................................................................................................................... 98 University of Ghana http://ugspace.ug.edu.gh VIII 5.1 General Characteristics of The Caregivers ..................................................................... 102 5.1.1 Kin Caregivers ........................................................................................................ 102 5.1.2 Non-Kin Caregivers ................................................................................................ 104 5.1.3 Economic Status of the Caregivers .......................................................................... 105 5.1.4 Caregivers’ Marital status and own children. ........................................................... 108 5.1.5 Caregivers’ own upbringing .................................................................................... 108 5.2 Some Specific Characteristics of Caregivers .................................................................. 110 5.2.1 Male Caregivers ...................................................................................................... 110 5.2.1Special Kin - Stayed-Behind Wives .......................................................................... 114 5.2.3 Paid Caregiver – An Emerging Phenomenon? ......................................................... 119 5.3 Issues emerging from Caregivers’ Characteristics .......................................................... 125 5.4 Summary ....................................................................................................................... 131 CHAPTER SIX: TRANSNATIONAL CAREGIVERS: PATHWAYS AND EXPECTATIONS ................................................................................................................. 133 6.0 Introduction .................................................................................................................. 133 6.1 Pathways into care ......................................................................................................... 133 6.1.1 Factors that informed the decision to be a caregiver ................................................. 141 6.2 Caregivers’ Expectations ............................................................................................... 147 6.2.1 Features of Caregivers’ Expectations ....................................................................... 148 6.3Sum mary ....................................................................................................................... 160 CHAPTER SEVEN: ‘W’A YE FINE’ – ‘PROVIDING CARE’ AND ‘MANAGING CARE’ IN TCRAS ............................................................................................................................. 161 7.0 Introduction ................................................................................................................... 162 7.1 Providing Care ............................................................................................................... 163 7.2Managing Care ............................................................................................................... 171 7.2.1 Managing remittances ............................................................................................. 172 7.2.2 Managing Communication ...................................................................................... 179 7.2.3 Managing the Assumptions of Outsiders.................................................................. 190 7.2.4 Managing a Change in Lifestyle ............................................................................. 194 7.2.5. Managing the Unforeseen in TCRAs ...................................................................... 195 University of Ghana http://ugspace.ug.edu.gh IX 7.3 Summary ..................................................................................................................... 197 CHAPTER EIGHT: CAREGIVERS’ NETWORKS OF CARE IN TCRAS ..................... 198 8.0 Introduction ................................................................................................................... 198 8.1 Social Network Analysis (SNA)........................................ Error! Bookmark not defined. 8.2Characteristics of the Care Networks .............................................................................. 199 8.2.1 Size and Composition .............................................................................................. 199 8.2.2 Caregivers’ Network Members and Support Functions ............................................ 201 8.2.3 Incidental and Other Helpers in Caregivers’ Network .............................................. 202 8.2.4Caregivers/Migrants Network ................................................................................... 204 8.3 Small Networks ............................................................................................................. 204 8.3.1 Mobility in Urban Centres ....................................................................................... 205 8.3.2. Economic Status of Members in Caregivers’ Network ............................................ 207 8.3.4 Caregivers’ and Migrants’ Capacity to Provide Care .............................................. 209 8.3.6 Caregivers Do Not Ask For Help ............................................................................. 211 8.4 Problems with Small Networks .................................................................................... 213 8.5 Summary ..................................................................................................................... 214 CHAPTER NINE: SUMMARY OF FINDINGS, CONCLUSION AND RECOMMENDATION ........................................................................................................ 216 9.0INTRODUCTION .......................................................................................................... 216 9.1 Summary of empirical findings ..................................................................................... 217 9.2 Conclusions ................................................................................................................... 224 9.3 Recommendations.......................................................................................................... 230 9.4 Recommendations for future research ............................................................................ 237 References.............................................................................................................................. 238 APPENDICES ....................................................................................................................... 252 APPENDIX A: INTERVIEW GUIDE ................................................................................. 252 APPENDIX B: GIFT SENT THROUGH ERNESTINA TO CAREGIVERS/CHILDREN .. 255 APPENDIX C: NETWORK MAPPING TOOL ................................................................... 257 University of Ghana http://ugspace.ug.edu.gh X List of Tables Table 5.1: Profile of Caregivers (Matched Sample) ……………………………………..105 Table 5.2 : Profile of Caregivers (Non-Matched Sample) ..………………………………105 Table 5.3: Caregiver’ Economic Activity ………………………………………………..114 Table 7.1: Cecilia’s weekly expenditure ……..………………………………………….187 Table 8.1: Caregivers’ Network-Size and Composition …………………………………220 Table 8.2: Caregivers’ Network – Spatial Dispersion ……………………………………220 Table 8.3: Caregivers’ Network by Support Function ……………………………………222 List of figures Figure 2. 1: Conceptual Framework ………………………………………………… 42 Figure 4.2 : Map of Accra Showing Study Areas ………………………………………….100 Figure 4.3: Map of Ashanti region Showing study Areas…………………………………..103 Figure 5.1: Age distribution of Caregivers ………………………………………………….111 Figure 6.1: Features of Caregivers ………………………………………………………….161 University of Ghana http://ugspace.ug.edu.gh 1 CHAPTER ONE: INTRODUCTION AND BACKGROUND TO THE STUDY 1.0 Introduction Studies on international migration have focused largely on the economic consequences of migration at the expense of the social. There have also been many studies linking migration to development, although these put more emphasis on migrant remittances (Page and Adam, 2004; Lazcko, 2005; Adepoju, 2007). Migration has been claimed to be beneficial, especially for developing countries as the volume of remittances flowing into them are estimated to exceed Official Direct Investments (Ratha, 2007). At the household level, migration is also essential both as a livelihood enhancement strategy and a means of cushioning against vulnerabilities in times of economic hardships (Siddiqui, 2003; Quartey, 2006; Awumbila and Ardayfio- Schandorf, 2008). International migration has, however, affected the households of migrants in more ways than just economically. There are also social impacts of migration on migrant households. Most prevalent among them is splitting household members over long distances and several years of separation, as whole household units seldom migrate together. This phenomenon gives rise to what is referred to in the academic literature as ‘transnational families’, i.e. families that live across different national boundaries (Hondagneu-Sotelo and Avila, 1997; Bryceson and Vuorela, 2002; Parrenas, 2005). When parents, especially mothers, migrate without their children, they delegate their roles as primary caregivers and nurturers of children to others, which results in Transnational Child Raising Arrangements (TCRAs). TCRAs are prevalent worldwide for many reasons, some of which include stringent migration policies in migrant-receiving countries that make it difficult for families to migrate together, families’ attempts to escape violent conflict or University of Ghana http://ugspace.ug.edu.gh 2 persecution, or family members’ preferences, especially in societies where child fostering is a common practice (Mazzucato and Schans, 2011). In Latin America and parts of East Asia where studies on this phenomenon are prevalent, there have been records of benefits as well as negative consequences of TCRAs on both migrants and those they leave behind. However, the negative consequences, especially on mothers and children, are more pronounced in these studies (Hondagneu-Sotelo and Avila, 1997; Thompson and Bauer, 2000; Wong, 2006; Dreby, 2007; Parrenas, 2005). Whereas some studies have noted that children of migrants adjusted well after the separation (Coe, 2008; Dankyi, 2011), other studies have noted the negative effects on the children who report feeling abandoned and who blame their poor academic performance on their parents’ migration ( Suarez-Orozco et al., 2002; Parrenas, 2005). Words such as ‘plight’ have been used to describe the state of children that are left behind (Yeoh and Lam, 2006; Reyes, 2007). In cases where children are noted to be doing well, credit is given to good communication powered by new technologies, migrant remittances that are sent back home to take care of the left-behind children, or strong support from family members and relatives. Studies mention caregivers who take care of children while the parents are away, but they do not focus solely on them and also do not go into much detail in their discussions of them. “Caregivers are often treated as isolated units in a linear process of cause and effect, rather than as multi-dimensional individuals whose relationships with the cared for operate in a complex social system. What is missing is an attempt to understand caregivers on their own terms and in their own social locations” (Quinn and Evans, 2010:53). Studies on transnational parenting have indicated that when mothers migrate without their children, they depend on other people, especially women, to help with child care. Most of the existing studies have found that children were left in the care of grandparents, aunts, fathers, University of Ghana http://ugspace.ug.edu.gh 3 and friends and, in very rare cases, on their own (Ehrenreich and Hochschild, 2002; Bernard et al., 2005; Parrenas, 2005; Dreby, 2010; Dankyi, 2011; Akesson et al., 2012). The place of caregivers and their importance for transnational parenthood cannot be over emphasized. Moran-Taylor (2008) describes them as 'unsung heroes' and advocates that they be accorded equal recognition as migrants, explaining that without them, migration would be difficult for some people. However, studies on transnational parenting have focused largely on the dyad relationship between the migrant parents (usually the mother) and the left-behind children, without giving equal attention to the caregiver who cares for the children in the home country. Transnational families, especially those involving parent-child separations, have long been associated primarily with the parents and their children (see Ehrenreich and Hochschild, 2002; Bernard et al., 2005; Parrenas, 2005; Dreby, 2010). Madiannou and Miller’s (2012) book on transnational families and polymedia is dedicated to understanding a new type of connected transnational family, and, yet, their focus remains on the parents and children alone. Transnational families in which parents are separated from their children are not complete without those with whom the parents leave their children when they migrate. In the few studies where caregivers are mentioned, they are usually referred to as the recipients of remittances for the care of the migrants’ children. Furthermore, the information about caregivers is usually derived from interviews with the migrant or the left-behind child. These studies (Dankyi 2012; Coe 2011a), however, indicate that caregivers can be a source of strength for the children because they provide almost the same quality of care that the children would have received from their parents. But the reverse situation is also possible, when, for example, a caregiver has a family of his/her own and feels overburdened with the additional child care responsibilities. Such a situation can put a strain on relationships within TCRAs. Nevertheless, studies that pay University of Ghana http://ugspace.ug.edu.gh 4 particular attention to this essential member of the triad in TCRAs are scanty (See Moran-Taylor, 2008; Dreby, 2010; Coe, 2011). This study will therefore contribute to the literature on transnational parenting by focusing primarily on the caregivers of the children migrants leave behind, not just as a passive member of the triad, but as an active actor in his/her own right, thereby enabling the caregiver's perspective of TCRAs to be seen. 1.1 The TCRA Project in Summary This study on caregivers in TCRAs forms part of a larger project that explores the effects of TCRAs on all the members involved in the care triad. The research project is funded by the Science for Global Development Division (WOTRO) of The Netherlands Organization for Scientific Research (NWO) in The Netherlands. It aims to contribute to debates on migration and development by examining the effects of migration on children who stay behind as well as on caregivers in Ghana and the parents in The Netherlands. The main research question of the project is: How do TCRAs affect the life-chances of children who remain in the country of origin, their parents in the Netherlands and their caregivers in Ghana? The project addresses this question by integrating a multi-sited and mixed-methods approach. It encompasses four sub-projects: Sub-project 1 focuses on the effects of TCRAs on the life-chances of children and on schools in Ghana. It studies 2,760 children in four regions in Ghana. This thesis, which is sub project 2, focuses on the experiences of caregivers in the TCRAs from their own perspectives. It also investigates if and how negotiations within TCRAs are affected by changing child-fostering norms. Sub-project 3 focuses on the effects of TCRAs on Ghanaian parents in The Netherlands and how their decisions are affected by migration policies. Sub-projects 2 and 3 are linked through matched samples in which sub-project 2 works with the University of Ghana http://ugspace.ug.edu.gh 5 caregivers who take care of children of the migrants selected for sub-project 3. They are both qualitative and ethnographic studies. Sub-project 4 studies the effects of TCRAs on parents’ integration and well-being in The Netherlands, based on a survey of 300 Ghanaian migrants in Amsterdam. The project contributes to literature in three distinct ways. One of the contributions is methodological. The project employs a Simultaneous Matched Sample (SMS) Methodology. As has been noted above, studies have usually examined TCRAs primarily from the perspective of one person at a time and in the context of a single nation. To address this gap, the project conducts simultaneous research in two countries using a number of researchers. This is a methodological innovation that emphasizes the need to include migrants as well as non-migrants in samples in order to consider the multiple sites and levels of transnational social fields and to use qualitative as well as quantitative data collection methods and analyses (Mazzucato, 2009a). The second contribution is geographical. Previous studies have been conducted mainly on Latin American and South East Asian migrants and their families (Parrenas, 2005; Dreby, 2007, Hondagneu-Sotelo and Avila 2007). This project adds the perspective from Africa to debates on transnational child-raising. Studies from Latin America and South East Asia indicate that there are some negative consequences for children and parents in transnational child-raising arrangements such as conflicts and depressive symptoms (Levitt, 2001; Suarez-Orozco et al., 2002; Parrenas, 2005; Dreby, 2007, Olwig, 2007). In a study on effects of migration in Romania, the Soros Foundation concluded that social services need to be developed and provided to children who stay behind when their parents migrate. The foundation made this conclusion because the data confirmed the existence of a strong association between parents’ absence and the frequency of depression in children (Soros Foundation, 2007). However, there is University of Ghana http://ugspace.ug.edu.gh 6 reason to believe that TCRAs in Africa--especially West Africa--may not have the same effects as in other parts of the world because child fosterage is a long-standing institution in many parts of the Sub-region (Fiawoo 1978; Goody, 1982; Isuigo-Abanihe, 1985; Verhoef, 2005). The third contribution is the institutional analysis included in this project. Various institutions in migrant-sending and receiving countries influence TCRAs either by setting limits or facilitating the provision of care. Institutions in this project are defined similarly to the way North (1990:7) defines them: “those commonly held categories such as rules, laws, or norms of conduct that guide people’s actions and govern social arrangements”. Three institutions are analysed in this project: child fostering, and schools in the migrant-sending countries, and legal systems in the migrant-receiving countries. As part of the project, this study focuses on the experiences of caregivers in TCRAs. It looks in particular at their motivation, expectations and challenges. I worked together in a team of four researchers who looked at the TCRAs in relation to the other actors involved, namely migrant parents and the children of migrants. My particular focus was on the caregivers in Ghana. However, I worked more closely with a PhD researcher who studied migrant parents in The Netherlands. The SMS methodology employed by the project as a whole allowed us to match parents in The Netherlands to their children and caregivers in Ghana. This means that even though my unit of analysis was the caregiver in the TCRAs, I did not study their experiences in isolation but with reference to the parents and the children within a given TCRA. The study's institutional focus is on the widely practiced tradition of child fostering in many Ghanaian societies. The norms and principles of this practice have shaped the living arrangements of children since time immemorial. The practice persists although it has seen some transformation over the years. One of the areas where it is likely to see further change is within University of Ghana http://ugspace.ug.edu.gh 7 the settings of international migration where parents who travel abroad leave their children behind to be raised by others. This study will explore how transnational migration has shaped the institution of child fostering. 1.2 Problem Statement Transnational Child Raising Arrangements (TCRAs) have been studied across many disciplines including family sociology (Schmalzbeaur, 2004; Dreby, 2007), gender studies (Hondagneu-Sotelo & Avila, 1997; Parrenas, 2001, 2005), psychology (Suarez-Orozco, 2002) and legal studies (Bernhard et al, 2005; Van Walsum, 2006). Mazzucato and Schans (2008), however, observed in a review of these studies that they did not incorporate the socio-cultural institutions and norms that underlie parenting in the migrant's home countries. In explaining the outcomes of TCRAs on the people involved, these studies assume the western model of parenting which emphasizes the role of the biological mother as the sole nurturer of the child. It also privileges face-to-face interactions between the mother and child as a prerequisite for effective child development. This assumption is problematized by socio-cultural settings like that of West Africa, where a child is considered to belong not only to the biological parents, but also to the extended family. Within such settings, it is common to hear phrases like 'a child has many mothers' (Verhoef, 2005), or ‘it takes a village to raise a child’ (Clinton, 2006). The work of Suarez-Orozco et al. (2002) corroborates this, noting that many immigrants come from cultures which include a wide supportive network of extended family members. Secondly, there is a substantial body of literature on transnational care practices (Horst, 2006; Banfi and Boccagni, 2010; Graham and Jordan, 2011) but these have focused largely on University of Ghana http://ugspace.ug.edu.gh 8 the advantages of these practices to migrant parents and the children they leave behind. These practices which includes communication, sending of remittances and return visits, usually serve as the ‘glue’ that binds members of transnational families (Vertovec, 2004; Parrenas, 2005; Dreby, 2010) The adverse effects that the absence of these care practices can pose to members in transnational families have been also examined (Coe, 2011; Graham and Jordan, 2011). However, what remains largely underexplored are the practical issues associated with these practices, and how these affect the rights and responsibilities of the caregiver in TCRAs. The issues of availability and affordability and the challenges associated with these practices are rarely examined in the literature on transnational families. This study provides a much needed addition to the scant number of studies which examine these issues by taking a broader look at transnational care practices. Finally, this study will add an African perspective to the debates on transnational child- raising. So far, the growing literature on transnational parenting revolves around studies that have largely been carried out in Latin American and East Asian contexts. Mazzucato, (2013) describes this as a geographical bias. The African situation has largely been neglected, and the few studies on the Ghanaian case (Coe, 2008; 2011; Dankyi, 2008) have not paid attention to the issue of child fosterage and its role in TCRAs. University of Ghana http://ugspace.ug.edu.gh 9 1.3 Research Objectives Drawing from the gaps identified, specifically the near absence of the perspectives of caregivers in the African context, the present study aims at exploring the experiences of caregivers in transnational child raising arrangements from their own perspective. The specific objectives are to: -Examine the socio-economic characteristics of the caregivers -Explore their motivations and expectations for entering into their care arrangements -Examine how the caregivers’ own upbringing affects their caregiving role -Add to the scanty African literature on care arrangements in TCRAs 1.4 Research Questions The study will be approached from a transnational perspective, where the features of transnationalism converge with local parenting practices and norms to present an understanding of the experiences of caregivers in TCRAs in Ghana. The study seeks to answer the following questions: 1. How do the caregivers come into their care giving roles? What are the processes and discussions that take place before they agree to take care of the children of migrants? 2. How do caregivers’ networks of care function in the TCRAs? 3. How do caregivers negotiate their rights and responsibilities in the contexts of child fosterage and transnational migration? University of Ghana http://ugspace.ug.edu.gh 10 4. How does transnational migration affect child-fostering norms in Ghana, and how are child fostering norms affecting TCRAs? 1.5 Definition of a caregiver A caregiver in this study is defined as any male or female, kin or non-kin, who a migrant has assigned to provide the material, emotional and social needs of the child/ren that s/he leave(s) behind. It is not uncommon to find more than one person providing care for the child/ren in many Ghanaian families, thus the one I select for the study must live in the same household as the child/ren of the migrant, be the one who receives and decides on the use of the remittances sent for the child/ren and must be the first point of call should the migrant demand accountability on the progress of the child/ren. I refer to these persons as transnational caregivers because they are caregivers of children who have one or more parents abroad and, even more so, because their daily activities find them actively engaging with both local and international actors to carry out their role. They are also able to live ‘here and there’ in real time, something which was not so possible in Ghana a decade and a half ago when modern communication technologies such as the internet and mobile telephony were not easily accessible. 1.6 Organization of the thesis This thesis comprises 9 chapters. Chapter 2 is a review of the relevant literature on transnational families and transnational parenting paying particular attention to the issues of context and transnational care practices. I also introduce transnationalism as the main approach to the analysis and propose a conceptual framework that takes into account transnationalism, child fosterage and transnational care practices. University of Ghana http://ugspace.ug.edu.gh 11 Chapter 3 sets the background to this study. It presents a brief overview of Ghana’s economy and a historical overview of international migration in Ghana. This is followed by a discussion on child care and child fosterage which sets the stage for the discussion of changes that have occurred in child fostering norms in the subsequent chapters. Chapter 4 describes the research design and the study sites for the study. The study employs an ethnographic approach. I discuss the Simultaneous Matched Sample (SMS) methodology and how it contributes to this study. The final section presents Greater Accra and the Ashanti Regions and a justification for why they were selected. Chapter 5 presents and discusses the profile of the caregivers. I highlight three categories of caregivers who present different sides to the studies on caregivers in TCRAs, namely the male caregiver, the stay-behind wife, and the paid caregiver. I conclude the chapter with some issues that emerge from the discussion on caregivers’ characteristics and how they contribute to the existing literature on TCRAs. Chapter 6 describes the pathways that caregivers follow into their care arrangement and their expectations. I also discuss the factors that inform the pathways that they follow. In Chapter 7, I argue that taking care of children of migrants consists of a combination of tangible and intangible activities which I describe as ‘providing care’ and ‘managing care’ respectively. I contend that the former which comprises the tangible activities is actually dependent on the latter. Chapter 8 examines the caregivers' care networks. Here I establish that the primary caregivers of the migrants’ children receive assistance from both local and international actors. I found that the University of Ghana http://ugspace.ug.edu.gh 12 caregivers had small networks contrary to what the literature on child fosterage and some studies on transnational parenting sometimes posit. I attempt to offer some explanations for the small networks. Chapter 9 synthesizes the study findings and relates them to the conceptual framework developed for this study in order to provide fresh insights into the debates on transnational parenting. I also make recommendations for future research. University of Ghana http://ugspace.ug.edu.gh 13 CHAPTER TWO: LITERATURE REVIEW AND CONCEPTUAL FRAMEWORK 2.0 Introduction In this chapter, I introduce transnationalism which serves as a central theoretical approach for this study. This is followed by a review of relevant literature, especially on transnational parenting practices. I focus in particular on how transnationalism has been applied to the study of parenting from afar, and how these studies have incorporated the perspectives of caregivers as an important aspect of transnational parenting. I also look at two broad areas which emerged in this study, namely context and transnational care activities. By the end of the chapter, I establish the gaps in the debate on transnational parenting that my study seeks to address and show how the findings of my study would improve upon existing work. I also present a conceptual framework that I have developed for the analysis in this study. 2.1 Transnationalism Basch, Glick Schiller and Szanton Blanc (1994:7) define transnationalism as “the processes by which immigrants forge and sustain multi-stranded social relations that link together their societies of origin and settlement”.. Transnationalism has been rapidly 'assimilated' by anthropologists, sociologists, political scientists, geographers and other scholars (Smith et al., 1998). However, in spite of the proliferation of studies on transnationalism, what appears to be consistent with the phenomenon since it was introduced and popularised into migration studies by anthropologists Basch, Glick Schiller and Szanton-Blanc is the near consensus that ‘transnationalism’ broadly refers to multiple ties and interactions linking people or institutions across the borders of nation-states, as noted by Mazzucato et al. (2004). The theory emphasizes University of Ghana http://ugspace.ug.edu.gh 14 the increased possibilities for migrants and their families to live transnationally and to adopt transnational identities (Vertovec, 1999; Guarnizo, 2003). Transnational migration may in fact not be new but has certainly reached a ‘certain intensity at global scale by the end of the 20th Century’ (Guarnizo and Smith, 1998: 4). Indeed it is true that 19th and early 20th century migrants also kept intensive transnational ties, but technological revolutions have radically increased the scope for migrants and their families to pursue transnational livelihoods on a more constant, day-to-day basis (Vertovec, 2004; de Haas, 2010a). Transnationalism has been studied both as a perspective (a way of looking at social phenomenon) and as a phenomenon describing social reality (Mazzucato et al., 2004). As a perspective, transnationalism presents a break from the approach that dominated migration studies in the 1960s and 70s. This approach split migrants’ lives into two: they either focused on the activities of the migrants in the receiving country or they examined the effects that migrants have on people back in their home countries (Itzigsohn et al, 1999; Mazzucato, 2008a). This approach portrays migration largely as a unidirectional movement in which migrants settle in well in their receiving countries, a situation which Levitt and Jaworsky (2007) describe as a one- way process of assimilation into a melting pot or a multicultural salad bowl. In most studies using this approach, the migrant is the sole focus of attention. Some migration literature, especially those related to remittances, (such as Page and Adam, 2004; Quartey, 2006; Ratha 2003, 2007; Carling 2008, Mazzucato et al 2008), focus mainly on the flow of resources from the migrant to his home country, with the added feature that most of these studies are conducted in the household of the migrant in the home country.. Such studies make very little mention of the existence of reciprocal and reverse relationships also referred to as reverse remittances between the migrants and actors in their country of origin. University of Ghana http://ugspace.ug.edu.gh 15 Transnationalism, however, recognises the individual migrant as a member of a larger migration community that extends beyond geographical boundaries. In this regard, to some extent transnationalism parallels the New Economics of Labour Migration (NELM) literature, in that they both emphasize placing the behavior of individual migrants in a wider societal context and by considering not the individual, but the family or the household as the most appropriate decision-making unit (De Haas, 2010a). NELM argues that migration is a deliberate attempt by social groups (typically, but not exclusively, households) to spread income risks, to improve their social and economic status and, hence, to overcome local development constraints (Lucas and Bloom, 1985). In this context, migration has increasingly been recognized as one of the main strategies households employ to diversify, secure and potentially improve their livelihoods. In particular, the emergence of NELM marked a departure from neo-classical and actor-oriented approaches towards a household-level based perspective which explains migration from the structural constraints and imperfect markets within which migration decisions are made (De Haas, 2010a). Whereas both approaches move away from an actor-oriented analysis toward a wider unit such as the household, transnationalism extends this further by looking at how the individual migrant actively engages with his new society and the society he leaves behind which includes his household. It emphasizes the linkages and bi-directional exchanges of informal, moral and material support that frequently and simultaneously takes place between the two societies. Therefore transnational migration studies asks questions not just about why migrants remit and to whom, but also about what help people back home offer to enable them to earn the remittances that migrants send back, whether the exchange is mutual, and whether the remittances are used for the purpose for which they were sent (Mazzucato, 2008a). University of Ghana http://ugspace.ug.edu.gh 16 The transnational approach also emphasizes that to varying degrees migrants are simultaneously embedded in the multiple sites and layers of transnational social fields in which they live. These social fields include the states from which the migrants originate and the states in which they have settled (Basch et al, 1994). These social fields are constructed through the daily lives and activities of immigrants covering all aspects of their lives, from their economic opportunities to their political behavior and their individual and group identities (Glick Schiller and Fouron, 2011; Tsuda, 2012). It has been shown that more and more aspects of social life take place across borders even as the political and cultural salience of nation-state boundaries remains clear (Levitt and Jaworsky, 2007). The social, economic and political ties forged by these linkages are so deep and widespread that they fundamentally change the way individuals earn their livelihoods, raise their families, enact religious rituals and express their political interests (Levitt, 2001). As a phenomenon, transnationalism highlights two important characteristics: simultaneity and networks (Mazzucato, 2009). Tsuda (2012) in his study on transnational migration theory and double engagement in sending and receiving countries also highlights transborder connections and simultaneity as the two main components of transnationalism. The fact that people communicate in real time means that actions are taken almost simultaneously in different parts of the world. Enhanced by new technologies, especially cheaper travel and communications, people are able to ‘live here and there’ simultaneously (Vertovec, 1999; Mazzucato, 2008). Transnationalism thus emphasizes the intensity, frequency and regularity with which these simultaneous flows take place (Mazzucato et al, 2004; Guarnizo and Smith, 1998). University of Ghana http://ugspace.ug.edu.gh 17 One aspect of transnational migration in which living ‘here and there’ is very pronounced is transnational child raising arrangements or transnational parenting. Aided by technology, the actors in TCRAs are ever present in each other’s lives. Literature on TCRAs has often presented the case of parents ‘living here and there’, but these have hardly taken into account the fact that those who stay behind especially the caregivers and the children are also able to ‘live here and there”. For instance, Alicea (1997:614) noted that performing care work in a transnational family means having to monitor and meet the physical and emotional needs of individuals in more than one household in more than one country, and to balance one's time and energy between the households. Her focus was largely on migrant parents especially mothers without an equal consideration of the caregivers’ ability to do same. Furthermore, such literature has often largely overlooked how ‘living here and there’ impacts the TCRAs and changes the practice of child-raising in the home country of the migrants. A few studies have focused on caregivers (Moran-Taylor, 2008; Dreby, 2010; Coe, 2011); others have focused on wives that stay behind (Pribilsky, 2004; Menjivar and Agadjanian, 2007; Mckenzie and Menjivar, 2010). Most of these studies cited indicate that transnational migration brings an added responsibility to caregivers in the home country. Stay-behind wives were more concerned about child care and financial decisions. With non-spouse caregivers, this entailed the additional task of sometimes having to raise their own children and run a business. Additionally, studies that focus on caregivers have particularly noted the importance of caregivers to the maintenance of transnational families by allowing the parents to travel and leave the children in their care. According to Dreby (2010), the idea that the children are being well taken care off gave migrants the peace of mind needed to work in the host country and send remittances for their children's upkeep. The existing studies which have focused on caregivers University of Ghana http://ugspace.ug.edu.gh 18 have also highlighted the mediating role caregivers play in re-enforcing the parent-child relationships (Pribilsky, 2004; Mckenzie and Menjivar, 2010; Dreby, 2010). Caregivers have been careful not to usurp the positions of parents and have allowed children to be reunited with their parents even when they thought and felt otherwise. At worst, they tried to negotiate and convince parents that leaving the children with them was a better idea (Moran-Taylor, 2008; Dreby, 2010). The relationship between caregivers and migrant parents is another important issue in transnational parenting. The factors which most enhanced or jeopardized the migrant-caregiver relationship were remittances and overextending the period of the care arrangement (Moran- Taylor, 2008; Dreby, 2010). Dreby (2010) noted however, that caregivers who are grandmothers to the children are able to maintain a cordial relationship with the migrants because they are more tolerant of situations such as delayed remittances than other relatives who take on the role of the caregiver. Caregivers have also expressed powerlessness, especially when the migrant parents are very much involved in their lives. Dreby (2010) observed how caregivers felt when their suggestions concerning the children’s upbringing were ignored by the migrant parent, especially in the case of a family reunification. What is missing in these few studies, however, is how the themes they have explored are shaped by factors such as the existing child-raising norms and practices in the home country. For instance, how is the powerlessness that caregivers experience as noted by Dreby (2010) and Moran-Taylor (2008) influenced by existing child- rearing practices? Both Dreby and Moran-Taylor give a background of the child-raising context in their studies, but they do not link these to the caregivers' experiences. I address this gap and contribute to the growing body of literature on caregivers by examining the experiences of caregivers through the lens of a conceptual framework which incorporates the context of child- University of Ghana http://ugspace.ug.edu.gh 19 raising practices that exist in the home country. In particular, I look at how existing child-raising practices combine with transnationalism to shape the experiences of caregivers. The second characteristic of transnationalism is the networks that link migrants and the people they leave behind. Migrant networks are made up of sets of ties that connect migrants, non-migrants and former migrants within webs of kinship, friendship and shared origin. Networks can be considered as a form of social capital stretched across migrant space, and therefore as a force that facilitates the likelihood of international movement by furnishing information which can lower the costs and risks of migration (Massey et al., 1998: 42-43). The Ghana Transnet Programme (Mazzucato, 2006, 2007; Kabki, 2007; Smith, 2007) has examined how migrants and their networks in both rural and urban sectors in Ghana exchange mutual and reciprocal help across borders. However, in the area of transnational child raising arrangements where migrants leave children behind in the home country, the importance of networks remain largely underexplored. Vertovec (1999) emphasizes the importance of people within networks by focusing on the 'multiple ties and interactions linking people or institutions across the borders of nation- states' (447). Networks reflect the fact that the most relevant unit of analysis is no longer the household, the individual migrant or the family members that live with him abroad. Instead, the unit of analysis comprises all the people that make up the transnational linkages that extend across geographical boundaries. This is considered to be a major contribution to the literature on transnationalism (Mazzucato, 2009), and it is one reason why this study focuses on caregivers, but does so within the context of TCRAs where the relationship with the migrant parent, the children in their care and others who help to raise the children, plays an important role in the analysis. In TCRAs, networks of caregivers comprise international actors, especially the migrant University of Ghana http://ugspace.ug.edu.gh 20 parents. But more importantly, networks are made up of actors in the home country with whom migrants often share some ties of kinship or friendship who help the caregivers with the raising of the children in their care. Transnationalism has been adopted as the theoretical approach for this study primarily for two reasons. First, it views migration as “a process rather than an event” (Levitt and Glick Schiller, 2004: 1012) and conceptualizes it as “a continuous flow of people, goods, money and ideas that transgress national boundaries...” (Mazzucato et al., 2004:1). This allows for the study of multiple relationships that exist among the caregivers, the migrant’s children in their care and the migrant parents as well as the bi-directional flow of exchanges that take place daily. Secondly, it emphasizes the point that movement is not a prerequisite for engaging in a transnational social field (Levitt & Nyberg-Sørensen, 2004). This is because there are people involved in TCRAs who do not cross borders, including migrants who have permanently settled in the host countries and are unlikely to return to their home countries, those who were born in the host country and therefore may never have set foot in their home country and those people which the migrants have left behind in the home country. Most of these different categories of people lay claims to membership in the transnational social field through heroic ancestors and family and blood ties (Glick Schiller and Fouron, 1999). Transnationalism, therefore, does not focus only on the agency of migrant, but emphasizes the agency of those who do not necessarily move but are connected to migrants through the networks of social relations they sustain across national borders. In this study, I focus on the category of people which the migrants leave behind in the home country: the caregivers who raise th eir children. University of Ghana http://ugspace.ug.edu.gh 21 2.2 Transnational families Despite the growing relevance of transnationalism to the study of migration, some areas of the migrants’ lives across borders remained largely unexplored possibly because the literature on migration focused for so long on questions of assimilation and integration in the host societies (Vertovec, 2009). One such area is the family (Mazzucato & Schans, 2008). Up until 2000, discussions on transnationalism paid more attention to the public arena of politics or citizenship at the expense of more domestic and private activities (Appadurai 1996; Cohen, 1997; Brah, 1998). These studies only captured part of the stories of transnational family life. They emphasized the issues of boundaries and space, whereas those related to agency and everyday practice which are even more central to understanding the mobility, future and orientation dynamics of networking within transnational families and transnational ways of life are far less worked out (Bryceson and Vuorela, 2002:4). To fully understand the meanings and implications of transnationalism for ordinary people, we also need to consider activities and relationships within households and families’ (Gardner and Grillo, 2002), despite the great distances and the presence of international borders. Bryceson and Vuorela (2002) were the first to set up transnational families as the central focus of research, and they defined the transnational family as “families that live some or most of the time separated from each other, yet hold together and create something that can be seen as a feeling of collective welfare and unity namely ‘familyhood’ even across national borders” (2002: 3). Through their work and that of many others (see Hondagneu-Sotelo and Avila, 1997; Parrenas, 2001, 2005; Schmalzbauer, 2004; Dreby, 2006, 2010), they have proved that studying transnationalism and families allows everyday practices that are more central to understanding transnational family-making and community-networking to be documented. University of Ghana http://ugspace.ug.edu.gh 22 In the introductory chapter of a special issue on family dynamics in transnational African migration to Europe, Konig and de Regt (2010) indicate that the articles published in the special issue document moments in contemporary transnational Afro-European history whereby women and men, parents and children, grandparents and offspring of all kinds constitute the core of migrants, rather than the patterns that lead to migration. They thus contributed to the efforts at examining the non-economic aspects of transnational migration by focusing more on what is termed ‘transnationalism from below’ (Guarnizo and Smith, 1998), a perspective from which the relationships and daily practices which take place within the transnational social fields cannot be missed. Bryceson and Vuorela express this idea in their statement that “by centre-staging transnational families, [they] are examining the processes of globalization from below, specifically from the point of view of individuals whose lives are largely inscribed by membership of transnationally mobile families” (2002:7). The fact that attention has been shifted onto what can be termed as ‘micro transnationalism’ (Gardner and Grillo, 2002) of the home does not dismiss the essence of the ‘macro transnationalism’ of the state and the economy. In fact, the interrelationship of the two is crucial because the one cannot be fully understood without the other (Gardner and Grillo, 2002; Zentgraf and Chincilla, 2012). Gardener and Grillo (2002) further explain that this interrelationship enables scholars to address the articulation of the global alongside issues of identity, status and culturally specific forms of hierarchy and inequality, as well as to consider how transnational activities are articulated with different forms of power in different locations. Family separation as a result of international migration is not a new phenomenon. However, what has become new in the last decade and half is the number of children who are left behind by parents who face declining economic opportunities at home along with high demand University of Ghana http://ugspace.ug.edu.gh 23 for the labour which these migrants can provide in other countries (Madiannou and Miller, 2012). Increasing female migration and the inability of mothers to travel with their children gives rise to new forms of transnational family life and new definitions of parenting and parent- child relationships. There is, therefore, an increasing phenomenon of 'parenting from afar'. Hondagneu-Sotelo and Avila (1997) call this 'transnational motherhood', while some other studies have referred to it as 'teleparenting' (Casteneda and Buck, 2011) or global parenting (Tanalega, 2002). Recent research on this phenomenon has resulted in a growing body of literature on transnational parenting (Ehrenreich and Hochschild, 2002; Bernard et al., 2005; Parrenas, 2005; Dreby, 2010). This literature covers broad themes on 'parenting from afar', including gender, care arrangements, legislation, class, communication and moral considerations (Carling et al., 2012). This body of literature has also been criticized for projecting to the center of the analysis the parent-child dyad instead of the triad or what Akesson et al. (2012) refer to as the transnational fostering triangle of care. This triangle includes the other people in the home country who take care of the migrant's children. Mazzucato (2013) has also identified some gaps in transnational family research which are now being addressed in emerging studies. One of these is the geographical bias in these studies: the majority of cases are from Latin America, closely followed by South East Asia. Almost no African cases have been examined. Thus, the current study seeks to add to the few existing African cases by looking at the care triads in transnational parenting in Ghana. Another gap in the research consists of perspectives on the caregivers and their roles in transnational parenting. In the subsequent sections, I draw on two broad areas that are significant to the study and analysis of transnational parenting, namely context and transnational care practices. University of Ghana http://ugspace.ug.edu.gh 24 2.3 Context in Transnational Parenting The role of context in the study and analysis of transnational parenting has begun to gain some attention. In this study, context refers to parenting norms and practices in Ghana, namely social parenthood of which child fostering forms a significant part. Mazzucato and Schans (2011) advocate the study of TCRAs within the context of parenting in the country being studied. Previous studies which have given priority to context focused mainly on the host countries in which the migrant parents lived, looking at their work-life conditions and the immigration policies that enhanced or prohibited smooth transnational parenting (Parrenas, 2005; Menjivar, 2012). Most of the early studies on transnational parenting were also criticised for carrying a western notion of parenting which contained the image of a nuclear heteronormative family as a hegemonic feature of society (Parrenas, 2005; Verhoef and Morelli, 2007; Gamburd, 2008). The critics pointed out that if a study is looked at from within the parenting context, one would realize that the costs and benefits of parent-child separation are not fixed and that they are influenced by the micro and macro level contexts within which they occur (Zentgraf and Chinchilla, 2012). In their seminal article on ‘transnational motherhood’, Hondagneu-Sotelo and Avila (1997) argue that perceptions, experiences and meanings of transnational motherhood should be understood as transformations of culture-specific mothering ideologies. Olwig (1999) also makes a similar point that the cultural construction of parenting is highly significant in appreciating, for instance, how children respond to their parents’ migration. This argument has been further corroborated by Coe (2011) and Dankyi (2011), who in separate studies examined the impact of parental migration on children who are raised in a system where child fosterage is widely practiced. They both found that the children they studied in Ghana were not adversely affected by the migration of their parent(s) largely because of the normality of parental University of Ghana http://ugspace.ug.edu.gh 25 separation. Adding to what Mazzucato and Schans (2011) advocated, Zentgraf and Chinchilla (2012) stressed the inclusion of context in what they described as a framework for analysis in transnational family separation. They further explained that a lack of recognition of different familial contexts and traditions often results in separated families being defined as pathological and in transnational parents being blamed for the problems of the children and youth left behind. When the context in which a study is situated is not made explicit or its role in understanding the study is not established, unanswered questions and gaps arise in the analysis. Castaneda and Buck (2011) studied 60 transnational families of immigrants in the United States. The authors observed that the transnational family was in limbo, living in a state of fear and anxiety which causes malaise and emotional stress for individuals who are often far away from the support of the extended family and their community of origin. Their study leaves a number of unresolved issues, one of which is the role of context. Castaneda and Buck (2011) did not clarify the context within which their study was situated. For instance, it must be made clear what the existing normative gender expectations of parents are in the sending country, and who migrates and whose departure is regarded as more traumatic by the children left behind. Other issues that need to be clarified include identifying whether there exists any pre-migration child raising practices, who the caregivers of these children are, and whether migration is new to the sending country. All these unanswered questions make the issue of the local parenting context, a more compelling component in the study of transnational parenting. By drawing attention to these lapses, I am by no means implying that situating the work within a context absorbs all the downsides that separation across borders can entail for those involved. To the contrary, my argument is that including context provides a more nuanced picture of the situation which parents and children and other people in transnational families experience. University of Ghana http://ugspace.ug.edu.gh 26 The obverse of the above argument is also valid because in cases where the child-raising practices in the country being studied are included in the analysis, it sheds light on why parent- child separation in some contexts is more traumatic than others. For instance, in their study on transnational, multi-local families among Latin Americans in Canada, Bernard et al. (2005) found that separation had a huge emotional toll on the mothers. The authors noted that apart from the social disapproval and stigmatization that mothers faced from members of their extended family who may have initially supported the idea to migrate, they also live in self-guilt, knowing all the time that they have fallen short of their duties as mothers. This is because in most Latin American cultures, mothers have the primary responsibility of attending to many areas related to their children’s development, a responsibility which in this context cannot be reassigned to others (Hondagneu-Sotelo and Avila, 1997). Glasgow and Gouse-Sheese (1995) found that a recurring theme in the study of mother-child separation among Caribbean children in Canada was a sense of rejection and abandonment by their mothers. Parrenas (2001) made a similar observation in her studies on Filipina mothers where there is an ideological component to the representation of “good mothering’ which exacerbates the already difficult situation mothers must face during separation. The painful situation induced by separation in the Philippines is intensified by the failure of mothers to meet the gender-based nurturing expectations of their children and their self-imposed expectations to follow culturally and ideologically-inscribed duties of the family in the homestead (Parrenas, 2001: 387). In line with the importance of context to the study and analysis of transnational parenting, this study is placed within the context of child fosterage in Ghana. Child fosterage is a common parenting practice in Ghana that involves the delegation of parental responsibilities to a trusted adult in the family, usually a close relation (Nukunya, 2003). Child fosterage is considered University of Ghana http://ugspace.ug.edu.gh 27 important in the analysis of caregivers' experiences because it largely informs their decision to take care of the children. Child fosterage also determines how much parental control can be exercised on children by caregivers because it always gives more control to the foster parents. For the purposes of the conceptual framework for this study, I focus on one of the characteristics of child fosterage: power relations. The institution of fosterage involves power relations engendered by the dependency of one party, the biological parent on another, the foster parent to meet the care needs of the child (Tronto, 1998). This dependency stems from several factors including the availability of resources to one person over another and the position one occupies in society. In the traditional fostering system, usually poorer rural families sent their children to more resourceful ones and the latter wielded more decision making power over the children. In TCRAs, the situation is almost a reverse where migrants’ parents are usually resourceful enough to provide for their children. This has implications for power relations in the TCRAs and also for how caregivers experience their role and thus the need to incorporate it into the analysis. In chapter 3, I discuss in detail the practice of child fosterage. 2. 4 Transnational Care Activities: Remittances and Communication Scholars such as Horst (2006), Banfi and Boccagni (2010) and Graham and Jordan (2011) who study transnational parenting activities are often impressed by migrant parents’ creative efforts to keep family members connected to and involved with each others’ lives, even in the case of prolonged separation. Through what some studies have termed transnational care practices (Graham and Jordan, 2011), rituals of transnational households (Horst, 2006) and a tool kit of transnational family relations (Banfi and Boccagni, 2010), migrant parents are able to keep close ties with the children they leave behind. These practices broadly encompass remittances to University of Ghana http://ugspace.ug.edu.gh 28 the home country, communication with those left behind, and return visits. These are said to influence both the quality of care and the well being of caregivers in the home country. Whereas transnational care practices help parents deal with separation from their children, they have also helped caregivers manage the burden of raising the children of migrants. In this study I focus on remittances and communication because migrants used them more frequently and therefore they have significant impact on the activities of those left behind. 2.4.1 Remittances and transnational parenting Migrants’ remittances are largely considered as one of the essential elements of transnationalism. According to Guarnizo (2003), remittances have become the most visible evidence and measuring stick for the ties connecting migrants with the society of origin. Until the pursuit of other elements such as social, cultural, and political consequences of transnationalism, the economic aspect of migration measured primarily in terms of migrant remittances and economic opportunities were the order of the day. This is compatible with the assertions of many studies that economic opportunities strongly motivate individuals to migrate. Migration has been said to be beneficial, especially in developing countries, as the volume of remittances flowing into them are estimated to exceed Official Direct Investments (Ratha, 2008). At the household level, migration is also viewed as helpful both as a livelihood enhancement strategy and a means of cushioning vulnerabilities in times of economic hardships (Siddiqui, 2003; Schmalzbauer, 2005; Quartey, 2006; Awumbila and Ardayfio-Schandorf, 2008). Castaneda (2009) describes remittances as representing the sweat, sacrifice, and loneliness that migrants endure in order to provide their families with basic goods and a humble increase in living standards. University of Ghana http://ugspace.ug.edu.gh 29 Migrant parents hope that the remittances they send, often at great sacrifice, will mitigate the potentially negative effects of their absence and open up opportunities for their children, such as private schooling or university studies, which might not be possible if they had remained in their home countries. In his study among Ecuadorian transmigrants in New York City and the Ecuadorian Andes, Pribilsky (2004) found that increased reliance on remittances in migrant households triggered a move away from agricultural production, children were relieved of their obligations to the family farm economy. Rather than waking up early to haul firewood or to help with the harvest, children of migrants increasingly rush to catch school buses, spending much of their day away from the household. Women (wives who remained in the home country) who received enough money to build a new home, which many respondents interpreted as the ultimate indicator of their migrant husband’s or son’s success, generally viewed migration as worth the hardships and socio-psychological costs of separation. They viewed remittances as a way to improve their families’ diet, purchase new clothes and pay for their child’s schooling (Pribilsky, 2004). The role of remittances in cementing already existing relationships especially among separated couples cannot be overemphasized. Wives who remain in the home country saw the regular flow of remittances as an assurance that their marriages were still strong (Pribilsky, 2004; Coe, 2011). Mothers, children and close relatives of migrants had similar feelings about receiving remittances. For non-spouse caregivers, regular and sufficient remittances also helped lessen the burden of child care. However, receiving remittances intended to pay for educating the children left behind, meant that the caregivers must by all means ensure that these children were in school regularly and excelling. This caused a shortage of domestic labour, unlike the traditional child fosterage situation in Ghana where the children sometimes provide commercial labour by joining the foster parent in working on the farm or at the market University of Ghana http://ugspace.ug.edu.gh 30 as well as undertaking domestic chores. Caregivers thus benefitted less from the children when parents sent sufficient remittances to cover the living expenses of their children. Migrants, especially mothers, have also sometimes raised concerns about misuse of remittances. Some of the respondents in the study of Akesson et al (2012) describe it as a ‘dilution of remittances’. Their respondents concerns included not trusting the caregivers to use the money for the intended purposes, and the problem of remittances attracting more people into the household of the migrant (as caregivers take on the responsibility of more children). Thus, sending remittances appears to breed some level of mistrust between migrants and the caregivers. What is missing in these analyses of how the remittances are used is the voice of the remittance ‘managers’, as Pribilsky (2004) describes the caregivers who receive them. At best, some studies capture the caregivers’ concerns about inadequate remittances (Moran-Taylor, 2008; Dreby, 2010; Mckenzie and Menjivar, 2010). In this study, I add to the existing literature on the importance of remittances in the TCRAs by examining what goes into ‘managing remittances’, such as issues related to delayed or insufficient remittances, mistrust, and the expectations of migrants and outsiders with regard to their remittances. I do this by focusing on the caregiver’s side of the story. 2.4.2 Communication and Transnational Parenting The dramatic change that revolutionized the way in which families maintain long distance communication was the emergence of a plethora of internet- and mobile phone-based platforms such as email, Instant Messaging (IM), Social Networking Sites (SNS) and webcam via Voice Over Internet Protocol (VOIP) (Madiannou and Miller, 2012:1). Madiannou and Miller noted that these new media have engendered the emergence of a new communicative environment which they call Polymedia (ibid). Polymedia allows for emotional and moral University of Ghana http://ugspace.ug.edu.gh 31 support, for the provision of practical support, for decision making and for the exchange and sharing of ideas to happen in real time and usually at the time it is most needed. Vertovec (2004) has also argued that nothing has facilitated global linkage more than the boom in ordinary, cheap international telephone calls. Indeed, along with remittances and packages, women interpreted regular telephone communication as the strongest reassurance that their families would remain intact through their period of separation. For stay-behind wives, communication meant more than just receiving a call with information about their loved ones, which the women longed for. It was also an indicator that their husbands still loved them (Mckenzie and Menjivar, 2010; Coe 2011). It has been argued that telephone and other forms of high-tech communication between migrants and non-migrant relatives ameliorates the effects of migration, contributing to building ties across borders ( Glick Schiller et al, 1995; Mahler, 2001). Communication between transnational families has progressed significantly. In her study in a rural town in Jamaica, Horst (2006) presented the metamorphosing stages of communication that transnational families have passed through, and how each stage improved the relationship within the TCRAs. Madianou and Miller (2012) made similar observations in their three-year study with Filipina mothers who left their children in the Philippines to work in the United Kingdom. The mothers shared their experiences from the era of writing letters, sending cassette tapes and making phone calls to public phones in the Philippines to this era of new technologies, especially mobile phones, where they can talk to their children at any time or send several instant messages in a day. The advent of the mobile phone added a level of spontaneity to communication between members of a family that lived apart. For instance, formerly the phone calls had to be scheduled for fixed times and there was no way one could tell ahead of time if the other was not going to be University of Ghana http://ugspace.ug.edu.gh 32 able to call. All of that had now changed. In Banfi and Boccagni's study (2011), one migrant mother was quoted as saying 'I brought my boys up by the phone', explaining that she used to call her children ten times a day. They found a similar trend in the progression of communication from slower media to the era of smart phones and social media networks. Similar changes could be noted in the Ghanaian context. In one of her several studies of Ghanaians in Toronto, Manuh (2001) found that besides sending letters and postcards, the main medium of communication was sending and receiving recorded messages on audio cassette tapes. With the advent of new media technologies, now, however, the children in Ghana who Dankyi (2011) and Coe (2011) studied communicated with their parents mainly via mobile telephone and emails. The use of telephones, especially mobile phones, has become very popular and affordable in Ghana. The ‘blessings’ of new technologies are enormous. Apart from the convenience and continuous access they offer, they are also of great help in emergency situations, such as when extra money is needed for medical attention. Additionally, migrants are able to send information ahead of time where they are not able to send remittances. Parents have never been closer to their children and the collaboration between parents and their children’s caregivers could not be any better. This notwithstanding, some scholars have cautioned against exaggerated good effects attributed to new technologies in enhancing the lives of transnational families (Pribilsky, 2004; Banfi and Boccagni, 2011; Horst, 2011). Concerns raised include the fact that communication is not available to all and in cases where they are, power relations still come into play since it is only those who can afford to initiate or make such calls that do so. It is usually more expensive to call from the home country to the host country and considering the already scanty remittances these caregivers have to live on, they are hardly able to make telephone calls to the migrants. University of Ghana http://ugspace.ug.edu.gh 33 Thus, caregiver calls to migrants are limited to very urgent situations. The caregivers therefore depend on the migrants to keep communication, especially by telephone, active and alive. Furthermore, communication means more than simply receiving telephone calls and messages from loved ones. There were often issues with the content of the communication. Both the migrants and the people they left behind were often selective about what to tell each other because they were worried that some information could be detrimental to the recipient. (McKenzie and Menjivar, 2010). Due to the constant and interactive nature of communication not just with migrants and their children but also with other relatives, or even neighbours who were willing to volunteer information, the concealment of problems and other unpleasant incidents that occur within the TCRAs is a problem (Madiannou and Miller, 2012). The wide usage of the telephone and the access it gives people back home to the migrants makes it difficult for members, especially caregivers in the transitional family, to control the content of information that circulates in their transnational space. As a result, gossip has been one of the burdens that the upsurge in communication has brought to families that are separated (Dreby, 2009). Some studies have looked at the communication between migrants and caregivers, however, these are mainly studies on wives who stay behind, and the communication focuses more on the conjugal relationships with their husbands (Pribilsky, 2004; Mckenzie and Menjivar, 2010; Menjivar, 2012). This study explores how caregivers experience communication within and outside the TCRAs. In particular, it explores how they manage the blessings and burdens of communication in their daily experiences. 2.5 Conceptual framework The conceptual framework comprises three main components namely transnationalism, child fosterage and transnational care practices, which are presented in Figure 2.1. For each University of Ghana http://ugspace.ug.edu.gh 34 component, I highlight the characteristic(s) that will be emphasized in the analysis. I use Basch, Glick Schiller and Blanc’s (1994)_definition of transnationalism which allows for an appreciation of activities and events in multiple sites. Under transnationalism, I emphasize two main characteristics, namely simultaneity and networks. Simultaneity emphasizes the ability of members in the transnational social space to live ‘here’ and ‘there’ at the same time. I include networks here to extend network analysis crucial to transnationalism to the networks that caregivers tap into perform their care roles. The next component is the institution of child fosterage which represents the local parenting context within which the study is situated. I have argued earlier that incorporating the parenting context into the analysis of TCRAs reduces the incidence of describing TCRAs as pathological or dysfunctional. Under child fosterage, I emphasize power relations between the migrant parents and caregivers because it is the aspect of fosterage that is likely to be mostly affected by transnational migration. The third component is transnational care practices. I include these in the framework for the many benefits they present to TCRAs. The individual components on their own have shaped the experiences of the actors especially the migrant parents and children. Most of the studies on transnational parenthood reviewed in this chapter have looked at the subject through one or two of the concepts mentioned, thus resulting in some gaps in our understanding, some of which will be addressed by this study. Thus, I focus on the inter-connections among the components (transnationalism, child fosterage and transnational care activities) to enable me to adequately capture the experiences of the caregivers in TCRAs. I argue that examining the experiences of caregivers through the lens of the three major components provided will reduce the incidence of unresolved debates and unanswered questions on the subject. It is the combination of their interconnections and how each shapes the experiences of caregivers that will provide new insights to the study of TCRAs. University of Ghana http://ugspace.ug.edu.gh 35 Figure 2.1 Conceptual Framework 2.6 Summary In this chapter, I have examined the broad theory of transnationalism, and how various scholars have applied it to the study of transnational parenthood and transnational child-raising arrangements (TCRAs). I have emphasized that TCRAs involve the maintenance of relationships among three key persons (migrant parent, child and caregiver/s) in two or more sites. Transnationalism Simultaneity Networks Context -child fosterage Power relations Transnational Care Practices Remittances Communication Experiences of Transnational Caregivers University of Ghana http://ugspace.ug.edu.gh 36 Importantly, it is the process of transnational migration that provides the context for these relationships. I have also highlighted the importance of situating a study of this nature in the appropriate context, which, for the purposes of this study, I have identified as the institution of child fosterage and have highlighted one of its key features to facilitate understanding caregivers' experiences in TCRAs. I also highlight some of the key features that hold the relationships in TCRAs together, namely transnational care activities. The existing literature has looked primarily at the advantages of these activities in TCRAs. I take this a step further and examine some of the challenges these activities pose for caregivers. I highlighted two main care practices, viz remittances and communication, because they are most crucial to the understanding of caregivers’ experiences and their capacity to engage in care giving activities. I have also provided a conceptual framework, comprising three major components namely, transnationalism, child fosterage and transnational care practices, all of which provide insight into the various factors that shapes the experiences of caregivers in transnational child raising arrangements. University of Ghana http://ugspace.ug.edu.gh 37 CHAPTER THREE: THE CONTEXT OF THE STUDY 3.1 Overview of International Migration in Ghana 3.1.1 Historical overview Anarfi et al. (2003) have identified four distinct phases of international migration in Ghana. There was the period of minimal emigration from pre-colonial times to the late 1960s, a period when Ghana was the destination of choice for many migrants, especially from West African nations. Ghana’s economy was comparatively strong and therefore it attracted a considerable number of political activists, scholars and foreigners. Additionally, migrant labourers from the neighbouring West African countries came to Ghana in large numbers to work in the mines, cocoa plantations and the expanding civil service and urban economic sectors (Adepoju, 2005: 26-8; Anarfi et al., 2003). International movement from Ghana involved a small number of people, most of whom were students and professionals who travelled mainly to the United Kingdom and other English speaking countries to pursue further education and training. Other Ghanaians, mostly from the fishing communities were involved in trading activities to neighboring countries such as Benin and Ivory Coast. The first phase was followed by the period of initial international migration among Ghanaians which accelerated as a result of the economic hardship that the country experienced in the 1970s. During the 1970s, Ghana experienced a period of economic decline and political instability. The deteriorating economic situation and the falling standard of living compelled many Ghanaians to seek greener pastures abroad (Tonah, 2007). The vast majority of Ghanaian migrants left for neighbouring West African countries, in particular Nigeria and Cote d’Ivoire. Most of these were unskilled, but there were also large numbers of skilled persons (masons, carpenters, mechanics, drivers) and professionals (lecturers, teachers, nurses, accountants, University of Ghana http://ugspace.ug.edu.gh 38 bankers) (Peil, 1995). The skilled migrants were perceived as ‘useful migrants’ who had come to assist in building the economies of the host countries (Tonah, 2007). While many West African countries needed skilled and unskilled labourers as well as professionals, European countries-- primarily the United Kingdom, Germany, and the Netherlands--needed cheap labourers which could not be obtained from the indigenous population for their rapidly expanding economies (ibid). The third was the phase of large-scale emigration between 1980 and 1990. By the 1980s, migration had become a coping strategy for individuals and families. The period witnessed large- scale emigration of Ghanaians seeking unskilled, semi-skilled and highly skilled jobs. During this period, most Ghanaians migrated to Côte d’Ivoire and Nigeria. Between 1974 and 1981, about 2 million Ghanaians left for these countries, with the highly skilled comprising a significant proportion of these (Anarfi et al. 2003). The fourth stage constitutes a period of intensification and diasporization of Ghanaians from the 1990s. The 1990s witnessed an increased emigration to Europe and North America. Nevertheless, migration to West Africa and other African countries remained the most important in terms of numbers. This could also be ascribed to the decline of Nigeria as a migration pole. Ghanaian emigrants are now well dispersed across the globe. 3.1.2 Estimates of Ghanaians abroad and destination The available estimates of Ghanaians working abroad are imprecise, largely because they have not been systematically documented (Peil, 1995; DRC 2006; IOM, 2009). There are a variety of reasons to account for this. In the first place, Ghanaians travel through informal channels, which make it difficult to document them. Smith (2007) argues that another reason for University of Ghana http://ugspace.ug.edu.gh 39 the problem has to do with the way organizations charged with collecting such information, such as the Ghana Immigration Service, classify and count who is departing the country. For these reasons, scholars have given different estimates of Ghanaians abroad. According to the Housing and Population Census that took place in 2010, there are 250,623 emigrants, comprising about 1 per cent of the country’s total population (GSS, 2010). This figure stands in sharp contrast to that provided by previous studies which suggested that the number of Ghanaian emigrants was between 1 to 2 million (Twum Baah 2005; Orozco, Bump and Fedewa 2005; Van der Wiel 2005; Kabki 2007; IOM 2009).. In terms of destination, it was estimated in 2006 that 189,461 Ghanaians resided in OECD countries, less than one percent of the estimated total population in Ghana (OECD Migration Database, 2006). In 2007, 689,431 Ghanaians--more than 70 percent of all Ghanaian emigrants--lived in ECOWAS countries (DRC, 2007). The principal destination of Ghanaian emigrants remains Côte d’Ivoire, with 305,648 Ghanaian residents, followed by Nigeria and Burkina Faso with 125,052 and 93,320 Ghanaian residents, respectively. Within OECD countries, the US and UK host the majority of Ghanaian emigrants, with 69,995 and 57,172 emigrants respectively, more than 67 per cent of the total (OECD Migration Database, 2006). However, according to the Housing and Population Census report (GSS 2010), Europe overall had the highest proportion with 38 percent, followed by Africa (36 percent) and the Americas (24 percent). Of the African population, ECOWAS countries accounted for 25 percent. The Housing and Population Census in 2010 (GSS 2012) found that almost a half of the Ghanaian emigrant populations came from the Ashanti and Greater Accra Regions, which is the reason these two were selected as the sites for this study. University of Ghana http://ugspace.ug.edu.gh 40 Recent Ghanaian emigrants work in a variety of occupational sectors. The health sector absorbs the largest number of Ghanaian who settled in OECD countries (19 per cent), followed by the manufacturing sector (18 per cent) (IOM, 2009). In the health sector, women comprise the majority, accounting for 13 per cent of the total number of emigrants in this sector, compared to 6 per cent of men (OECD, 2008). Independent female emigrants are another significant feature of international migration in Ghana. In the year 2000, the Ghanaian males in the United States slightly outnumbered their female counterparts at 56 and 44 per cent respectively (Bump, 2006). According to the latest housing and population census, more than 30 percent of Ghanaian emigrants abroad were women. Adepoju (2010) attributes the increase in the number of Ghanaian women who emigrate to the availability of formal education for African women and the corresponding job opportunities this offers them in the global market. That is not to say that these women always found jobs that matched their skills and competence. Most of these migrants, both male and female, settle for jobs in the host countries that are below their qualifications, which lead to what has been termed the “down skilling” of these migrants (Pitman et al, 2007). 3.1.3 Ghanaian Migrants’ Remittances Migrant remittances form an important source of development finance (Ratha, 2007). Some scholars assert that there has been a remarkable renaissance in interest in migration and development both in policy and academic research and the new interest is undoubtedly triggered by the increase in remittance flows to developing countries (Ratha, 2003; De Haas, 2010b). Migrant remittances from abroad especially are an important source of foreign exchange, whose magnitude exceeded the amount of Official Development Assistance (ODA) in Ghana (Quartey, University of Ghana http://ugspace.ug.edu.gh 41 2006). Governments in migrant sending countries have been said to put renewed hopes on transnationally oriented migrants and ‘Diasporans’ as potential investors and actors of development after decades of concern over brain drain (De Haas 2010b). According to Mazzucato et al (2008), conservative estimates by the Bank of Ghana show that migrant remittances received from abroad constitute approximately 13 per cent of Ghana’s GDP. Remittances can be in cash and kind such as food, goods and what Levitt (1998) calls social remittances which encompass ideas, behaviours, identities and social capital that flows from receiving countries to sending countries and vice versa. However, most studies have only paid attention to the monetary transfers that migrants send to their homes or communities. This is probably because it is easier to quantify and are mostly sent through formal and accessible channels. Remittances are also both internal and international; however, the latter has received more attention both in policy and research. Mazzucato et al (2008) in their study on remittances in Ghana observed and cautioned against a growing neglect of internal remittances in research. They indicated that internal remittances especially from the urban centres to the rural areas formed a significant part of the income of recipient households. Most studies on remittances whether it is internal or international have noted that households of the senders of remittances are the highest beneficiaries. (Kabki et al, 2004; Quartey, 2006; Mazzucato et al, 2008; Adaawen and Owusu, 2013). Others include their extended kin and sometimes members in their communities. Remittances are sent to serve a wide array of purposes. These include household consumption, building projects, business investments and a buffer in times of economic shock. Households and individuals with relatives abroad count on these migrants as social capital, a form of social security especially in times of crisis. Remittances as a form of social security are University of Ghana http://ugspace.ug.edu.gh 42 even more vital in many developing countries where there are no formal social security or welfare systems. Kabki (2007) made this observation in a qualitative study in some rural areas in one of the highest out-migration areas in Ghana. She observed that mass migration from Ghana has made it common for rural dwellers to have migrants as social network members and these migrants can play an important role in the social security provision of rural dwellers. Migrants abroad have also been noted to use remittances as part of activities to keep ties with those back at home (Guarnizo, 2003). Through sending of remittances, migrants not only assist their families and friends back home with personal needs but also use this to cement their relationship with them. Smith (2007) provides insights into how remittances are allocated to diverse economic activities in the urban economy of Ghana. Migrants invested in businesses, education of family members and were also regular sponsors of funerals in the event of the demise of a family member. His work indicated how these activities cemented transnational relationships between migrants and their folks in urban centres. Particularly, remittances are an expression of strong transnational social bonds and of the wish to improve the lives of those left behind. In cultures where meeting a child’s needs materially encapsulates one’s love for the child, remittances help to make the separation less painful for both parents and children. Parents see themselves as fulfilling their obligations while children feel loved since their basic needs are met through their parents’ remittances (Wong, 2006; Coe, 2011; Dankyi, 2011; Poeze and Mazzucato, 2013). 3.1.4 Transnational Activities of Ghanaians Abroad Another important feature of Ghana’s migration landscape is the double engagements of Ghanaians living abroad (Ter Haar, 2005; Mazzucato, 2005, 2008a, 2009b; Tonah, 2007). Tonah (2007) observed that the newer generation of Ghanaian emigrants in Europe were not as fortunate as the previous generation who were considered ‘useful migrants’. The former are University of Ghana http://ugspace.ug.edu.gh 43 mainly young, unskilled and adventurous persons looking for prospects of a better life elsewhere. Those that are trained professionals are often under employed. Most of the recent migrants especially the unskilled belong to the lowest echelons of European society and are forced to survive by doing menial jobs. They face enormous social discrimination and exclusion and a generally hostile host population and government. Tonah (2007) further argues that increasingly, the hostile environment, racial discrimination and social discrimination under which many migrant Ghanaians abroad live compel them to remain transnational. Ghanaian migrants engage in activities that help them build and maintain close social, cultural and religious networks in the countries where they live, while at the same time continuing to cultivate links with their relatives, friends and institutions back in Ghana. The creation of a sizable array of different social groups, clubs, networks and associations in both the host country and Ghana which function to bring Ghanaians with common interests together attest to this. The proliferation of Ghanaian churches abroad is also an indication of these emigrants' desperate need for staying connected both in the host and home country. Among other things, the activities these churches and associations offer help to foster unity among Ghanaian migrants as well as help them to establish good rapport with the host country and maintain ties with the home country. The most common form of maintaining links with Ghana by the various associations abroad include adopting deprived groups, churches and benevolent institutions at home which they regularly support by sending funds, equipment and experts (Owusu, 2000; Tonah 2007) from abroad. Most of the philanthropic activities of Ghanaian associations abroad are concentrated in the health and education sectors and focus on sending help to deprived communities, government institutions, churches and nongovernmental organizations in Ghana. University of Ghana http://ugspace.ug.edu.gh 44 On an individual level, migrants maintain ties with relatives in Ghana by sending remittances for both domestic and commercial purposes, communicating, making return visits and staying involved in social activities such as funerals and festivals (Mazzucato, 2006; Kabki, 2006; Smith, 2007; Tonah, 2007). 3.2 Care and Care Giving in the Ghanaian Social Context Fisher and Tronto (1990) have defined care as a species of activity that includes everything we do to maintain, continue and repair our ‘world’ so that we can live in it as well as possible. According to Waerness (2004), care is both particular and universal. Waerness explains that all people need care even if the need requirement varies in both cultural and biological dimensions. Tronto (1998) has echoed the claim of the universality of care by arguing that the way people care for each other is one of the things that make people human. This view can be summed up in the philosophical concept originating from South Africa, ‘umuntu ngumuntu nga bantu (ubuntu)’, meaning 'A person is a person because of other people'. This concept expresses a strong sense of community where people co-exist in a mutually supportive life style. These relationships are couched within the norms of balanced reciprocity (Waerness, 2004; Coe, 2011). Care giving is an activity that encompasses both instrumental tasks and affective relations. It demands that the caregiver takes some actions when the care needs of others are identified. Wherever care is present, there is always the element of concern. Care practices are engaged in out of a concern for the recipient of the care. Although care is very crucial to human existence and has a long history in many cultures, it has been largely ignored both in research and policy. This neglect appears to be the result of care being understood primarily as a family responsibility, with the tasks routinely falling to women for whom they are viewed as natural, University of Ghana http://ugspace.ug.edu.gh 45 taken-for-granted acts (Graham, 1983) and secondly of care being associated with the quintessentially gendered role of mothering (England, 2005). According to Tronto (1998), the main assumption has been that care is mainly a job for women and it is also menial. Care seems to have been subsumed under the general topic of social support, with the personal devotion and duty involved being implicitly attributed to gender, operating through the primacy of kinship and marriage. Under these conditions, the fundamental importance of care for social life is easily ignored (Fine, 2005). Direct caregiving has always been a job assigned to the lowest groups in a hierarchy: women, slaves and servants. The direct and specific care for children as well as the sick and the elderly has been delegated almost exclusively to women. Tronto challenges the notion that care is usually left in the hands of those who have the least power in society when she posits that care involves power relations wherein the caregiver possesses some skill, ability, resources and/or knowledge that the care receiver does not (Tronto, 1998). Tronto’s claim is supported by England's (2005) ‘public good’ perspective on care. England argues that care work, whether paid or unpaid, often includes investments in the capabilities of the recipients. The care of children for instance imbues cognitive skills in them which can increase their earning potential and, more generally, helps them develop skills, values and habits which will benefit them and others in the future (England 2005). The implication is that care cannot be left in the hands of persons who lack the skill and resources to provide care. To a considerable extent, this perspective is reflected in the practice of child fosterage in Ghana. In child fosterage, a child is placed with ‘parents’ who have more material and sometimes non-material resources than the child's biological parents to raise the child into an independent and responsible adult, capable of returning the investments made in him by serving the wider community. University of Ghana http://ugspace.ug.edu.gh 46 Caregivers are said to be motivated by altruism which provides some intrinsic reward for them and which also makes them prisoners of altruism when the conditions under which they are providing care are poor (England, 2005). The emotional bonds caregivers develop with those they care for puts them in a vulnerable position. For instance, it can discourage them from demanding changes in working conditions because their demands might have adverse effects on those under their care. Whereas caregivers in the Ghanaian context can be said to be motivated by altruism to some extent, especially for those who take care of children, one finds that in addition to altruism is also the expectation of intergenerational reciprocity where care for a child is linked with the child taking care of the adult in their old age (Coe, 2011). Altruism is the immediate impression created by caregivers and foster parents when asked why they decided to take care of children. They also usually say that they are performing their God-given task in helping the children and derive the utmost fulfillment from this. In a study on caregivers in a small town in Ghana, Coe (2011) noted that the caregivers said that they did not expect anything in return for looking after the children in their care. There is an adage in Twi1 which says ‘obi nnim nea obehwe obi daakye', meaning that nobody knows who will take care of another in the future. The adage is used mostly in the context of child care as an admonishment of parents or other caregivers to not neglect the responsibility of taking care of a child. Another popular Ghanaian adage, that is expressed in almost every language, which reflects intergenerational reciprocity, is one which suggests that if your mother or father or someone else looks after you while your teeth are coming in, when the time comes when their teeth are falling out, you should look after them. In spite of the assurance 1 The Twi language is spoken by the Ashanti, largest sub grouping in the Akan ethnic group in Ghana A majority of my respondents belong to this tribe and most of the interviews were held in Twi . University of Ghana http://ugspace.ug.edu.gh 47 derived from this and several other adages on the need to care for the elderly, the caregivers were still careful not to rely completely on the children for care in their old age. Their skeptism can be explained by what Aboderin (2004) observed to be the decline in willingness and ability to support the elderly in her study on material support for the elderly among three generations of Akans in Ghana. Caregivers’ actions cannot be entirely motivated by altruism, however, they prefer to keep a neutral position with respect to their expectations. Caregiving is sometimes defined as a face- to-face activity. Cancian and Oliker (2000:2) define caring as a combination of feelings and actions that 'provide responsively for an individual's personal needs or well-being, in a face-to-face relationship'. Folbre (1995) defined the labour involved in caring as work that provides services based on sustained personal (usually face-to-face) interaction, motivated (at least in part) by concern for the recipient's welfare. These definitions are made problematic by the care performed in TCRAs where parents are said to care across borders (Hondagneu Sotelo and Avila, 1997; Parrenas, 2005; Dreby, 2006, 2010). The issue has become even more problematic in this era of improved information technology where parents and children in TCRAs are able to ‘live here and there’ simultaneously. Although the above discussion captures caregiving in general, in this study, care is restricted to the domestic care of children, that is, care that takes place within a household and does not require formal or scientific training to be performed. Care for children in the domestic sphere is usually performed by kin and is most often work that is delegated to women. Within a kinship group, members are responsible for caring for each other. In Ghana, as in many African countries, older people take care of the children, and it is expected that the children will reciprocate this gesture when they become old and successful. It is therefore important for parents (biological or social) to show a lot of concern and commitment to building the University of Ghana http://ugspace.ug.edu.gh 48 capabilities of the children they care for to earn higher wages in order to ensure their security in old age. This study focuses on the care of children who have one or more parent residing and working abroad and is situated within the context of child fosterage in Ghana. To better appreciate the study, I provide a background into parenting and care for children in families in Ghana. I focus on social parenthood and child fosterage. 3.3 Parenting in the Ghanaian Family While it is difficult to come up with a universally accepted definition of parenthood, anthropologists have traditionally defined it as being concerned with fulfilling the tasks involved in bearing and rearing children and helping them to become effective adults (Goody 1982:7). Goody further adds to these tasks the training, sponsorship and endowment of civil birth status. Ghanaian societies view the task of bearing a child as the sole responsibility of the biological parent, but the act of rearing, training, sponsorship and endowment may be overseen by many. In other words, beyond the birth of a child, the roles pertaining to the welfare of a child can be socially organized. The division of labour in the home permeates almost all facets of family life. Roles are distributed according to gender and age. In the Ghanaian home, fathers are required to provide the material needs of the children as well as play the role of a disciplinarian. According to Nukunya (2003), in most patrilineal societies, the fathers are more concerned with economic provisions and disciplining to ensure children are raised in a manner commensurate with social norms. In matrilineal societies such as the Asante, fathers are not necessarily the ones who are doing the disciplining: a mother’s brothers may also take over this function (Clark,1999; Nukunya, 2003). University of Ghana http://ugspace.ug.edu.gh 49 Mothers are expected to care for and nurture their children in the home. This may sound as though mothers are confined to the domestic sphere and invest more socially and emotionally in their household and children. This is, however, not always the case. Many Ghanaian women wield the hoe (farming) or the tray (commerce). They therefore also contribute economically to the home. According to Clark (1999), in what she calls the urban Asante ideology, the Asante express their maternal devotion mostly through tireless work to meet their children’s financial needs rather than by staying home with them. To the Asante mother, the amount of time spent on caring for a child at home could have been spent working to meet the financial needs of the children. They thus combine child care either by carrying the children along or arranging for others to take care of them while they are away on the farm or market. Mothers who work delegate the day-to-day needs of their children to older children, or, in the case of young mothers, women who are too old to go to work. Fayorsey (1993) observed this in her study on mothers and their matrikin among the Ga of Central Accra, where the economically active women go to work while the older ones take care of children who are left at home. These social mothers provide care for the children in exchange for gifts, in particular cash, because being unable to go to work means that they need another source of income to take care of their needs. In many West African countries, child raising has for many years and to a large extent been a cooperative enterprise drawing upon resources and help from kin, in laws, fathers and others (Oppong, 2012). The biological mother usually enjoys wide spread support especially from her relatives. This yields very large benefits in terms of the survival and well being of the child as well as enabling the mother to speed up her reproductive rate and also maintain her subsistence, economic and productive activities required for material success (ibid). This practice is known as social parenthood. (Goody, 1982; Oppong, 2012). Social parenthood is an essential University of Ghana http://ugspace.ug.edu.gh 50 part of the Ghanaian family system. Both biological and social parents are required to provide the care, maintenance and protection that a child needs before reaching adulthood. Social parents include both kin and non-kin alike who extend to the child the same rights and responsibilities they would to their own biological children. Social parenthood also endorses that every adult of reasonable age can discipline the child of another if the child is caught misbehaving. Disciplining the child may involve spanking, insults, offering advice and withholding privileges (Nukunya, 2003; Twum-Danso, 2009). In return for the care provided, a child is considered obliged to render services to his or her parents within the household, duties which are not considered child labour but a necessary part of socialization. (Mensah-Bonsu and Dowuona-Hammond,1996). As part of the support enjoyed by the biological parents, their children can be separated from them for longer periods to be raised either jointly but most of the time solely by other parents and this is known as child fosterage. 3.3.1 Child Fosterage Child fosterage is a common parenting practice in Ghana that involves the delegation of parental responsibilities to a trusted adult in the family, usually a close relation (Nukunya, 2003). It is also practiced among non-kin where the biological parents entrust the care of their child to a trusted friend or a person who is highly respected in society, such as a priest or a teacher. Child fosterage is practiced in Ghana for several reasons. These include forging links between adults and generations and strengthening familial bonds (Goody, 1982), sharing the cost of raising children (Goody, 1982; Isuigo-Abanihe, 1985) and raising responsible adults by subjecting the children to strict discipline (Bledsoe, 1990; Nukunya, 2003). Child fosterage is also largely considered as a child care strategy in response to life situations (Blanc and Lloyd, 1990). University of Ghana http://ugspace.ug.edu.gh 51 Parental migration, especially international migration, where parents travel to work abroad leaving one or more children behind to be raised by a relative or non-relative, is an example of a situation where child fosterage becomes useful as a child care strategy. Fosterage concerns the process of rearing, and thus the relationship between foster-parent and foster-child would be expected to be essentially affective and moral in character (Goody, 1982). It also thrives on reciprocity where child care is linked to adult care (Coe, 2011) in that one is encouraged to take care of a child in order to secure his/her own care in the future. Several types of child fosterage are practiced in Ghana. The most predominant type is that which Isuigo-Abanihe (1985) describes as kinship fostering. Most fosterage in Ghana takes place within a kinship framework. Among some groups in Ghana, fosterage is widespread and thus well arranged often ahead of a child's birth. Forinstance, the Gonja put great emphasis on the institution of kinship fostering whereby from the age of about five, children are sent away to be brought up by parental kin (Goody and Goody 1967). Kinship fostering is largely a consequence of the need to reallocate resources within the extended family or the kin group, to ensure the maximum survival of the unit and to strengthen kinship ties. In some societies, however, these arrangements are more flexible; biological parents seek people with whom they know their children will be safe and will have the prospects for a brighter future. These people sometimes include non-kin such as in the cases of the Fanti, Ewe, Ga and Krobo (Goody 1982). There is another form of fostering that is based on the function of the fostering arrangement. Goody (1982) calls this crisis fostering. Crisis fostering is initiated out of necessity, usually because the natal family of orientation is unable to fulfill these roles due to illness, death or divorce. Even though fostering in this case is initiated out of necessity, it has to be entered into voluntarily. Crisis fostering is generally thought to improve the survival of University of Ghana http://ugspace.ug.edu.gh 52 children by removing them from the source of a crisis, whether real or imagined (Isuigo- Abanihe, 1984). Goody (1982) has also identified other forms of fostering which are based on the nurturance requirement of the child. She described fostering that involved infants and young children as nurturant fostering because they required care such as the provision of food and early socialization to control impulses and bodily functions. In contrast, even though older children require food and shelter, the primary task for the foster parents of these children is preparing them for the adult roles they will take on and teaching them the skills necessary for this role and the values of society more generally. In particular, female children are sent to experienced women where they are expected to learn the domestic roles they will perform in their future homes (Fiawoo, 1978; Goody, 1982). Goody termed this type of child care educational or apprentice fostering, depending on which activity the fostered child is sent to be trained in (Goody 1982). Closely related to apprentice or educational fostering is wardship and alliance fostering (Isuigo-Abanihe, 1984). In many West African societies, there is a specific emphasis on the use of fostering to establish and strengthen social, economic or political alliances (Goody 1982). Children are sent as wards to the homes of non-relatives, friends and people of certain social standing, primarily teachers and clergy. In Muslim culture, children, especially boys, are often sent at a young age to live with influential religious or political leaders or landlords to receive care, training and instruction in the Koran. Because alliance and wardship fostering often combines the responsibilities of training and sponsoring of children, it goes hand in hand with apprentice fostering. Child fosterage constitutes part of the trend in which the welfare of rural dwellers depends on their solidarity ties with urban kin families. A key component of this practice is the University of Ghana http://ugspace.ug.edu.gh 53 channeling of remittances from urban workers to rural areas through educational support. By conferring parental responsibilities to their urban relatives, fosterage guarantees the mobility of children from rural families. In addition, when poor and large families transferred the guardianship of their children to smaller and wealthier families, fosterage mitigated inequalities in resource endowment among children (Bigombe and Khadiagala, 2003). Child fosterage is not intended as a permanent legal transfer to another family or institution. However, it transfers the majority of the responsibilities, including decision making regarding the child’s welfare to the foster parent. Since most fosterage situations in Ghana constitute an exchange between one party that has economic resources, or a higher social position in society (male, parent, chief and/or highly educated), and another who does not, the former most often has greater say in the upbringing of the child. Whereas it is described as sharing of parenting roles among members in lineage, the burden almost always falls on one or two people, which in most cases is the immediate benefactor of the child. If they are still alive, the natal parents of the child are welcome to contribute whatever they can in the child’s life, such as paying visits to the child, sending foodstuffs to the family or offering advice when necessary. However, in cases where the child is sent away to be trained to be an independent adult, parental involvement is usually discouraged because it is seen as defeating the purpose of the arrangement. Power relations in child fosterage, where more control is vested in the foster parent or caregivers, is emphasized in the conceptual framework because this is likely to change in the event of international migration where parents are also actively involved in the care of their children via remittances and modern communication technologies. University of Ghana http://ugspace.ug.edu.gh 54 3.3.2 Child Fosterage and Social Change The traditional ways of arranging familial and kinship fosterage has changed significantly (Findley, 1997; Ocholla-Ayayo, 1997; Nukunya, 2003). A number of studies argue that the ties that hold members of a group together have been weakened. The decline in support has been attributed to urbanization, migration, family nucleation and increasing literacy, rising expectations and aspirations and general economic hardship in most African countries (Bigombe and Khadiagala, 2003; Aboderin, 2004; Badasu, 2004; Adepoju, 2005; Oheneba-Sakyi and Takyi, 2006, Wusu and Isuigo-Abanihe, 2006). Some scholars also argue that individualism and family nucleation, which is characteristic of urban life demands its own code of behavior (Assimeng, 1999; Nukunya, 2003) The emphases have shifted from ‘our’ to ‘mine’ and from the ‘we’ of traditional corporate life to the ‘I’ of modern individualism (Ocholla-Ayayo, 1997: 64- 65). Family size is becoming smaller and more dispersed, and the traditional extended family concept is giving way to one in which the parents and their children constitute the family in the modern sense of the word (ibid). These changes appear to have affected child care and intergenerational support in both rural and urban areas. The resulting situation, reduced kin support for child care is, however, more pronounced in urban centers. One reason for this is that members of the same lineage are dispersed across different parts of the urban areas (Nukunya, 2003). In her work on declining support of the aged in Ghana, Aboderin (2004) also found that incapacity, especially economic among young people, contributes to the general decline of support for the aged. Even though Aboderin's work was about care for the aged, the reason for the decline of aged support can also be applied to child care as many families are finding it difficult to make ends meet and now rarely go beyond their nuclear families to take care of members of the extended family University of Ghana http://ugspace.ug.edu.gh 55 According to Nukunya, what such forces of social change--colonialism, urbanization, commercialization--does to an institution is to introduce flexibilities into its rules, principles and modes of operation (Nukunya, 2003). This means that the extended family has experienced a lot of changes in how it operates and governs its members. For instance, colonialism, urbanization and commercialization have changed the economic position of children with respect to their parents. Formerly, parents extended economic responsibilities over their children and the children likewise contributed economically to the family income by way of providing labour. Children worked as farm hands as well as helped at the market. This practice still exists in both rural and urban areas, but now children divide their time between going to school and performing these chores. This change has been facilitated primarily by government policies which provide free access to education. The Free Compulsory Universal Basic Education (fCUBE) was adopted in 1995, and included a cost-sharing scheme to cover non-tuition fees, which was intended to limit the expenses parents must bear for basic education. The ‘Capitation Grant’ system was also introduced in early 2005, which provides every public kindergarten, primary school and junior secondary school about $3.30 (GHC 3) per pupil per year. Finally, the Ghana School Feeding Programme, launched in September 2005 under the African Union-New Partnership for Africa’s Development (AU-NEPAD) initiative, was introduced to improve education service delivery and agricultural development and to reduce malnutrition among school-aged children. The government’s intervention through these three programmes achieved some success, in particular high enrolment rates in the country's most deprived areas where children would otherwise not have attended school (Afoakwah, 2007; UNICEF, 2007; Osei- Fosu, 2011). These programmes have reduced the cost of education for parents and resulted in University of Ghana http://ugspace.ug.edu.gh 56 more children enrolled in school, which also means fewer hands for helping parents on the farm and in the market. 3.4 Summary This background chapter began with an overview of international migration in Ghana where I discussed migration trends, estimates of migrants abroad and their transnational activities. It was followed by a discussion of the parenting context (in social parenthood and child fosterage) within which the study is situated. This context was introduced in the previous chapter as part of the conceptual framework that was developed for the study. In this chapter, I looked at some characteristics of social parenthood and child fosterage. I also made the case in this chapter that the institution of child fosterage has been affected by factors of social changes such as urbanization. University of Ghana http://ugspace.ug.edu.gh 57 CHAPTER FOUR: RESEARCH METHODOLOGY AND STUDY AREAS 4.0 Introduction This chapter describes the research design and methods employed in this study. The chapter is divided into two main sections. The first describes the research approach and provides the researcher’s justification for embarking on a qualitative study and, more importantly, for using ethnographic case studies. This is followed by a description of the Simultaneous Matched Sample (SMS) methodology which informed the selection of the cases used in the study. Next, I examine the instruments used in data collection (name generator tool, in-depth interviews, observations made during visits and my involvement in respondents’ social activities) and the data analysis. I conclude this section with some reflections on my position as a researcher and the ethical issues entailed in this study. The second section looks at the study areas – the Greater Accra and Ashanti Regions – and provides a justification for selecting these areas for the study. 4.1 Qualitative research This study uses a qualitative research approach. Theories of qualitative research stress that knowledge is created through the understanding of a phenomenon, text, object within a particular context and not merely gained through the discovery of objective truths. This study is, therefore, situated within a context in which child fosterage is a normal practice; however, examining child fosterage in a transnational context changes the dynamics of the practice. The involvement of parents in the daily lives of their children as a result of technological advancements raises questions about the role of caregivers in a transnational context and about the motives, expectations and challenges involved in raising children of parents who are in a position to pay for the material upkeep of their children. In exploring these issues, I had to ask University of Ghana http://ugspace.ug.edu.gh 58 the caregivers questions centered on how they got into the care arrangements, their motivations for doing so and their expectations of the migrants, among other things. Other issues I had to examine include the caregivers’ general satisfaction with the care arrangements and remittances, and, most critically, their willingness to disclose the legal status of their migrant relatives. Not only are these issues deemed private because they are related to the family but they are also sensitive, especially the latter on the migrants’ legal status in the host country since such disclosures could get their relatives into trouble. Due to the sensitive and vulnerable nature of the issues examined, the study required a research approach that allowed for a lot of time and contact with the respondents so that the researcher could win and build their trust. These experiences are best captured and reported descriptively in the words of the participants and sometimes with pictures rather than in numbers (Creswell, 2003). The use of qualitative research methods for this study builds on quantitative studies of transnational arrangements in the TCRA project of which this study forms a part. Certain processes and outcomes realized in the quantitative study demanded a deeper understanding which could not be met with the use of questionnaires administered by research assistants. In a qualitative study, the researcher, with the intent of elucidating the subject or phenomenon under study, is the primary instrument in data collection and is able to gradually make sense of a phenomenon. Participants’ perceptions and experiences are best captured through ongoing interactions, which a qualitative study allows. To validate the data, prolonged engagement, persistent observation through numerous visits to caregivers' houses and engaging in their activities constituted an important part of this study. Furthermore, data triangulation was also used extensively in this research to help reduce researchers’ bias. Data triangulation was achieved by exchanging notes with my Dutch colleague in the Netherlands, interacting with University of Ghana http://ugspace.ug.edu.gh 59 children and other members of the caregivers’ household, seeking clarification on previously obtained information during visits with caregivers and debriefing sessions with my supervisors. According to qualitative researchers, these features of qualitative research help the researcher check the validity, also referred to as credibility, authenticity, adequacy, verisimilitude, goodness and trustworthiness (Creswell and Miller, 2000) of her/ his findings. To gain access to a private sphere such as a household does not only require time to build trust, but it also requires intensive research methods that allow one to collect data from multiple sources. In light of this, the qualitative research used in this study is multi-methodical in focus, involving an interpretive, naturalistic approach towards the subject matter. It deploys a wide range of interconnected methods; always in the hope of getting a better insight on the subject matter at hand (Denzin and Lincoln, 2005). In this regard, this study makes use of ethnographic case studies and the main methods for collecting data from the participants were name generator questionnaires, in-depth interviews, and observation during visits, telephone conversations and participation in respondents’ activities. 4.2 Ethnographic Case Study According to Yin (2003:13), a case study is “an empirical inquiry that investigates a contemporary phenomenon within its real-life context, especially when the boundaries between phenomenon and context are not clearly evident… [and it] relies on multiple sources of evidence.” The emphasis here is on the deliberate attempt to investigate a particular situation in its real life context. Each case is an object of interest in its own right, and the researcher provides an in-depth elucidation of it. A case in this study denotes the relationship that binds the three University of Ghana http://ugspace.ug.edu.gh 60 actors which make up the TCRAs: the migrants, the caregivers and the children that are left behind. In each case in this study, the caregiver is the unit of analysis. Ethnography was the main research method used in each case. Falzon (2009) describes ethnography as an “eclectic methodological choice which privileges an engaged, contextually rich and nuanced type of qualitative research in which fine-grained daily interactions constitute the lifeblood of the data produced” (1). Another scholar defines ethnography as social research based on the close-up, on-the-ground observation of people and institutions in real time and space in which the investigator embeds him or herself near or within the phenomenon in order to detect how and why agents at the scene act, think and behave the way they do (Wacquant, 2003: 5). A lot goes into a transnational relationship that can easily be missed in one-off interviews or in a series of a few qualitative interviews. This is where ethnographic research methods come into play. Ethnographic research methods allow for understanding the experiences, interpretations, activities and actions of caregivers in TCRAs by triangulating different tools. It encourages moving beyond just what participants in a study say to observing what they actually do (Mazzucato,2009a). The intent of ethnography research is, among other things, to obtain a holistic picture of the subject with emphasis on portraying the everyday experiences of individuals by observing and interviewing them and relevant others (Creswell, 2003). Conventionally, ethnographic studies have favored the practice of the researcher living in a single study site over a lengthy period and living as the participants do. However, these conventions have evolved over time, and a lot of flexibility has been introduced into ethnographic research where researchers spend shorter times than originally prescribed in the field (Bryman, 2008). Ultimately, an ethnographic study aims to attain a ‘thick’ description of the phenomenon being studied, and this usually entails continuous University of Ghana http://ugspace.ug.edu.gh 61 interaction between the researcher and the researched. Furthermore, because of the time needed to build relationships of trust between the researcher and the researched, most ethnographic studies encourage the use of small numbers rather than large numbers of respondents because this helps to ensure building and sustaining strong relationships. Establishing a strong relationship and building a rapport between the researcher and the researched formed an important part of this study. Spradley (1979) has outlined certain stages of building a rapport, namely apprehension, exploration, cooperation and participation, all of which I experienced during my fieldwork. According to Spradley, the initial stage of apprehension is quite natural. It is characterized by uncertainty that stems from the strange context within which both the interviewer and the interviewee find themselves. Such apprehension occurred quite often in the initial stage of data collection, especially in TCRAs where the migrant had an undocumented status in the host country. It often took respondents a while to understand my mission and to become convinced that I was not working for a government organization before they opened up. The second stage is the exploration phase where the interviewee engages in in- depth descriptions. This descriptive process is accompanied by learning, listening, testing, sharing and bonding between the researcher and the respondent. The third stage, which is that of cooperation, is where the interviewer uses different methods to seek clarification and validation on the issues raised in the previous stages. At this point, the interviewee has the opportunity to correct the interviewer on certain points or on the assumptions drawn from previous discussions in a process in which they try to work together to make sense of the interviewee’s world. The last stage is the participation stage, where the greatest amount of rapport is achieved. At this stage, the interviewee may take on the role of a guide. The researcher and the researched become familiar with each other; the researched as something more than just a human subject – a University of Ghana http://ugspace.ug.edu.gh 62 knowing and authentic individual, and the researcher as something less an exalted and authoritative social scientist. I recall going through these phases with my interviewees during my fieldwork. However, the time it took to move from one stage to the other varied from respondent to respondent. These transitions were facilitated by other activities, such as making phone calls and paying informal visits to respondents as well as interventions by the migrant. In a few cases, we never got to the final phase of participation. 4.3 Selection of Cases -The Simultaneous Matched Sample (SMS) methodology2 As mentioned in the introduction, this study is part of a larger project to examine transnational child raising arrangements between Ghana and the Netherlands. The project involves parents in the Netherlands and their children and their children’s caregivers in Ghana. The larger study explores how care relationships are developed, (re-)configured and maintained across the two nation-states by employing a multi-sited ethnography. This kind of exploration draws heavily on two elements of transnationalism - networks and simultaneity (Mazzucato, 2009a). The element of simultaneity in transnationalism allows for studies in a transnational flow to move beyond methodological nationalism (Wimmer and Glick-Schiller, 2003) where one studies one site (nation-state) at a time. Stressing the importance of multi-sited research, Mazzucato (2007) explains that because such studies collect information from both sides they can best investigate the inner workings of transnational flows and link the actions of migrants with those of the people back home. These studies also highlight reciprocal relations that exist among persons in different countries and how they work. This method of studying transnational phenomena has been corroborated by Madianou and Miller (2012) who noted that 2 For more on our experiences of using the SMS methodology see Poeze and Dankyi, 2013 University of Ghana http://ugspace.ug.edu.gh 63 transnationalism is all about relationships and following them (rather than assuming them) is one way of dealing with the perils of methodological nationalism. To best achieve this, a multi-sited approach is necessary. The most common approach which has been applied to the study of transnational networks is a multi-sited methodology applied in a step-wise fashion, meaning that one researcher would follow a transnational network in two or more countries. The step wise approach has been used in the past by authors such as Madianou and Miller (2012), Asima (2010), Dreby (2010), Erel (2002) and Coe (2002). Mazzucato (2009a), however, argued that, as a result of practical limitations, this often led to short periods of observation, and, with that, a lack of depth. In addition, this approach is unable to capture the simultaneity of impact of events and transnational flows in two different nation-states. In order, therefore, to “bridge the boundary between breadth and depth and to incorporate simultaneity and networks”, Mazzucato (2009a) proposed the SMS methodology, which she describes as ‘using a sample of informants who are linked to each other by being part of a same social network and studying these informants in a simultaneous fashion so that information obtained from one informant in one locality can be immediately linked up with information obtained from another informant in another locality (217). The SMS methodology encourages the use of multiple researchers at multiple sites in real time. This helps to capture the instantaneity of transnational flows and its impact on people’s daily lives and relationships in different localities. What this means for this study is that I conducted ethnographic fieldwork among caregivers in Accra and Kumasi, who were matched to parents with whom a Dutch researcher conducted ethnographic study in Amsterdam. 4.3.1 How the SMS methodology was adopted for this study The SMS methodology was instrumental to this study in three ways: the selection of cases, the exchange of field notes and the exchange of field visits. University of Ghana http://ugspace.ug.edu.gh 64 Selection of Cases – Matched Samples We (the researcher in the host country and I) began our search for matched samples together in Bijlmer, the Netherlands, where a large number in the Dutch Ghanaian community reside. We deployed different means to find respondents. We made announcements on Ghanaian radio stations and in churches, and we used gatekeepers and Ghanaian interviewers who had worked for the larger TCRA project as well as the previous contacts which the TCRA team of researchers had made. In order for me to maintain regular contact with the respondents in Ghana, we restricted ourselves to parents in Amsterdam whose children and their caregivers were residing in Accra and Kumasi. We were able to select 15 matched samples, which represented a variety in legal and socioeconomic status (migrants), sex (all respondents), and type of caregiver (Poeze and Dankyi, 2013). The first step after identifying parents in the Netherlands and receiving their consent to talk to their families back home was to contact the families in Ghana via telephone. In the initial telephone call, I introduced myself by making reference to the migrant by name and then ask if the migrant had mentioned my research to them. In a few instances, the migrants had already informed the caregivers prior to my call, while in other instances no prior notice had been given. The initial challenge I had with both those who had prior knowledge of my study and those who did not was that they lacked the full details of my research. There was also the bigger challenge of having to correct the impression that the interviewees had that I had come to interview the caregiver and the children in order to facilitate a reunification with the parents abroad. It took the first two and sometimes three visits as well as a constant reminder from the parents to clear them of this impression. Some caregivers during this initialising stage lost interest when they realised University of Ghana http://ugspace.ug.edu.gh 65 that I had not come to help initiate a reunification process. My colleague in the host country also lost some parents for this same reason. Exchange of field notes Another feature of the SMS methodology was the constant sharing of field notes between my Dutch colleague and myself. We exchanged notes with each other at the end of each interaction with a respondent. In order not to overload the other researcher with interview notes, we decided to share summaries of each field interaction be it an informal visit, a telephone call or a more formal interview. We took particular notice of information on which we needed the other to obtain clarification or we deemed useful for subsequent interactions with the interviewees at our respective sites. We also discussed the most important observations, (signs of) tensions and issues we believed would be of particular interest to the other researcher. This provided a huge advantage as it helped to enhance the questions I asked during my interviews. I also learned what can be best described as the unspoken side of transnational relationships, especially between the caregiver and the migrant (Poeze et al., in press). Without knowing what was happening on the other side, it would have been difficult to glean some of the information I got by following my normal interview routines and observations during visits. In comparing the other's summaries with our own observations, we often detected inconsistencies in what was said by different respondents at the different research sites. Through almost daily emailing we discussed these inconsistencies and asked each other to elaborate on them. In situations where they were due to a miscommunication between the respondent and the researcher, we tried to ask our questions more accurately during the next visit. We also came to understand that some of the inconsistencies reflected important transnational relationship dynamics, most notably the information gap that existed between the respondents (Poeze and Dankyi, 2013). Despite the fact University of Ghana http://ugspace.ug.edu.gh 66 that they often faced challenging circumstances, the respondents on both sides withheld information from each other in an effort to keep relationships intact and to diminish distrust. It was by being present at both ends of the relationship and by sharing information on the nitty- gritty details of daily life among multiple transnational families that we gained important analytical insights into the underlying dynamics of transnational relationships. The sharing of field notes also informed us about important events that we otherwise would not have been able to capture. For instance, one caregiver complained bitterly to me about how the migrant had moved them (caregiver and the migrant’s children) from the house they were renting into their own building that was not yet fully built. She told me that on countless occasions she had asked the migrant to fix some things in the house to make it more habitable, but all of her requests had fallen on deaf ears. In her estimation, the migrant was deliberately ignoring her concerns. Through the sharing of notes after this interaction, I learnt from my colleague that the migrant had lost his job and had also lost his legal pursuit for a residence permit. He was, therefore, in very dire straits, but had deliberately not told the caregiver and the children because he did not want to overwhelm them with worry. Situations like these presented some ethical issues for me which I discuss later in the chapter. Shared Field Visits During my research period, I spent a total of eight months spread over three consecutive years in the host country. One of the key objectives of these trips was to familiarise myself with the context within which the Ghanaian in the host country lived. On the first trip, I joined my colleague in the search for interviewees. On subsequent trips, I spent the time visiting the existing interviewees and meeting new ones. For these field visits I teamed up with my colleague research partner because we deemed that she was more familiar with the people and University of Ghana http://ugspace.ug.edu.gh 67 may have already gained their trust. These team visits were also important because it was possible that some useful information might be revealed during the visits that would prove useful to her, the primary researcher in this context. In turn, my colleague visited Ghana once during the research period and, similarly, I took her with me on visits to interviewees in Accra. These visits helped in several ways. First, they helped to build trust between the interviewees and us, and the interviewees became aware of the researchers working on both sides. The migrant parents were always excited to see me. This was especially so in the case of those who, as a result of their undocumented status, could not visit home. I sometimes carried pictures (that I took and others that the caregivers and children had already had taken) and messages (video and audio recorded messages) to the parents. The parents often sent back in return gifts, pictures and food items to their children in Ghana through me. They also often made the same request of my colleague when she was visiting Ghana. The exchange of gifts, pictures and messages from the one side to the other always generated interesting information. Mazzucato (2009a) noted that sending these gifts and messages to the other side provided a wealth of information in terms of what was sent and how these flows were organized. Secondly, a lot was revealed during the visits which a researcher working on just one side would not have been privy to. For instance, on one occasion, while on a visit to one of the caregivers and the children she cared for in Ghana, my colleague was allowed into the caregiver's bedroom so she could see for herself the mess the children had made of the room and report this to the migrant in the Netherlands. In this manner, the caregiver visibly showed her frustrations with the care situation. Meanwhile, even though I had visited this woman’s house frequently, I University of Ghana http://ugspace.ug.edu.gh 68 had never been let into that part of the house. With this revelation, I got a better understanding of the frustrations often expressed by this caregiver during our interviews. In addition to gaining some new understanding about the interviewees, after the visits each researcher was able to better contextualize and interpret the data of the other researcher. Moreover, these visits enabled the researchers to witness the interviewees' living conditions. They also helped to expose the anxieties of the caregivers who often complained about lack or insufficiency of the remittances and of stay-behind wives who sometimes thought their migrant husbands were spending money on other women in the host country while neglecting to care for the family back home. Although as a team we shared our findings and observations in field notes, to better understand the actual circumstances of respondents, it was important for us to witness their living conditions in person. 4.3.2 Challenges with the SMS Finding matched samples proved time consuming as a result of having to establish the trust needed for respondents to give the researcher the contact information of those in Ghana. There were some instances in which the migrant parent had given consent but the caregiver did not agree to participate. Keeping the matched samples was also challenging because the ethnographic research proved time consuming for interviewees who often had to combine caring for young children with busy work schedules. This, in combination with the sensitive topics we addressed, led to some interviewees opting out of the survey. Once a member of the triad dropped out of the matched sample, we lost the whole triad and had to look for a replacement. University of Ghana http://ugspace.ug.edu.gh 69 4.4 Selection of Cases – Non-Matched Samples In addition to the 15 caregivers in the matched sample, I interviewed 19 caregivers in a non-matched sample category. I met them for a one-off semi-structured interview. The main justification for conducting additional interviews with these caregivers is the assumption that many of the matched samples were considered ‘problem free’ because of the double consent needed from the respondents on both sites. It is assumed that only parents and caregivers who had a cordial relationship will consent to participate in the research (Poeze and Dankyi, 2013). I, therefore, included additional caregivers who were not matched to any migrants abroad in order to tease out the ‘not so nice aspects of TCRAs’. The 19 caregivers were selected through the schools in Ghana where the quantitative aspects of the larger TCRA project had been conducted. The quantitative study engaged 2,760 children in randomly selected schools in four locations in Ghana including Accra and Kumasi. The destination countries for migrants in the non-matched sample were not restricted to the Netherlands; they included other European countries, the United States of America and Canada. I visited some of the schools both in Accra and in Kumasi to meet some of the children who participated in the research. With the approval of school authorities, I took the telephone numbers of the caregivers from the children. I then called the caregivers and sought permission to interview them. Some caregivers immediately agreed and we fixed an appointment for an interview. Others took longer to respond because they wanted to check with the parents abroad first. Some of them declined my request immediately, while others only did so later after I had made several calls to follow up on my request. I ended up with 19 because I had reached theoretical saturation. The saturation was more of what Rebar et al. (2011) called a descriptive saturation in that at a certain point I found there were no more new themes emerging from the interviews. I therefore ended the sampling to pursue the themes University of Ghana http://ugspace.ug.edu.gh 70 that had emerged in greater depth with interviewees in the ethnographic study. Second, trying to get a larger number of respondents took longer than I estimated, and I needed to spend more time on the 15 caregivers with whom I was already conducting ethnographic research. The characteristics of the 19 caregivers are presented in detail in addition to the matched sample in the next chapter. 4.5 Data Collection Data collection took place in the Greater Accra and the Ashanti Regions from August 2011 to August 2013. Throughout that period, I made visits to the Netherlands for durations of 3 months in 2011, 2 months in 2012 and another 3 months in 2013 in order to enable me to develop part of the SMS methodology. The remaining 19 months were spent in the two sites selected in Ghana. Within a period of about eight months I made nine trips to Kumasi during which I spent a maximum of 21 days each. The rest of the fieldwork period was spent in Accra. Data collection started with phone calls to the caregivers. During the initial phone interaction, I made an appointment with the family for my first visit. The initial visit, which was to be the first of many during the period, was to introduce myself and the project to the caregivers. Even though I had already explained the project to them on the phone, it appeared that this had not helped at all. I always had to repeat fully everything that I had said during the initial phone call as well as explain that I had already spoken with their migrant relative from whom I had obtained consent and their telephone numbers so I could contact them. During the introductory visit I often spent some time making general conversation about the children, the migrant and answering any questions they had. Their questions and the focus of the conversation University of Ghana http://ugspace.ug.edu.gh 71 were often on the migrants and how they were doing because I mentioned that I had been to the Netherlands and had met their migrant relative. I often used the first visits to observe the physical living conditions of the house as well as to take note of the number of people in it. Another thing I did on the first day which I also often repeated on subsequent visits to respondents was to call the migrant relative in the Netherlands to inform them that I had arrived safely in their house in Accra or Kumasi. I would then hand the phone over to the caregiver and the children who would take turns talking to the migrant. These activities were geared towards establishing a good rapport with and trust from the respondents. Often, the phone call somehow automatically broke the ice; everyone who talked to the migrant ended up laughing about one thing or another. In most cases, these initial visits lasted between 45 minutes to an hour. At the end of the visit, I would make an appointment for our next meeting at the respondents’ earliest convenience. I also made them understand that I was going to be visiting more than once and that, if they did not mind, there were going to be times when I would visit to find out how they were faring. On the average, with a minimum of three and a maximum of six visits, I visited each of the 15 caregivers three times, including the initial visit. As mentioned earlier, ethnographic research allows for what Falzon (2009) calls a situational combination of field techniques. I employed a number of these in order to obtain a thick description of the experiences of the caregivers in the study. I used the name generator questionnaire mainly to establish networks of caregivers. These were followed by in-depth interviews. Moreover, through observation during visits, I also paid a lot of attention to the things that were not said but were implied. Furthermore, I paid informal visits in which I did not conduct any interviews, and I also made a lot of telephone calls in-between visits. In addition to these sources of data were the summaries of field notes that I received from my Dutch colleague University of Ghana http://ugspace.ug.edu.gh 72 and the observations I made during my visits to the host country. I also engaged in a lot of activities with the participants. For instance, I joined some caregivers for church services or visited the boarding schools of the children under a caregiver's care upon request from their parents. One parent also asked me to visit his hometown where he had left one of his children in the care of his wife’s relatives, which I did. I also attended graduation ceremonies of some of the children and sometimes invited them out. For this, I usually asked the caregivers for permission and also I personally went to pick the children from their house and drop them off afterwards. This often gave me the opportunity to visit the house without making a prior appointment for an interview. In general, fieldwork with the caregivers did not follow any strict schedule, but, as encouraged by most qualitative and ethnographic studies, each visit, phone call or activity engaged in was aimed at reaching a deeper understanding of the caregiver’s experiences in the TCRAs. Regardless of the unstructured nature of the schedule, there were two mandatory activities in which I engaged with respondents--completing the network mapping tool (Appendix C) and in-depth interviews. The network mapping tool exercise kick-started my interviews. After this exercise, the subsequent interactions were geared towards asking more questions that were generated from the initial interview and building upon what had been done in the mapping exercise. 4.5.1 The Network Mapping Exercise and Social Network Analysis A social network is the amalgamation of ties among actors in a given setting (Wetherel, 1998). It is considered as a structure, made up of individuals or organizations (nodes) or egos that are tied by one specific type, or more, of interdependency, such as friendship, kinship, University of Ghana http://ugspace.ug.edu.gh 73 common interest, financial exchange, dislike, sexual relationships or relationships of belief, knowledge or prestige (ibid) SNA also stresses the notion that individual actors are a part of social relations and, in interaction with network partners, are actively engaged in the exchange of information, material and immaterial goods and services (Bernardi, 2011). Using SNA, families can be analyzed as a configuration of relationships that goes beyond shared housing, residential proximity, legal membership and national and cultural barriers (Bernardi, 2011). This approach is well suited to the study of transnational families and TCRAs, which involve short and long term separation and dispersion across national borders. A study of these transnational social networks helps us to understand how the relations between members of a triad (parent, caregiver and child) help to address the social, psychological and practical challenges that voluntary separation may pose to members, especially children who are separated from their parents (Bernardi, 2011). Despite the potential wealth and understanding that transnational social networks promises to bring to the study of transnational families, researchers have hardly looked at what a transnational social network looks like (Bernardi, 2011). Moreover, as far as I am aware, there are no studies that systematically examine the wider networks of caregivers in transnational families, while assertions have been made that caregivers can rely on large networks of support (Orozco 2006, Carling et al 2012). In this study, I describe the transnational network of my interviewees using the characteristic features of networks that SNA highlights, namely their size, composition, spatial dispersion and their function (type of support they render). I identify the ties in a network primarily through a name generator questionnaire, another important methodological feature borrowed from quantitative SNA (Burt 1984; Campbell and Lee 1991). The questionnaire helps motivate caregivers to name all the persons who are important to them University of Ghana http://ugspace.ug.edu.gh 74 in some way, especially with respect to the care of a migrant’s child. I asked caregivers to mention the names of the people who assist them with care of the migrants’ children and with the help of the questionnaire, to classify them under the particular type of care they provided: material, emotional and/or social. The issue of material care was discussed using questions such as who provides money to pay for school fees, food and cooking and medical bills, etc. Emotional care was probed using questions about from whom the caregiver sought advice and with whom the caregiver had arguments regarding the care of the child/ren. The social dimension of the children's care was examined through questions about who educates the child/ren on how to behave, who gives advice to the child/ren and with whom does/do the child/ren spend their free time. Questions were also asked about positive relationships (such as friendships) as well as negative relationships (such as people one argues with) with regard to kinship, consanguineous and non-kinship relationships (Mazzucato et al, forth coming). I asked respondents to mention as many names as they could think of for each question and in addition to providing names, to note the location of the person, his or her relationship with the child (kin/non-kin), and, for each name mentioned, to specify what kind of relationship (such as ‘business partner’ or ‘mother’s sister’) exists between that person and the respondent, as well as the frequency and means of contact. Follow-up in-depth interviews were conducted with each of the caregivers, and the network tool was used as a reference, either to cross-check information or to obtain further clarification on the roles of specific people in the caregiver’s care network. 4.5.2 In-depth Interviews To delve deeper into various aspects of the research, in-depth interviews were conducted. With the basic research question in mind on how caregivers experienced their TCRAs, several University of Ghana http://ugspace.ug.edu.gh 75 other questions were developed to focus on specific areas of the basic research question. In consonance with qualitative studies methods, some of the questions lost relevance as the study proceeded and had to be adjusted to fit the parameters of the preliminary analyses which had already begun alongside the interviews. In line with the characteristics of qualitative research, the researcher demonstrated the ability to know what to drop and what to pursue as well as to adjust or add to the existing set of questions as the occasion demanded. The themes around which I conducted the in-depth interviews focused specifically on the caregiver’s experiences. They were designed to investigate the practical ways in which care was arranged, such as how decisions were made in the care arrangement. The in-depth interviews also looked at the caregivers’ responsibilities with respect to the children in their day-to-day care as well as to the migrants. Another issue was that of compensation and expectations. Did caregivers have expectations in the form of rewards from the parents? Did they have expectations of future reciprocal care from the children in their care? Did they receive help with household chores, especially from the older children? The interviews also explored the respondents’ sense of satisfaction, such as with the care arrangement itself and whether there were aspects of it that were stressful. Yet another line of exploration related to conflicts and conflict resolution. The themes pursued in the interviews were adjusted to suit the experiences of the various respondents, especially when the respondent was the child’s biological mother. In the case of biological mothers, the theme of rewards or compensation was taken up to learn if the mother had any expectations of her child in terms of future reciprocal care. 4.5.3 Life history interviews with a focus on respondents upbringing. A life history interview is an approach that uses a form of individual interviews directed at documenting the respondent’s life or an aspect of it that has developed over the life course University of Ghana http://ugspace.ug.edu.gh 76 (Fielding, 2006). This approach can be demanding on the researcher and the respondents as it seeks to go into depth and to recover experiences across long periods of time. In this study, we sought to uncover a certain aspect of the caregivers’ life – their own upbringing. This demanded caregivers to recall activities and situations that took place several years ago, which some of the older respondents could do only with difficulty. Some of the caregivers found this exercise nostalgic and spoke fondly about events that took place when they were growing up, while others expressed disappointment in how children of today are missing out on a good upbringing. The aim of these interviews was to help establish if the caregiver’s own upbringing had any bearing on how the caregiver was raising the migrant’s child. Another aim was to help ascertain if some of the changes in child fosterage as per the caregiver’s own upbringing could in some ways be linked to migration, that is, the fact that the parents of the child in their care are abroad. 4.5.4 Other Data Collection Activities Telephone Calls Besides the interviews and contact interactions, I made phone calls to the respondents, especially to those in the matched sample, in-between visits. My phone calls were aimed mainly at checking on the caregivers and the children, including the caregivers’ own children. I also often asked when the last time was since they had heard from their migrant relatives. Normally, I made these calls at least once in-between visits. For respondents in Kumasi, I took advantage of my return trip to inform them about my safe arrival, and that was how I was able to start a conversation. Initially, especially after the first visits, these calls were quite difficult to make because the caregivers were not very friendly on the phone, and it usually took some time before University of Ghana http://ugspace.ug.edu.gh 77 they could recognize me. However, after the first, second and sometimes third visit, these follow- up calls became easier to make, and in many cases it became possible to stay longer on the phone and discuss the child, the migrant or more general issues. For instance, through these calls I was able to corroborate the information I had gotten from those who told me that the migrant called very often, by asking when the migrant last called. I often got responses like “yesterday,” “the day before” or “we just finished talking.” I usually made one phone call to check on them and then another to make an appointment for my next visit. In addition, I usually also called the migrant when I visited the caregivers, during which time I had the advantage of observing how the caregivers talked with the migrants on the phone. This helped me to understand, confirm or disprove what they had told me about their relationship with the migrant. Informal Visits Informal visits occurred more frequently with the respondents in Accra than those in Kumasi simply because I was based in Accra. Fortunately, some of the respondents lived close to my house or close to the University where I was doing my research. As a result, on my way back and forth, I often seized the opportunity to pay them a visit. Since I had already informed respondents ahead of time that I would be having multiple interview sessions with them, as soon as they saw me, they got ready to be interviewed. On the occasions when I had not come to conduct an interview, I then had to tell them that I was just passing through to see how they were doing. These informal visits were also aimed at establishing trust. Moreover, I also inadvertently got to observe interactions between caregivers and the children or other household members. I also sometimes chanced upon caregivers in the middle of doing one of their chores, such as cooking or tending to young children, or find the children running errands for caregivers. Despite University of Ghana http://ugspace.ug.edu.gh 78 the distance, I also did a similar thing with some of the caregivers in Kumasi, even though, in their case, I often called before visiting. I usually told them that I was visiting because it had been a while since I had seen them. On such occasions we talked about the migrant, the child or the caregiver’s life in general. Another activity includes the last visit I made to most caregivers after I had collected enough data from them. This visit was to express my gratitude and to let them know that I had ended my interviews with them. However, I often added that I would speak with them on the telephone should the need arise. I found this important because, on this last visit, I often got some valuable information from some respondents. Some reiterated things they had told me in the past which they deemed important whereas in the case of others I got new and useful information. Participating in Respondents’ Activities Participating in caregivers’ activities proved a valuable technique for building rapport, establishing trust, observing them in different settings and to triangulate data. As a researcher, I was able to get more clarification and sometimes confirmation of previously obtained information through participating in the respondents' activities with them. On one occasion, I had the opportunity to attend the funeral of a respondent’s grandmother. The migrant had told me about the funeral when I met him in the Netherlands, and on my return to Accra I followed up and attended on the scheduled date. At the funeral I had the opportunity to meet some of the people whom the caregiver had named in her care network and some whose names had come up during the in-depth interviews. On another occasion, I also accompanied a caregiver and the children in his care to church. This occurred because the caregiver had indicated that the only University of Ghana http://ugspace.ug.edu.gh 79 time he could make it for an interview was on Sunday mornings before he left for church at 8 am. I, therefore, decided to join her and the children to church on my first visit to their house. I arrived there early enough to speak with them and introduce myself and my research. I then joined them to church and sat beside the caregiver throughout the service. In other instances, I joined participants while they attended school activities related to the children in their care. In one case, this included a Parent Teacher Association (PTA) meeting and, in another, the graduation ceremony of the migrant’s child. There were also times when I offered to take children to school or, in the case of those in boarding schools, to visit the children during visiting times instead of the caregiver. This saved the caregivers time and money, but it also gave me more access to the caregivers because I had to communicate with them prior to the visit and give them feedback after it. 4.6 Recording Data The main medium of recording data was note taking. I refrained from audio recording interviews right from the commencement of data collection because I noticed some of the caregivers were not comfortable. I, however, did audio-tape interviews with a few caregivers who gave sample who gave their consent. In the field, I did field jotting (Dooley, 1995) on the spot because writing longer notes sometimes affected the flow of the conversation. As time went on, I realised that some of the respondents preferred unstructured conversations to a formal environment of asking questions and writing. With these caregivers, I had to refrain from taking field jottings. Emerson et al. (2011) talks about the dilemma that many ethnographers are faced with in the field with respect to the desire to preserve the immediacy of a moment by jotting down words as they are spoken and capturing details of scenes as they are enacted, while, on the University of Ghana http://ugspace.ug.edu.gh 80 other hand, they feel that by taking out note pads or smart phones they will ruin the moment and plant seeds of distrust. In my case, I took a cue from what Dooley (1995) and Emerson et al. (2011) advised by seeking the opportunity in fleeting moments of privacy to make brief notes of key words and phrases. Sometimes I asked to go to the bathroom where I would quickly write something on my phone. In other cases, the respondent asked to be excused to attend to some situation or someone, and then I quickly wrote something. I, therefore, have phrases like “remember party” or “remember shoe incident” in my notebook. This approach had the down side of being subject to the general problems of bias, poor recall, inaccurate articulation (Yin, 1994) and the absence of direct quotes from respondents. However, the benefits of employing this approach, especially with the respondents mentioned above, far outweighed the disadvantage, and it is clear that the pursuit of direct quotes from respondents would have adversely affected the study. I also kept a book in which I recorded personal observations and reactions. Additionally, I shared a lot of my personal reactions with my Dutch colleague, especially if the reactions were in response to some inconsistency in data we had gathered from both sides on the same subject. The summaries of field notes that I exchanged with my Dutch colleague formed another essential source of data for this study. 4.7 Data Analysis According to Creswell (2009) and Bryman (2008), qualitative data analysis must be carried out simultaneously with the data collection process. Similarly, Dooley (1995) states that qualitative analysis begins with the first observation. Creswell (2009) again asserts that since qualitative data analysis depends on the purpose of the research, it is important that the proposed method of analysis be carefully considered in planning the research, and that it be integrated University of Ghana http://ugspace.ug.edu.gh 81 from the start with other parts of the research rather than being an afterthought. There are several methods to analyze qualitative and ethnographic data, and for this study I employ a thematic analysis. As noted above, the initial in-depth interviews were conducted to cover some of the thematic areas. These areas were derived primarily from the literature on TCRAs and data originating from the quantitative survey of the larger TCRA project. As the various steps of qualitative analysis data played out and as data collection progressed, new themes deriving from the data being collected began to emerge. I went back into the field to explore further the new themes that emerged and, following the processes of a thematic analysis, I looked out for sub- themes of the main themes. This process led me into the field for more feedback on what I had identified as emerging themes and sub-themes from research questions based on the literature. This process continued until saturation point where further themes were no longer discernible. Some of the main themes that emerged from the field included pathways into care, reciprocity, communication, power relations, remittances and small networks. I then interwove both the main themes and the sub-themes from my findings with the literature to help answer the various questions and make meaningful contributions to the debate on transnational parenting from the perspective of caregivers. 4.8 Ethical Considerations 4.8.1 Informed Consent, Confidentiality and Anonymity Seeking informed consent for participation in this project started during the search for interviewees in Amsterdam. My colleague and I designed a summary of the project, detailing the research objectives and participants’ involvement in the study, emphasizing that participation was voluntary. In most of the churches and on the radio station where we publicized the project, University of Ghana http://ugspace.ug.edu.gh 82 we introduced ourselves and the project, gave an oral summary of the research as presented on the paper, and invited those who might be interested or who sought further clarification to meet with us at the end of the service or at a specified time stated in the radio broadcast. During those one-on-one interactions, we explained the project in more detail to the prospective participants and then handed out the written summary which contained a part where they could fill out their details upon agreeing to join the project. Those who agreed to join gave us their contact details immediately. Others took the form which had our contact details away with them and then promised to call us after they had given the idea some thought. Even for those who agreed and gave us their numbers to call for interviews, we still had to spend a better part of our first meeting elaborating even further on what our research was about and, more importantly, to make them understand that we were independent researchers who were not working with any government institution. We also made them understand that if at any point at all they felt uncomfortable about continuing they were free to opt out. In our case, it became necessary to spend a lot of time explaining the project to the prospective participants because the mere mention of working with children and their caregivers back in Ghana gave them the impression that we were going to assist them with family reunification. We had a good number of people opting out when they came to understand our project and were certain that we were not working on reunification. I went through a similar exercise with the caregivers in Ghana and again some people opted out when they understood what the project was actually about. I am, therefore, convinced that those who stayed with us had received enough information to enable them to give informed consent for participating in the research. I assured respondents of anonymity and confidentiality by telling them that I would use pseudonyms and avoid narrating incidents that could easily be traced to them. Subsequently, University of Ghana http://ugspace.ug.edu.gh 83 pseudonyms have been used in the thesis to protect the identity of the respondents. Additionally, I did not just have to ensure confidentiality of respondents’ information from the outside world. I also needed to be very careful because the study set up gave me the opportunity to glean information about my respondents from the other members of the triangulated set, which the other respondents in the same set were not privy to, especially for the matched sample. Due to the exchange of field notes and the number of field visits, I had information from the migrants’ side that the caregivers and sometimes the children did not know. Consequently, I also had the task of ensuring that what I shared from my field site between my colleague and me did not come to the migrants’ notice. There were some instances where we had to check the inconsistency in information that we had received from both sides. We employed very subtle ways of doing this, especially in the way we asked questions in subsequent interviews. However, in most cases we had to forget about pursuing the matter altogether as that could have led to divulging information from the other side and was likely to break trust and compromise our findings as well. In addition to checking for inconsistency, during visits to the migrants’ side, respondents sometimes asked a lot of questions in their bid to know what was going on back at home. This became a bit disturbing in the case of migrants who did not really trust the caregivers. Here, I still refrained from giving any information that I had been told in confidence or narrate any scenes that I had observed. Fortunately, I was not confronted with any cases where caregivers were abusing the children in their care. Should that have occurred, I would have been faced with the moral dilemma of reporting to either the parents or an institution that was in the position to handle the situation. University of Ghana http://ugspace.ug.edu.gh 84 4.8.2 Positionality and Representation One of the main criticisms of ethnography has been the question of representation. Loon (2001) questions whether ethnographers adequately represent their participants in writing, and whether the writing adequately represents what is happening. This has created concerns because of the underlying propensity of ethnographers to become enmeshed in the settings and in the lives of the subjects they study. Stacey (1991) affirms that this had mostly put participants at a grave risk of manipulation and betrayal. In addition to being enmeshed in the social world they are studying and its consequent challenges, another factor that is likely to affect representation is the positioning of the researcher since there are a variety of factors that ‘filter knowledge’ (Fontana and Frey, 2005:712). A dimension of positionality that is particularly important in social science research is power relations between the researcher and the researched (Cormode & Hughes, 1999) where adapting to changing power relations with the researched can be a daunting task for a researcher. In my study, where a majority of the caregivers were 60 years and above, there was bound to be the hierarchical power relation between these caregivers and more so in cultural setting where seniority is highly cherished. I also worked with caregivers who had very low levels of education and considering my level of education, there was also the tendency of power relations between the very highly educated researcher and a low educated researched. These realities notwithstanding, they did not affect my data collection activities in any way. One dimension of positionality that significantly impacted my data collection activities is the insider/outsider perception of the researcher. An insider has been described as one that studies a group that s/he belongs to and identifies with. In this respect, I consider myself as an insider in this research. This is so because I am a Ghanaian who speaks and understands the language of my respondents, and this helped greatly in both Ghana and especially the University of Ghana http://ugspace.ug.edu.gh 85 Netherlands. My presence helped to build trust in the respondents because they usually associated my colleague with the Dutch government. They also knew I would return to Ghana to interview their children’s caregivers, and that assured them of the independence of our research and made it easier for us to reach a large number of our respondents. Another advantage afforded by my language abilities was that it allowed me to make observations because I understood conversations that took place on the phone with the migrants and also the conversations that took place in the home of the caregivers. I also know about child-raising in Ghana both anecdotally and experientially, having been born and bred in Ghana. Furthermore, my parents had foster children throughout my youth and, therefore, I can identify with what most of the literature says about child fosterage. In addition, I fostered two children in the course of my study under two different circumstances and could, therefore, identify with what the literature and my own research findings present about child fosterage. Finally, I am Ghanaian, and even though I may not qualify as a migrant owing to the purpose and duration of my stay abroad, I have had the experience of travelling and leaving my children in the care of my mother and have had to make all the kinds of arrangements and decisions that migrant parents who leave their children in the home country make. I have tried mothering from afar, sometimes checking and instructing what dress my daughter should wear to attend an event at school or feeling upset and helpless about my son being given a haircut against my wish. I have had to call my daughter’s school once and have had to keep in touch with all the people I consider helpful in caring for my children while I am away. I am also often faced with the headache of getting gifts, no matter how small, for everybody who contributed to my children’s upkeep while I was away. On my last trip, three of the migrant parents had asked me University of Ghana http://ugspace.ug.edu.gh 86 to send gifts to their children back home, and I watched them as they struggled to get a gift3 for everyone, especially for the caregiver’s own children and for members of the caregiver’s household who, they may have heard, may have been helping the caregiver with the children. I am also very certain that if my mother were to be interviewed, she would have expressed similar opinions to those of the caregivers in my study. I know this because I am very much involved with her while I am away. It has been stated that when the researcher is an insider, s/he has the greater advantage of having easy access in the field and obtaining data, but also the disadvantage of subjectivity. The major challenge I had as an insider was the tendency to take a lot of information for granted or at face value. I assumed that I knew what the answers to certain situations were and, therefore, I skipped asking certain questions. A good example was the caregiver's motivation for taking care of a migrant's child. I found it difficult to ask some caregivers, especially grandmothers, what their motivation was because I knew the answer. Many Ghanaian maternal grandmothers will not allow their daughter to leave her grandchildren in the care of anybody else while she is still alive. When I informed my mother about my trip to the Netherlands, she immediately told me to leave the children with her. With this understanding of a grandmother’s position on the care of their grandchildren, I found it unnecessary to ask grandmothers what their motivation was for taking care of their grandchildren. Sometimes when respondents gave certain responses to my questions, I would not ask them to elaborate because I assumed I knew the explanation to be given. It took the ‘cultural naivety’ of my Dutch colleague, through exchanging fieldnotes, to 3 I made a list of gifts that migrants sent to their children, caregivers and others in their network (see appendix) University of Ghana http://ugspace.ug.edu.gh 87 help avert this problem. My colleague considered the information I found ‘normal’ to be interesting and wanted more explanation of it. I realized then that the explanation I gave her most of the time was mine and not the respondents'. This made me go back to probe further and ask for more elaboration on responses, even when I assumed I knew what the answers would be. I ended up finding out that even though the explanations were not far from what I had in mind, there was always something new that I would not have found out had I not asked. According to Mullings (1999), Merriam et al. (2001) and Teye (2012), the ‘insider’- ‘outsider’ position of the researcher shifts, and, depending on the situation in which the researcher finds her or himself, s/he can be either an outsider or an insider. There were times when I considered myself an outsider such as when the respondents were not sure what my intentions were and for whom I was working. It also took longer to gain my ‘insider’ position with some interviewees than others, although for the most part, I was an ‘insider’. 4.9 Reflection: The Dilemma of an Ethnographer Starting as outsiders to a field setting, many fieldworkers find themselves pulled toward involvement as insiders in ways that make maintaining a research stance difficult (Emerson et al. 2011: 41). Although I considered myself an ‘insider’ both in Ghana and in the Netherlands for the greater part of the research period, this was not the case at the beginning of my research with the caregivers in Ghana. Because almost 50 per cent of the parents in the matched samples had undocumented status in the Netherlands, I was not easily accepted as an ‘insider’ in this respect. Additionally, it did not help when the caregivers learned that I had been to the Netherlands and was conducting my research together with a Dutch researcher. In my quest to build rapport and University of Ghana http://ugspace.ug.edu.gh 88 trust, I had to work very hard by constantly keeping in touch in several different ways which I have described earlier, an experience that corroborates the quotation above. In some cases, this worked so well that, as caregivers started drawing me into their worlds, the boundaries between being the researcher and the ‘friend’ (the insider status I had worked hard to earn), sometimes became blurred. There were times when I felt that in studying them I was exploiting them. Sometimes on informal visits, they gave me information without my soliciting it, and other times some incidents became difficult to write about because documenting them would have involved betrayals or resulted in revelations that I personally found ethically intolerable (Emerson et al. 2011). Furthermore, in the host country, parents who felt grateful that I had brought messages from their caregivers and children drew me into personal conversations about the care arrangement. I became very much aware of the fact that they did this because they knew I was not the one researching them. There were also times when caregivers, children and parents drew me into mediating roles. Children often sent me messages to give to their parents, so I sometimes videotaped their requests in order to provide a direct representation of the children’s message. Caregivers, on the other hand, have asked me on several occasions to advise children who they believed were going wayward or were not taking schoolwork seriously enough. I have also had some parents ask me to explain their living conditions to the caregivers and children back home so that they will not become suspicious when they did not receive enough remittances. There have also been instances where children sought my opinion on what to do about a situation with a caregiver. In their bid to get through to me to deliver their messages and perform these intermediary roles, I often got some valuable information on the care arrangement. For instance, I might be informed about how caregivers felt about the migrants and vice versa. University of Ghana http://ugspace.ug.edu.gh 89 Heyl states that “[E]motions are a critical part of an ethnographic interview and could sometimes create discomfort and ethical dilemmas,” (2001: 374). I had my fair share of these kinds of experiences during the research period, and that was one of the burdens which the SMS methodology laid on me. Knowing what was going on ‘here’ and ‘there’ had the effect of making the burden greater. I met caregivers who felt the migrants were deliberately ignoring their concerns, did not appreciate their efforts and had burdened them with the care of the children. Meanwhile, having met the migrants myself and witnessed the situation in which they had found themselves in the host country, I knew that their actions were not deliberate but, since it was not my place to say anything about it, there was little I could do to help dispel the caregivers' anxieties. To cite one example, there was a caregiver who happened to be the wife of a migrant, but who had had a quarrel with the migrant. This problem had begun some 6 months before I started my field research, and it was still not resolved even at the time the research ended. Having made contact with both sides and listened to both of them separately, there were times when I had come very close to letting them know what they individually thought of each other and where their marriage was headed. I also came into contact with children who were making excessive demands on their parents or who thought their parents were being mean because they did not meet all their needs. Having witnessed the conditions in which their parents lived, I had often come very close to telling them the truth, but this would have breached confidence. There was another incident involving another caregiver-wife whose husband had been imprisoned for being undocumented. I got wind of this information when I went to the Netherlands. I had just had two sessions with the caregiver and was still establishing a rapport with her when the incident occurred. This incident presented quite an interesting situation University of Ghana http://ugspace.ug.edu.gh 90 because I saw in it an opportunity to learn firsthand what happens when something goes wrong in the care arrangement. However, I was also very reluctant to keep visiting them under the persistent state of affairs. I was sure there was no way I was going to pull out some paper to write down something when I visited, and yet I knew that without a pen and notebook she would consider me ‘a friend’ and give me information in that capacity. It was a big dilemma, and I did not want to go because I did not want to appear as if I was exploiting them. I decided to go, and, as expected, she narrated the whole incident: she mentioned people both in Ghana and the Netherlands who had been helpful through difficult periods and how she had rallied support both in Ghana and the Netherlands. This was all very useful data. I obtained more names that had not come up during the name generator exercise because these people came in to assist in the period of crisis. Dealing with the dilemma To manage this type of dilemma, ethnographers are advised to step aside from the relationships, remind themselves of their commitment to the research and strike a balance between their commitments on both sides. In my case, as noted earlier, I declined requests to spend lengthy periods with the respondents such as a weekend stay-over in their home. Additionally, in Accra, after establishing the needed trust and rapport, I cut down on a lot of the niceties which I had used in the beginning. Finally, to best deal with the situation, I used a pseudonym for all the respondents and sometimes changed the location, particularly the destination areas of the migrants, so that the respondents could not be traced through the sequence of events narrated in the study. In this way, I upheld the confidentiality of the respondents and the data, while still being able to use the relevant information for my analysis. University of Ghana http://ugspace.ug.edu.gh 91 4.10 Study Sites The Greater Accra and Ashanti Regions were selected for the study. This selection was made for two main reasons: First, these regions have been cited as the regions with the highest number of Ghanaian emigrants (GSS, 2012). The nature of this study and the use of the SMS methodology where we searched for migrants first before matching them with others in Ghana meant we did not have the luxury of working in other areas in the country, which reflected only a minority of the migrant population in Europe. Thus, it was easier to find respondents from these two regions in the Netherlands. Furthermore, because I was conducting an ethnographic study and needed to spend extensive time with the respondents, I could not spread my selection of respondents across too many regions and, therefore, had to settle for these two, which contained the majority population of respondents that I needed for our kind of study. Second, most anthropological studies on child-raising conducted in Ghana have taken place in rural areas and, for this reason I estimated that having one with a focus on an urban context was appropriate since the urban areas were likely to present different dynamics. Below is a brief over-view of the study areas. 4.10.1 Accra Metropolitan Area The Accra Metropolis falls within the Greater Accra Region. It has a total land size of 200 sq. km. Accra has been Ghana’s capital since 1877. It is the seat of government, and it also hosts the headquarters of the Ministries, Departments and Agencies (MDAs), thus making it the most well-endowed city in Ghana with the best hospitals, banks, universities and other educational facilities (Spio 2011). According to the latest Housing and Population Census held in 2010, the Greater Accra Region was the second most populous region in the country, with a population of more than 4 million or 16.3 per cent of the total population. During the colonial era, there was more development in the Southern sector of the country of which Accra is a part at University of Ghana http://ugspace.ug.edu.gh 92 the expense of the Northern sector. This still persists today and has engendered a lot of migration from the underdeveloped Northern parts to the south, especially to Accra. According to the GLSS 5 report (2008), the Greater Accra Region has the largest number of in-migrants, with about four in every ten residents being in-migrants. Migration accounts for almost 35 percent of Accra’s population. Most of the in-migrants are found in Accra and a few in the rural and peri- urban areas surrounding the city. For some time now, the increasing rate of population growth in the city has earned it the reputation of one of the fastest growing cities on the sub-Saharan continent (Spio, 2011). The economically active population of the metropolis is estimated to be 823,327. The economy consists of the Primary Sector (farming, fishing, mining and quarrying); Secondary Sector (manufacturing, electricity, gas, water, construction); and Tertiary Sector (wholesale trade, retail trade, hotel, restaurant, transportation, storage, communication, financial intermediation, real estate service, public administration, education, health and other social services). As an urban economy, the service sector is the largest, employing about 531,670 people (64.58%) (ISSER, 2013). The second largest, being the secondary sector, employs 22.34% of the labour force (that is 183,934 people). Yet, it has 114,198 of its labour unemployed - rate of 12.2%. Despite the Metropolis’ ability to employ some of its labour population, it also has a high unemployment rate of about 114,198 representing 12.2% (www.ghanadistricts.com, retrieved on 10/09/2013). The rapid growth of Accra’s population over the years has placed an increasing demand on its infrastructure, which has not expanded proportionally to population growth. The demand for services such as education, health, decent housing and portable water is also on the rise. University of Ghana http://ugspace.ug.edu.gh 93 Another notable feature of Accra's increasing urbanization and modernization is the expansion of distances between locations that were otherwise deemed very close to each other. Accra's diverse population deploy a variety of schedules with respect to work, school, social activities and operating hours for business, which create a diverse and high demand for transportation services that the existing transport services are unable to meet. This has increased traffic congestion during rush hour, coupled with the small number of buses available on some routes that makes movement in and around Accra, a rather unpleasant chore. Urban poverty has also increased with the increasing urbanization. Although Ghana has been touted as having attained a middle-income economy, and Accra has been tipped as one of the fastest growing cities in sub-Saharan Africa as well as one of Africa’s fastest growing cities for millionaires, poverty remains in Accra, especially among the low-educated, low-skilled population, the majority of whom are immigrants who have fled to Accra to seek greener pastures. By migrating in this manner, they increase Accra's poverty level. According to a UNICEF (2012) report on poverty in Ghana, poverty rates across Ghana had fallen considerably, however in Accra they have increased. Below is a map of Accra, indicating the various locations of my interviewees. University of Ghana http://ugspace.ug.edu.gh 94 Figure 4.2 Map of Accra showing study areas 4.10.2 Kumasi Metropolitan Area Given its strategic location and political dominance, Kumasi developed into a major commercial centre where all the major trade routes converged. However, it came under the influence of British rule in 1890 (Adu Boahen 1986). With time, the city began to expand and grow, thereby, making it second only to Accra in terms of land area, population size, social life and economic activity. Its strategic location has also endowed it with the status of being the principal transport terminal, assuring its pivotal role in the vast and profitable distribution of goods within the country as well as beyond. University of Ghana http://ugspace.ug.edu.gh 95 Seventy-nine per cent of the regional population, aged 15 years and older, is economically active. The Metropolis is made up of formal and informal sectors. Kumasi's formal sector is characterized by businesses with corporate ownership and large-scale operation. The informal sector is 'a confusing maze of thousands of tiny workshops and enterprises producing everything under the sun, with a complicated distribution and communication network at their disposal' (www.ghanadistricts.com, retrieved on 10/01/2013). Nevertheless, it contributes much to Kumasi's economy. The main sectors of its economy fall under the category of Trade/Commerce/Services (ibid) which accounts for about 71 per cent of the economy. Manufacturing/Industry makes up 24 per cent and the Primary Production sector only 5 per cent. The service sectors absorb most of the active labour force because they consist of an integrated system of markets at Adum, the Central Business District and Kumasi Central Market, which is the single largest market in West Africa with linkages to satellite markets in other parts of Kumasi. In addition to these, other services such as banking, insurance, transportation, hotels, restaurants and traditional caterers and other tourist sites are also found in the city, thus contributing to its establishment as a major commercial centre. The positioning of the area as a centre for commercial activities accounts for the large- scale immigration into Kumasi. According to the latest Population and Housing Census (GSS 2012), about two-thirds (65.7 per cent) of the population in the Ashanti Region were born in the place where they were counted in the census. The remaining one-third (34.3 per cent) consists of immigrants. Indeed, migration has been a key factor in the phenomenal growth in Kumasi population which reached 60.8 per cent in 1960 and was 53.1 per cent a decade later. In 2000, it slightly reduced to 48.6 per cent, but migration again pushed it upwards. In 2005, the rate grew to 57.4 per cent and in 2009 it remained little changed at 56.7 per cent. University of Ghana http://ugspace.ug.edu.gh 96 Kumasi's strategic location and its status as the main administrative hub and brisk commercial centre have helped make the city a popular destination for both internal and international migrants. Like Accra, Kumasi attracts a number of migrants from several parts of Ghana as well as from neighbouring African countries such as Togo, Burkina Faso, Mali, Nigeria, Cote D’Ivoire and abroad, especially Europe. Although many people in Kumasi are engaged in some form of employment, about 60 per cent of its residents continue to have a lower standard of living as a result of their low level of income. In addition, the proportion of unemployed in the city is currently 16.0 per cent. The unemployment rates are more pronounced in the Metropolis than in the remaining parts of the region The pursuit of non-existent jobs has led to a lot of out-migration, both internally and internationally, especially among the young and economically active. According to the latest Population and Housing census (GSS 2012), the Ashanti Region has the highest number of Ghanaians living abroad and the highest number living in Europe. Most of the respondents from the Ashanti Region were drawn from the Kumasi Metropolis, except three who came from other parts of the region including Dompoase, Offinso and Sekyere Banko. The locations of the respondents are indicated in the map below. University of Ghana http://ugspace.ug.edu.gh 97 Figure 4.3: Map of the Ashanti Region Showing study Areas University of Ghana http://ugspace.ug.edu.gh 98 CHAPTER FIVE: PROFILE OF THE CAREGIVERS 5.0 Introduction The aim of this chapter is to present and discuss the profile of the caregivers in the study and to highlight some characteristics that are important for the analysis in the subsequent chapters. Additionally, I also highlight important issues that emerge from the discussion on caregivers’ characteristics and how they contribute to the existing literature on TCRAs. The chapter has two main sections. The first section presents characteristics of the 34 caregivers, namely: age, sex, relation to the child, marital status, number of their own children, economic status and the characteristics of the children in their care. The first section begins with two tables. Table (5.1) presents the profile of caregivers in the matched sample (n=15) and Table 5.2 profiles the caregivers in the non-matched sample (N=19). The second section highlights three categories of caregivers who have special characteristics: the male caregiver, the stay-behind wife (special kin), and the paid caregiver. These categories of caregivers are highlighted because they present different sides to the studies on caregivers in TCRAs: male primary caregivers are highlighted because they constitute a contrast to the parenting norm; the live-in paid caregiver is selected because they constitute an emerging phenomenon; while the stay behind wife have different experiences because the child and the migrant in the TCRA are their biological child and husband respectively. I conclude the chapter by discussing some important themes that emerge from the discussion on the characteristics of caregivers, namely power relations in caregiving and the gendered notion of care. Tables 5.1 and 5.2 present the characteristics of caregivers, the children in their care and some data on the migrant parents. University of Ghana http://ugspace.ug.edu.gh 99 Table 5.1 Profile of caregivers (CG) in the Matched Sample CG Age CG Children (HH) Migrant Child (sex) Age Sibship Size Previous CGs Duration of Stay with CG Mother 42 2 Father Male 14 3 - 14 Mother 48 2 Father Female 16 3 - 16 Father’s nephew 34 - Father Male 19 1 2 2 Father’s wife’s brother 36 - Father Male 17 2 1 8 Mother 44 4 Father Male 14 5 - 14 Father’s mother 73 *1 Father Male 14 1 1 3 Mother’s sister 38 - Father&Mother Female 19 2 1 8 Mother’s sister 40 *4 Father &Mother Female 14 1 - 14 Mother’s mother 71 - Father &Mother Male 19 2 - 18 Sister 26 2 Father &Mother Female 14 1 1 3 Mother’s mother 58 3 Mother Male 16 2 - 14 Father’s mother 76 1 Father&Mother Female 16 2 - 15 Mother’s nephew 46 - Mother Female 18 1 1 2 Mother’s mother’s sister 64 1 Mother Female 17 5 1 2 Mother’s mother’s mother 80 - Mother Female 12 1 - 10 Table 5.2 Profile of caregivers in the non-matched Sample CG Age CG Children (HH) Migrant Child (sex) Age Sibship Size Previous CGs Duration of Stay with CG Mother’s mother 77 - Mother Female 15 3 - 8 Mother’s mother 68 - Mother Female 16 2 - 5 Father’s mother 63 - Father&Mother Male 14 1 - 7 Mother’s mother 76 - Mother Female 14 3 - 8 Mother’s sister 55 1 Mother Male 16 3 - 9 Father’s sister 51 3 Father&Mother Female 16 1 - 2 Father’s brother 59 - Father&Mother Male 15 1 - 7 Female Non-kin 52 2 Father Female 17 1 1 2 Mother’s mother 58 2 Mother Female 14 8 - 14 Mother’s Father 58 - Father&Mother Male 11 2 - 4 Mother’s mother 56 - Father&Mother Female 14 4 - 6 Mother’s sister 34 1 Mother Male 19 1 2 2 Mother’s Male Friend 34 - Mother Male 16 3 1 3 Female Non-Kin 52 2 Father&Mother Male 11 1 1 1 Female Non-Kin 35 - Father&Mother Male 13 3 1 1 Mother’s mother 51 1 Father&Mother Female 15 2 - 12 Mother’s mother 57 1 Mother Male 16 1 1 10 Father’s mother 73 - Mother Female 16 2 - 12 Mother’s mother 65 *1 Father&Mother Female 19 3 1 6 Notes: HH (living in same household as the migrant’s child) *Caregiver’s fostered child University of Ghana http://ugspace.ug.edu.gh 100 The caregivers were in different types of TCRAs depending on who had migrated. There were those in TCRAs where the migrants were mothers (n=13), fathers (n=7) and both mothers and fathers (n=17). The total number of female and male migrants whose children were with caregivers was 30 and 24 respectively. The large number of female migrants in the study supports the notion that international migration of women is on the increase. It also accounts for the large number of caregivers who were not the other parent of the child and supports the observations that when women migrate without their children, they rely more on others mostly women and less on the biological fathers of the children to take over their roles as primary caregivers of their children. None of the caregivers in the 13 TCRAs where only mothers had migrated was the other parent of the child which contrasts the TCRAs where fathers were migrants in which 3 out of the 7 caregivers were the other parent of the child. This phenomenon could be explained by gender role division in the home where women were the primary caregivers of children. However, Asima (2010) found in his study that more fathers were willing to assume the responsibilities of primary caregivers when their wives migrated in lieu of sending the children to female relatives. The fathers played this role with or without the assistance of women. Thus, gender role division in the home cannot adequately explain why women and men sometimes choose not to leave the children with the other parent. I observed that almost all the women and the four men whose children were with other caregivers were no longer married to the other parent of the children they left behind. The three men in my study and women and men who left their children with the children’s other parents in Asima’s study were still married to them. Divorce either before or in the course of migration is therefore one possible explanation for migrants’ need for caregivers other than the other parent of their children in TCRAs. University of Ghana http://ugspace.ug.edu.gh 101 The tables also provide some information about the children being raised by these caregivers. These include the sex and age of one of the children and the sibship size to show that some of the caregivers had more than one child of migrants in their care. As part of the SMS methodology only one of the children was selected to be in the matched sample, but during interaction with children and others in the caregivers’ household, I also included the other children of the migrants when there was the need. The experiences of caregivers were thus in relation to all the children and not just the one selected. It is also for the same purpose of a possible interview with children that the age provided in the tables ranged from 11 to 19 even though some caregivers had children younger than 11 years in their care. The age of the children vis-à-vis the duration of stay with caregivers implies that some of the children had been with caregivers for a minimum of a year and a maximum of 18 years. I noticed that children stayed for the longest duration in TCRAs where grandmothers were the caregivers and in these TCRAs caregivers were seldom changed. Some studies ( Moran-Taylor, 2008; Dreby 2010) on child fostering practices and also on transnational families have asserted that grandmothers were the most preferred choice of caregivers for migrants’ children. The assertion is supported by the large numbers of grandmothers in my study and the duration of time the children stayed with them. Later in this chapter, I will show why migrants prefer grandmothers of the children to others. The tables also provide information about the number of caregivers the children had been with before they were sent to the present caregiver. Moving a child from one caregiver to another or what Akesson et al (2012) refers to as re-arranging care in the course of parental migration is normal in TCRAs and is necessitated by many factors. I examine some of the factors in chapter six (6). University of Ghana http://ugspace.ug.edu.gh 102 The other characteristics featured on the table are discussed in more detail in the subsequent sections. 5.1 General Characteristics of the Caregivers 5.1.1 Kin Caregivers Kin caregivers constituted the majority (88.6 per cent) of the caregivers. There were 28 kin caregivers in this study. They were predominantly maternal (n=21) and to a lesser extent paternal (n=7). The choice of maternal relatives was usually made in the case of a migrant mother or when both the mother and father migrated. This can partly be explained by the fact that most of the people in my sample were from the Akan speaking group in Ghana who are matrilineal. It was very rare to find children entrusted to paternal relatives when the migrant was a woman. Children were mostly left in the care of paternal relatives when fathers had migrated especially if they were no longer married to the mothers of the child/ren that were left behind. Usually in these cases, the children were either living with the fathers or fathers’ relatives prior to migration, or in some cases they were moved from their mothers to the fathers’ relatives during migration. The latter was most prevalent when the mothers of the children had remarried and the fathers could not trust the children with their step fathers or as in the case of Georgina, her migrant son was not comfortable with the living conditions of the child’s mother. Georgina, the paternal grandmother, explained that her grandson was moved from his mother’s house into her care because her migrant son could not trust that remittances he sent to the child would be used solely for the child. She added that the child was previously living in a compound house with his mother and several members of her extended family. In a few cases (4), both parents had travelled and left the children with the father’s relatives. One key reason why the women agreed University of Ghana http://ugspace.ug.edu.gh 103 for the children to be left with paternal relatives was the availability of quality schools in the areas where they lived, as opposed to that of the maternal relatives. Grandmothers were the majority in this group. They were 16 or 45.7 per cent of the total number of caregivers in the sample. These also constituted more than half of the caregivers who were kin, and included 11 maternal grandmothers, one maternal great-grandmother, one maternal grand-aunt4 and three paternal grandmothers. The maternal relatives also included sisters of the migrants, that is, the maternal aunts of the left behind child. There are different reasons why grandmothers are preferred caregivers. In a study on Mexican transnational families, it was indicated that grandmothers consider taking care of their grandchildren as an extension of their motherly roles towards their adult children and are therefore more sacrificial in their endeavors than most of the caregivers who fall into the other categories (Dreby, 2010). Grandmothers also explained that their migrant children ended up ‘killing two birds with one stone’ when they left children in their care as they were able to fulfill their care obligations towards their own mothers and provide for their children at the same time. Furthermore, grandmothers tended to have fewer concerns, such as raising other children or actively engaging in economic activities which competed for their attention. The biggest disadvantage with this arrangement, however, is when old age and ill health sometimes impeded the provision of child care (see chapter 6). There were 5 maternal aunts in addition to the 13 maternal grandmothers. This is more in line with what Dolphyne describes as ‘mothering,’ that encompasses a collective responsibility for the welfare of each child through co-mothers, grandmothers, aunts, sisters and community 4 She is categorized as a grandmother because in Ghana there are not separate terms for grandparents and their siblings. They were all referred to as grandmothers or grandfathers. In the matrilineal system especially there are no separate terms for mothers and their sisters. They are referred to as mothers so a child would refer to her mother’s sister as ‘my mother’ University of Ghana http://ugspace.ug.edu.gh 104 people, especially during long periods of separation from parents (Dolphyne, 1991). Other kin caregivers (from both sides) included a paternal aunt, a migrant’s brother-in-law, a maternal nephew, a paternal nephew, a child’s older sibling, a paternal uncle, and a maternal grandfather. 5.1.2 Non-Kin Caregivers There were four caregivers in my study who were not kin, reflecting the notion prevalent in several communities in Ghana that a child is born to an entire community and must not necessarily be related to another person by blood to be raised by him or her. As is well known and documented, there are fosterage situations in which parents send a child to somebody of higher status or influence in society, because they want their child to learn his/her vocation or to attain their level of expertise in a particular field, in the expectation that this would shape their child's future in a positive way, and a teacher or priest is often sought out for such a purpose (Goody, 1982). In this study, three out of the four non kin caregivers were female. Two of them were teachers, which is similar to the traditional practice. Interestingly, the parents did not know the teachers prior to the care arrangement. In one of the cases, the teacher was approached based on the recommendation of the migrant’s friend, and in the second instance, the child knew the teacher and asked her father to let her stay with the teacher when her own mother died. The third non kin caregiver was a friend of the migrant, while the fourth was hired as a live-in caregiver. She did not know the migrants prior to her employment and it was her uncle who informed her about the migrants’ intention to hire a caregiver. This caregiver is discussed in detail later in this chapter University of Ghana http://ugspace.ug.edu.gh 105 Figure 5.1 Age distribution of caregivers Majority of the caregivers were 60 and above years of age and most of these were grandmothers of the children. The youngest caregiver was 26 years and she was an older sibling to the child. Considering the age of most of the grandmothers and the duration of children’s stay with them, it is apparent they started taking care of these children when they were younger and more physically active. There were also others who started providing care for children when they were already old and quite unfit physically. In the next chapter I will discuss how the age of the children vis-à-vis that of the grandmothers could create problems for some of the TCRAs and also the benefits that some grandmothers enjoy from the children they had raised into young adults now that they are aging and becoming frail. 5.1.3 Economic Status of the Caregivers Most of the caregivers in my sample were economically active. A majority of them had their own sources of income and were not dependent on the remittances of the migrant for their University of Ghana http://ugspace.ug.edu.gh 106 personal living expenses. This situation stands in contrast to Hochchild’s (2000) work where caregivers were employed or requested by the migrant to take care of their children in exchange for a means of sustenance and were, therefore, dependent on allowances from them. Historically, Ghanaian women have been economically active and when married, remain financially independent from their husbands (Clark, 1999; Fayorsey, 1999; Oppong, 2006). Most of the caregivers (n=21) were already working. A majority of them (n=29), including some of those who were not economically active, had other sources of income such as financial support from their spouses or from their adult children. There were five caregivers who depended solely on migrants’ remittances for theirs and the children’s upkeep. These were four grandmothers and one maternal aunt who quit her work as a hair dresser so she could take care of her sister’s children. Most of the caregivers, especially grandmothers who were not working at the time of the interviews, said they had been working prior to that time and had only recently retired as a result of ill health or old age. Some of these also had to give up their work entirely because the children were very young--a few of the children were as young as four-months at the time they were left in their care--and they could not combine their work and the care of the children. Some of the caregivers especially the grandmothers, who were directly involved in economic activities employed house-helps to help them at home so they could keep working as they needed the income from their work to survive. According to them, they did not receive remittances especially during the initial period after migration when their migrant relatives were establishing themselves. In such cases, it was necessary for them to continue with their economic activities to ensure their survival. Table 5.3 presents caregivers economic activities. Out of the 21 who were engaged in some form of income-generating activity, there were nine who were formally employed and earned University of Ghana http://ugspace.ug.edu.gh 107 monthly salaries. The remaining 12 were self-employed, with the majority working in petty commerce. Petty commerce, as defined in this study, includes business activities that fall below the Ghana’s National Board on Small Scale Industry’s (NBSSI) definition of micro enterprises under the micro, small and medium Enterprises. Micro enterprises are those that employ between 1 and 5 people with fixed assets not exceeding $10,000 excluding land and building (NBSSI, 2014). Per this definition, only one of my respondents partially fell under the micro enterprise with capital investments of GHc 50,000($17,000)5, one regular sales girl and two casual labourers. None of the remaining respondents had businesses which met these specifications. I have therefore placed them under petty commerce based on the general definition of petty commerce as a trading activity that is conducted on a small scale, or sale of small inexpensive items, and my own observation of what constitutes petty commerce in Ghana, which includes selling items on a table-top, in a small shed at the market, in one's home or head-carried on a tray to be sold from door-to-door. The capital investments of my respondents’ trading activities ranged from GHC 100 to GHC 2,000. Table 5.3 Caregivers’ economic activity # Caregiver Care giver’s sex Relation to child(ren) Type of economic activity 1. Naki Female Mother Petty commerce 2. Pomaa Female Mother Petty commerce/ farmer 3. Takyi Male Father’s brother-in-law Auto- mechanic 4. Kwesi Male Father’s nephew Fabricator 5. Awurakua Female Mother sister Petty commerce 6. Fosua Female Father’s mother Micro Enterprise 7. Tettey Male Mother’s nephew Civil servant 8. Christie Female Mother’s sister Civil servant 9. Akua Female Mother Seamstress/petty 5 She said the amount was the estimated cost of goods in her shop at the time I visited. Some of them items had been paid for out rightly and a part of them was supplied on credit. University of Ghana http://ugspace.ug.edu.gh 108 commerce 10. Hajia Female Mother’s mother Petty commerce 11. Theresa Female Mother’s sister Petty commerce 12. Maame Ama Female Father’s sister Petty Commerce 13. George Male Father’s brother Civil servant 14. Mama comfort Female Non-kin Teacher 15. B. Manu Male Mother father Head teacher 16. Abla Female Mother’s sister Research assistant 17. Koo Frimpong Male Non- Kin Petty Commerce 18. Margaret Female Non-Kin Teacher 19. Ophelia Female Non-kin Paid caregiver 20. Boadi Male Father’s brother University Librarian 21 Kate Female Mother’s mother Petty commerce Source: field data (2011/2012) 5.1.4 Caregivers’ Marital Status and own Children. Twenty-one out of the 34 caregivers were married. Of these, 17 lived with their husbands in the same house as the migrants’ children. Four out of the 21 married caregivers were male and the remainder female. All five maternal and paternal aunts were married, and all, except one, lived elsewhere with the migrants’ children. Only two of the grandmothers were still married at the time of the research. The remaining unmarried grandmothers were mostly widows. There were 18 caregivers who had children of their own and they lived in the same house as the migrants’ children. Three of these were left-behind wives. Of the 16 who did not have their own children in the house, three did not have any children of their own, whereas in the case of most of the grandmothers, their biological children were adults who lived elsewhere. 5.1.5 Caregivers’ own upbringing As part of this study, I conducted life history interviews with the caregivers. These interviews focused more on the upbringing of the caregiver. I was particularly interested in those who had also been fostered as children or whose parents had fostered other children when they were growing up. I was interested in their upbringing to help map the changes in child fosterage from University of Ghana http://ugspace.ug.edu.gh 109 their own experiences as/with foster children and currently as foster parents themselves, and to see to what extent the changes can be attributed to transnational migration. Twenty-eight out of the 34 caregivers had been fostered as children. All of these, with the exception of three, were fostered by non-kin. Most of the fostered ones lived as house maids to rich market women who were known to their parents. The three, who were fostered by kin, either moved to the city to live with a rich relative or to nearby towns or villages to help with farm work. One of them was fostered together with her brother by her maternal grandmother because she was lonely and needed company after her children had all left home to start their families. They were fully sponsored by their grandmother even though their parents could afford to take care of them. One caregiver mentioned that her father and uncles ‘exchanged’ her and her siblings, (i.e., each sibling spent a considerable period of their youth in different uncles' houses) to help familiarize them with their cousins and other family members so they could live as one big family. She said her father and her uncles were wealthy and, therefore they could support each other’s children when they were in their care. The third caregiver in this category said she was fostered by her maternal uncle following her parents’ divorce. She later moved to her paternal relatives when her uncle died. The other fosterage situations involving non-kin were characterized by poor parents who chose to send their children to relatives who were in a position to help them. Eleven of these were fostered at an early age (before they turned 10-years- old), while the majority of them left their parents’ houses in their teens. Some of them changed foster parents during their fosterage period for different reasons, such as the death of a foster parent, maltreatment by a foster parent or unfulfilled promises. The caregivers mentioned that they have observed some changes in child fosterage, particularly in TCRAs. These changes have been facilitated primarily by the parents’ ability to University of Ghana http://ugspace.ug.edu.gh 110 send remittances to cover their children's living experiences and establishing a communicative environment, which among other things guarantees their active involvement in their children’s lives. 5.2 Some Specific Characteristics of Caregivers In this section, I highlight some characteristics of caregivers that present different kinds of insights into the literature on care giving within Ghana and across borders. These are male caregivers, stay behind wives, and live-in paid care givers. Males as primary caregivers have been singled out because care has and is still largely considered a female sphere, and men who are directly involved in care giving usually do so with the help of a female (Asima, 2010). Wives who stayed behind to raise the children when their husbands travelled experienced the role quite differently from that of other caregivers owing to their conjugal and parental obligations towards the migrant and the children, respectively. Finally in this section, I present the live-in paid caregiver who was hired to take care of the migrants' children. 5.2.1 Male Caregivers Women have been described as primary caregivers of children. This is true in most transnational child-raising arrangements, and studies on transnational parenting have mentioned that migrant mothers often rely on other women to take care of the children that they leave behind (Hondagneu-Sotelo and Avila, 1997; Schmalzbauer, 2004; Parrenas, 2005; Dankyi, 2011). This observation has also been corroborated by the findings of this study, where women comprise more than 80 per cent of the caregivers. In this section, I dwell more on the male caregivers, mainly because there has been very little attention paid to them in studies on transnational families. Some studies have mentioned men as caregivers. However, these were University of Ghana http://ugspace.ug.edu.gh 111 primarily stay-behind husbands who took care of the children with the help of female relatives when the wives migrated (Parrenas, 2005; Asima, 2010). In this study, there were six male caregivers who were not stay-behind husbands, with four of them not having females to assist them with the day-to-day care of the children. Five out of the six males were kin caregivers. Two of these male caregivers were assisted by females; one was assisted by his wife and the other by a female relative. In these arrangements, the man was in charge of making major decisions concerning the children’s education and health, disciplining the children and receiving and managing the remittances for the children’s upkeep, while the women took care of feeding, bathing and other day to day chores for the children. One of the two was the maternal grandfather of the children, and he was raising these children together with his wife. I interviewed both him and his wife, but used him in the analysis because of his involvement in the care arrangement from the beginning. He said that his daughter first sought his approval of her decision to bring the children to Ghana before discussing it with her mother. He was the one who received remittances on behalf of the children and made most of the major decisions concerning their education and health. He also mentioned that he was a hands-on ‘grandpa’: He played with them all the time and helped with their school work to relieve their grandmother whenever he returned from work. His job as an accountant in an educational institution was not time-consuming, and it gave him a lot of time to help with the care of his grandsons. The other caregiver, Boadi, is an elderly man nearing 60, who is taking care of his younger brother’s son. The boy was sent to him at age 8 when his parents moved abroad. He looked after the child with the help of his cousin’s wife who was living with them, as his own wife lives in another part of the country. At the time of the research, the boy was in Senior High School and was a boarder, but he lives with Boadi during the school University of Ghana http://ugspace.ug.edu.gh 112 holidays. Takyi, another caregiver, is the brother of the migrant’s wife and is raising his sister’s step-sons in Kumasi. He is in his mid-thirties and is a bachelor who works as an auto mechanic. The first boy was sent to him three years before the second. According to Takyi, the migrant called from Amsterdam to inform him about his intention to relocate his children from the village to the city where they could have access to quality education. He then requested Takyi to take the children into his care. The first boy was barely 10 years old when he was sent to Takyi. About three years later, the second boy, then 11, was sent to him. Takyi was in charge of the day to day care of the children. He started teaching the older boy to cook and wash his own clothes when he was about 13, at which time the younger boy had joined them. The older boy could therefore help with some ‘light chores’ while Takyi performed the more substantive chores. The situation with Victor and Kwesi was similar. Their charges were brought to them when they were in their late teens (16 and 17, respectively). Victor, a civil servant, lived in the migrants’ house as a care taker and preformed tasks for the migrant as and when necessary. The migrant sent his daughter to live with Victor when the grandmother with whom she had been living died. Kwesi, on the other hand, is a self-employed steel fabricator, who took care of his maternal uncle’s son. He used to live in his uncle’s house and this was the reason his uncle sent his son to live with him. At the time of the study, Kwesi had moved out of his uncle’s house and lived elsewhere with his wife and two young children. Even though his uncle’s son remains in his father’s house, he spends virtually every day in Kwesi’s house. Victor and Kwesi did not have to provide nurturant care for the children because they were older and could take care of their personal grooming. In Victor’s case, his aunt’s daughter was the one who did the cooking and cleaning while Victor provided money for daily feeding and payment of utility bills. Kwesi’s University of Ghana http://ugspace.ug.edu.gh 113 cousin in his care ate food prepared by his wife everyday, while Kwesi has to ensure that the boy concentrates on the apprenticeship in which he has been enrolled. The last caregiver in this category is Koo Frimpong, who presents an interesting case which in many ways is counter to popular notions of care-giving. As has been noted above, in most cases in which there are male caregivers, they are aided by a female helper, or the children are old enough to take care of themselves, as was the case with Victor and Kwesi. However, in Koo Frimpong’s case, he was raising a combination of boys and a girl, without the assistance of a female. At the time of the research, Koo was 34 years-old, single and ran a petty business. When I met him for the first interview in 2011, he was making sandals for a living. When I visited again in early 2013, he had expanded his business to include renting out motor-bikes and running a small shop where he sold frozen fish and meat products. He lives together with the children in the house of the migrant. Koo is not kin, but a friend of the migrant. When Koo took over the care arrangement in 2010, there had been a woman caregiver (the migrant’s friend) with whom the migrant had left the children when she was travelling. When the migrant heard reports from Koo and the children about the woman’s near abandonment of the children to concentrate on her business activities, the migrant reluctantly agreed to Koo’s offer to move in and help the woman to take care of the children. Koo said the migrant was reluctant because she did not want to burden him with that responsibility of raising children especially because she knew her two older boys were stubborn. Several months after Koo took over, the former caregiver was still the one in charge of receiving and managing the remittances. Later, the migrant felt it made no sense to send the money to the former caregiver when she virtually did nothing for the children. So she started sending the money directly to Koo. Finally, the migrant asked the woman to move out of University of Ghana http://ugspace.ug.edu.gh 114 the house since she was not fulfilling her part of their agreement. Shortly after our first interview in 2011, the two boys gained admission to Senior High School (SHS) and became boarders. Once they were both in boarding School, Koo was left with only with the youngest girl. Because Koo runs most of his businesses from home, he has a lot of time to pay more attention to the girl, who, according to Koo, could barely bathe properly when he took over full-time care of the children. As presented, the last four male caregivers throw up some interesting issues for discussion including whether migration is changing gender roles in homes. Is it the case that migrants get so desperate that, as in Koo’s case, they would leave a young girl in the care of a male caregiver, a responsibility that has typically been performed by a woman? I return to these issues in the latter part of the chapter. 5.2.1 Special Kin - Stay-Behind Wives The stay-behind wives are referred to as special kin and separated from the larger group of kin caregivers because their unique experiences in the TCRAs largely are as a result of their status as wives of the migrants and the biological mothers of the left-behind children. There were three caregivers in my study sample who belonged to the special kin category. These women shared a common characteristic of having migrant husbands who were undocumented and had therefore, not been able to visit them since they travelled abroad. The husbands of two of the wives had been arrested and incarcerated for living illegally in the host country. One of the migrants had been released shortly before the commencement of the research. The other was incarcerated during the course of the research. University of Ghana http://ugspace.ug.edu.gh 115 Naki, one of the three stay-behind wives, lives with her three sons in the Greater Accra Region. At the time the research began, her husband had been away for 15 years. Akua’s husband had left some eight years earlier. They have five children who were aged between 2 and 14 at the time he left. Pomaa’s husband has been away for six years. She lives with her three children in the Ashanti Region. Why Are They ‘Special Kin’? The stay-behind wives have been described as special kin because they experience the TCRAs differently. The underlying difference is that their experiences are more in consonance with their conjugal obligations towards their migrant husbands, unlike the other caregivers whose obligation to care for the children are more in consonance with the tenets of child fostering and negotiated commitments. The experiences of the wives, however, bear more similarity to those of the mothers of migrants. Not only do they have to worry about delayed remittances and other contributions migrants make towards their children’s upbringing, these women also have an emotional stake in the separation. The mothers miss their migrant children and are filled with fear that they might abandon them, while the wives miss their husbands and are worried about the status of their marriages, especially in the case of prolonged separation. Naki said: This marriage has been under a lot of pressure since my husband travelled. If it is not my family members telling me to leave the marriage because I am wasting my time then it is sometimes my husband saying things that implies that if I am tired of waiting for him, I can leave and find myself another man. I myself have sometimes questioned why I stay married to him after all these years of being separated (Accra, May 2012). During the period following the migration of their husbands, the lives of stay-behind wives changed significantly on a number of levels. On a personal level, they often must deal with the University of Ghana http://ugspace.ug.edu.gh 116 stress and anxiety the departure of a husband can bring (de Snyder 1993). And in the home or the family, they frequently take on the responsibilities which their husbands were in charge of before migration (de Snyder 1993; Menjivar and McKenzie 2010). These responsibilities include disciplining the children and taking major decisions regarding the children’s upbringing, especially education. Under normal circumstances, these decisions would have been taken by the husbands or jointly with the wives, but in their absence, these rest on the shoulders of the wives. Naki said this was very difficult at the initial stages of her husband’s migration because it was difficult to reach her husband to take the decisions together. She added: Sometimes, I was too broke to afford to make an international call and on the few occasions that I managed to call, I was asked to leave a message on the voice mail. And because we did not have a phone in the house, it was very difficult for him to return my call. Making plans with him was really difficult at the time (Accra, May 2102). Nowadays, as a result of developments in ICT in Ghana, accessibility to and affordability of communication technologies are easing these burdens on wives and caregivers in general (see chapters 6 and 7). The wives thus become the mothers and fathers in the home with the fathers trying to parent from afar via ICT. In addition, wives also become managers of the remittances and household wealth. These wives wait in anguish when remittances fail to arrive, or weeks passed without telephone calls from husbands abroad, as was the case for Akua whose husband stopped communicating with her in the previous two years because of a misunderstanding that had ensued between them. She told me that she had overcome the emotional stress that he had put her through, and that she had given herself a few more months, after which she would ask for the marriage to be dissolved, if this continued. McKenzie and Menjivar (2010) observed that sometimes these gaps were intentional. Some migrant husbands admitted that they occasionally intimidated their wives with deliberate silence when they were upset or suspected the women University of Ghana http://ugspace.ug.edu.gh 117 were not behaving properly. Additionally, when remittances were delayed or weeks passed without telephone calls, the wives began to suspect their husbands of infidelity. When I met Akua for the first time, she complained bitterly about her husband’s failure to communicate with her. In the course of our conversation, she said that if her husband was having an affair with another woman, he should just tell her and she would move on. In the case of Naki, it was her uncle who suspected her husband of infidelity when the husband could not send remittances regularly. Her uncle sent another relative who lived in the same country as Naki’s husband to check on him and, according to Naki, his suspicions were not confirmed. These cases also illustrate that distance instigates all kinds of tensions in the relationships and have consequences which may not be found or tolerated in care arrangements in which the caregiver was not the wife. For instance, it is quite unimaginable that a migrant will stop communicating with the caregiver of his child for over two years because they had had a misunderstanding, as in Akua’s case. She reported that her husband sometimes called to speak with the children via the phone of their co-tenants and sent remittances through their oldest daughter. She fumed as she said I just don’t get him; even if he left the children with a house help would he not be calling that house help to check how she is doing and how she is taking care of his children? (Kumasi, July 2012). Naki also expressed similar sentiments when she recalled a period where there was a break in communication because she had refused to heed to her husband’s warning concerning a new business she had wanted to venture into: The last time we had a quarrel, he would just call and when I picked up he would ask how we are all doing and then tell me to give the phone to our first child. He was the only one who was old enough to communicate properly on phone then. After speaking with him he will hang up. I was so hurt by that and thought that at least he should even ask about the well being of the one taking care of his children even if not to ask me as his wife (Accra, March 2012). University of Ghana http://ugspace.ug.edu.gh 118 The attitude exhibited by the husbands as described by Akua and Naki would rarely happen in other care arrangements. Through the exchange of notes with my colleague who is studying the parents’ situation, I understood that parents tried as much as they could not to court the displeasure of the caregivers, at least for the sake of the happiness of their children. This was especially where the children were left with caregivers who were neither wives nor grandmothers, or in the case of migrants who had very limited options with care giving and those whose legal and socio-economic status did not allow them to easily bring their children over to the Netherlands. Even when they had good reasons to be upset with the caregivers, they found more subtle ways of handling the situation to avoid tension in the relationship. Another difficulty that stay-behind wives were saddled with was petty gossip from their husbands’ families and sometimes also from neighbours. They had to deal with gossip about alleged infidelity and the misuse of remittances, and sometimes the migrant husbands’ reaction to this gossip (Coe, 2011; McKenzie and Menjivar, 2010). In addition, left-behind wives typically found that their mobility was closely monitored by the extended family. In-laws, in particular, often assumed that their chief role in the constitution of transnational households was in safeguarding their son’s respect by ensuring the honour of their daughters-in-law. Even when family relationships appeared amicable, women worried that their in-laws might call their husbands at the slightest sign of impropriety. Akua lived in the same village as her in-laws, and she confided that she would not be surprised if her husband’s attitude towards her had something to do with such gossip from her in- laws and people in the community. Similarly, Naki lost a benefactor as a result of the gossip by her neighbours who called her husband to tell him that she was having an affair with the benefactor, who happened to know her husband. The benefactor’s wife also heard the rumours, University of Ghana http://ugspace.ug.edu.gh 119 so the benefactor decided to stay away from Naki and her children to avoid any unpleasant actions on the part of her husband or his wife. Some of these incidents indicate that the wives have faced some hardships and heartache which they could hardly have anticipated at the beginning of the migration process. However, in the middle or by the end of the interviews, they still found a reason to say that their husbands’ migration had been beneficial to them and their children. They recounted how in spite of insufficient and delayed remittances, they did not believe that their children could have had the education they were currently having without the migration of their fathers. Naki’s first son attended one of the best Senior High schools in the country and she puts it down to the fact that they could afford to give him quality education at the basic level. The boy is currently pursuing a Bachelor of Science degree in Biochemistry in one of the country’s best universities. She said this may not have been possible on the income her husband earned when he was in Ghana, especially considering the number of extended family members who depended on him financially. Pomaa also mentioned that because of her husband’s migration their house is almost completed and her three children are receiving quality education. She explained that this would not have been possible without his migration as for several months prior to his migration; her husband did not have any regular job or regular source of income. This affirms the observation by Smelser (1998:5) that ambivalence is more characteristic of how individuals experience social life than certainty is. The majority of the women who expressed anxiety or fear qualified their assessments by asserting that their husband’s migration had been a good decision. 5.2.3 Paid Caregiver – An Emerging Phenomenon? The paid caregiver in this study was singled out because she was hired and paid a monthly salary to take care of children of a migrant couple. She also lived with the children in the University of Ghana http://ugspace.ug.edu.gh 120 migrant’s house, took care of them without the assistance of a relative and was not under the surveillance of a family member, which is usually the case in the other kinds of arrangements, and depended entirely on the migrants for all expenses. She did not have to use her own money on the children unlike in the cases of most of the caregivers, kin and non-kin alike. Ophelia was a 35-year-old single woman who was hired to raise three boys (14, 12, and 10 years-old) at the time of the research. This is a TCRA in which both the mother and father have migrated and are not a part of the SMS. The first time I met Ophelia, she was in her third month of her new job. After that first interview, I stayed in touch with her through phone calls and returned to visit her seven months after the first visit. This time I did not go for a formal interview as in the first instance, but I went to visit her and the boys in order to spend some time with them. Before taking this job, Ophelia worked full time with a sanitation company in Ghana. She left that job to devote her time to her new job as a paid caregiver. Ophelia found out about the job through her uncle who lived in the same area as the migrants. Her uncle mentioned to her that the migrant and his wife needed a caretaker to look after their children while they were away and that he had recommended her to them. She met them to express interest in the job offer. Next, a meeting was set up between her relatives and theirs. Ophelia was accompanied by her uncle and another family member. The migrant mentioned to her and her relatives that this was necessary because they did not know each other. One of the items on the agenda was the negotiation of her salary. Ophelia’s relatives were the first to propose a figure and told the migrant and his wife that they should bear in mind that she left her job to take up this one. The migrants negotiated to reduce the amount of pay, and, after a few back and forth sessions, they settled on an agreed amount which Ophelia did not disclose to me. University of Ghana http://ugspace.ug.edu.gh 121 Also discussed were the expectations each party had of the other. Both parties took turns voicing what they expected from the arrangement, what would make them happy to maintain it, and the possible reasons for its termination. Ophelia’s expectations included prompt payment of her salary and that her employers confront her directly if they were not happy with her performance. Her employers also told her to dedicate a lot of time to the care of the boys, even if that meant her boyfriend moving into the house with her. In addition, they asked Ophelia to discuss with them when anything she disliked about the care arrangement transpired. She was also to consult with them on all major decisions concerning the children’s education or health. However, she was at liberty to make minor everyday decisions without asking them, including disciplining the children when necessary. These arrangements were all discussed in the presence of the witnesses. In contrast to the ‘formality’ of the negotiat ions, there was no written agreement on the arrangement, with the entire transaction brokered verbally. After the meeting, the couple lived with them for three weeks for observation before they finally left the country. Part of Ophelia’s job was to move in and live with the children in the mansion the couple was building and to oversee its completion. Therefore, in-between the time that the children left for school and returned, Ophelia juggled house-keeping and oversight of the construction of the house. According to Ophelia, she received money from the migrants for everything they needed, which included regular monthly remittances including her salary. During my second visit, I mentioned that because it was so expensive to send money from abroad, migrants used all sorts of means to do this. She said she had not realized this because she received money so often from the migrants that she assumed it was almost free to send money home. University of Ghana http://ugspace.ug.edu.gh 122 According to Ophelia, so far she had not experienced any challenges, which she attributed to the fact that the migrants were happy with the service she was providing. She was also happy that they were honoring their part of the agreement. So far, neither of them had breached their side of the verbal agreement, so things were moving along very well. The couple had visited Ghana once since they left, and she indicated that it was shortly after my first visit and that they had stayed for one month. I asked her if she had taken the opportunity to go on ‘leave’ (vacation) when they visited. She laughed and said, “The parents told me that they had come on vacation so I should even pretend they were not around.” During the negotiation meeting, Ophelia told me that another thing that they discussed was an exit plan. The migrants told her that she was to feel free to leave at anytime, and that she only had to give them ample notice to enable them find a suitable replacement. ‘Ample time’ was not defined, but in Ophelia’s mind if she decided to leave, she would make sure that she stayed until they found a replacement, however long that might take. I found two similarities between Ophelia and the other caregivers. The first was their motivation for being in the care arrangement. Like the other caregivers, especially those who were not dependent on the migrant’s remittances, Ophelia saw the care arrangement as doing the migrant and their children a ‘favour’. The other motivation was the expectation to gain from taking care of the children, as in the case of the some of the caregivers who relied on remittances from migrants. The second similarity is that for Ophelia and other caregivers, their role required their maximum attention and was thus very time consuming. Ophelia’s responsibilities were very much like all the other caregivers who had younger children in their care. She was responsible for the day to day care of feeding, bathing, clothing, and helping the children to stay out of trouble. She was also mandated by the parents to discipline them which included using the University of Ghana http://ugspace.ug.edu.gh 123 cane when necessary. Ophelia was not performing any activities different from the other caregivers because she was hired. She made major decisions in consultation with parents just like the other caregivers did, such as arranging for extra tuition in the home for the children as they were not in a position to help the children with that. There was hardly any break in the caregiving for as long as the children lived with her. The fact that Ophelia received a salary did not give her any rights to a vacation, as is the custom for salaried workers in Ghana, such as for caregivers who worked in institutional foster homes. I also observed that the major difference between this care arrangement and all the others is the paid nature of the job. There was a similar situation involving one of the teachers in which the migrant offered to pay her for her services, but she refused to accept the ‘salary’. She preferred the extra amount the migrant added to the figure they agreed upon for the child’s upkeep and saw it more as a gift. It appears that this caregiver felt more comfortable with the fact that it was the migrant’s idea to send the extra amount despite her reluctance to accept it. Thus she did not consider herself as being paid by the migrant. Her resistance to being paid to take care of the child reflects the norms of social parenthood in Ghana where childcare is a shared responsibility and nobody expects to be paid or openly asked to be rewarded to take care of another’s child within any given community, making Ophelia’s accepting to be paid to take care of the children unique. But these two cases are also different in the sense that the child moved into the teacher’s home, whereas Ophelia did the opposite. Paying for domestic services is not exactly new. However, the usual trend is that people hire the services of live-in house-helps who live under the same roof as their employer and take instructions from them. They help the mothers to take care of the home but they do not take on a ‘motherly’ role as Ophelia did. A salary, in the strict sense of the word, comes into play if the University of Ghana http://ugspace.ug.edu.gh 124 house-help is contracted through a recruitment agency. Otherwise if the house-help is recruited through the employer’s own network, they agree with the house help’s family for her to stay with them for an agreed number of years, and, when she leaves, the employer gives her some working capital as well as places her into an apprenticeship of her choice. To ask the caregiver to live in the home of the migrant is also a common practice, as some of the grandmothers and aunts were living in houses built by the migrants. What makes Ophelia’s case peculiar then, is that she is not kin and does not have any ‘family’ member of the migrant living in the house to supervise what she does, which is what usually happens when people leave their children behind in the care of house-helps. In such instances, there was usually a ‘grandmother’ or an older person keeping an eye on them. The paid nature of this type of caregiving leads to a significant observation: It is costly because the job entails paying a salary and allowing the caregiver to move into the migrant’s house. As such only a certain class of migrant can engage the services of a live-in paid caregiver. This case demonstrates that the practice is only possible for a particular class of migrants. The two caregivers in the cases cited (the teacher who was offered payment but declined it, and Ophelia) admitted that they lack nothing as far as the children’s upkeep was concerned. Even though the teacher was gainfully employed, she said that she had never had to use her own money for anything because the parents sent money on a regular basis, and what they sent was always more than enough. The cost was even more in the case of Ophelia because she lived in the migrants’ house and the migrants provided for everything including the payment of the utility bills. All these costs were factored into the remittances they sent. Furthermore, because the children in Ophelia’s care had once lived abroad, the parents preferred to ship boxes of food and sometimes medication for them. Ophelia only bought food when they ran out of the stock of University of Ghana http://ugspace.ug.edu.gh 125 food that had been sent to them. If for some reason she spent money meant for the building project on the children, such as visiting the clinic unexpectedly, the parents immediately reimbursed her. All the electronic gadgets in the house were bought by the migrants and would be replaced or repaired at their cost if they broke down. The additional cost of having a live-in caregiver was something other migrants escaped from because their children moved into the house of the caregiver. Furthermore, this practice is also more costly because the caregiver’s source of income depends on the migrants’ financial state and should anything go wrong with them, the consequences would be dire, just as in the case of the migrants who had caregivers that did not work for pay. Therefore, to be in a care arrangement like that of Ophelia’s required that the migrant have a particular economic status such as that of Ophelia’s employers and the migrant whose child was with the teacher. I found out from the caregivers (Ophelia and the teacher) that the migrants whose children are in their care have had tertiary education, documented legal status and good jobs in the host country. The caregivers in these two TCRAs admitted that they did not lack any material resource because everything was provided for by the migrants. Low skilled migrants (documented and undocumented) may not be able to afford a paid caregiver because it may be quite costly. Moreover, because these categories of migrants usually have fewer financial resources and may have to rely on the caregivers to supplement what they provide, this option may not appeal to them. 5.3 Issues Emerging from Caregivers’ Characteristics Transnational Parenting and the Gendered Nature of Care University of Ghana http://ugspace.ug.edu.gh 126 In most societies, care-giving falls naturally to women, more so when it is childcare. However, the involvement of the male respondents in the care of the migrants' children in this study suggests that this norm may be changing. There are some characteristics about the four primary male caregivers that, to a large extent, reinforce the fact that women are the preferred caregivers and that it is the circumstances surrounding care-giving that are leading to the change in norms. First, in all four of these cases, the caregivers were not the original caregivers that migrants had chosen. The children were initially left in the care of female caregivers, but had to switch to the ones discussed during their parents’ migration. In Koo's case, the migrant preferred her friend, a woman with whom she had had a long friendship and may have established strong ties and trust, to take over her child-caring role when she was away. Koo only came into the picture because the caregiver of choice did not work out. Takyi came onto the scene because he was the only relative of the migrant who lived in the city where the migrant wanted his children to live so they could get good quality education. Victor had to take charge of Ata because her grandmother with whom she had lived passed away. On his part, Kwesi became responsible for his uncle’s son because the boy was getting out of control in his mother’s hands. Thus, the male caregivers had been drawn into care-giving arrangements because of changing circumstances within the TCRAs and a lack of other options from which the migrants could choose in order to secure a care arrangement for their children. I raise the latter point to emphasize that care arrangements in transnational families are fluid, of a temporary nature and change over the life course (Baldassar, 2007). They are malleable and adjustable depending on the circumstances. Yet, while men enter into the care arrangements as primary caregivers due to changing circumstances, women continued to be the primary providers of child care changing in TCRAs. University of Ghana http://ugspace.ug.edu.gh 127 The cases of Takyi and Koo, who took care of children who needed maximum time and attention because they were young are similar to what Asima (2010) found in his study of male caregivers although he focused mainly on stay-behind husbands. Asima examined shifting gendered norms (renegotiations of the household division of labour) in the home when one partner travels. He observed that stay-behind fathers were gradually taking over some of the roles ascribed to mothers. This, he noted, posed a challenge to the maintenance of the ideology of separate spheres as well as the traditional gender division of labour in the household. He posited that transnational migration is accentuating the process whereby men are getting more involved in the female space and taking over female functions, and may, therefore, be transforming social relations in the absence of their migrant wives. The findings in my study move this notion further to include men who are not part of the migrant’s household nor of the migration process taking on these ‘female roles’, such as cleaning the house, cooking for the children, bathing them and helping them with their laundry until they are able to do so for themselves. Such were the situations of Koo and Kwesi when they took over the care of the migrants’ children. These two caregivers were both bachelors and it may be argued that they were performing some of these chores such as cooking or cleaning the house before they became caregivers of young children. However, Takyi’s response when I asked if he was doing these chores before is quite telling. He said: Before the first boy was sent to me, I did not care about what I ate. I cooked once in a while maybe the weekends. On the weekdays I spent most of the day at my workshop and ended up eating there with my friends before coming home. I usually went straight to bed when I got home from work. But with these children in the picture I could not live like that any longer. I had to make sure there was always food in the house. I had to clean regularly in order to set a good example for them and I believe they learn quickly when they watch you do it. I had to bathe them, choose the attire they will wear, dress them up and make sure they appear clean and neat all the time. I taught them how to brush their teeth, comb their hair etc until they were able to do so themselves. All this meant that I had to be home most of the time. I left for the workshop late and came back early (Kumasi, June 2012). University of Ghana http://ugspace.ug.edu.gh 128 Takyi’s response illustrates that even though these bachelors may have been cooking prior to the care arrangement that changed when they lived with the children and food had to be provided on a regular basis. The chores they found themselves performing were those that mothers will do in a typical Ghanaian home where both mothers and fathers were present. Indeed, this situation also corroborates what Toyota et al. (2007) confirmed that ‘leaving behind’ often means a reorganization of significant facets of daily life. Another interesting feature about these male caregivers is that they are self-employed or have jobs with flexible schedules enabling them to provide full-time care to the children without the assistance of other relatives, especially women. The ability to dedicate more of one’s time to care-giving is one of the reasons women are preferred to men; women are assumed to have a lot of time to devote to the care of children. Takyi’s example of ensuring to leave for work late and return early just to be available all the time because of the children is a case in point. The nature of the jobs of the male caregivers discussed above enabled them to exercise the level of dedication that women are known to exhibit in caregiving. Thus, even though more males appear to be entering the female sphere, they must possess certain attributes that female caregivers typically possess to perform the role of a caregiver, thereby reinforcing to a large extent the fact that care-giving is female business. Inequality and Power Relations in TCRAs As part of the SMS method, I had the opportunity to meet some of the migrants with my colleague who was interviewing them during one of our site visits. Information I gathered during those visits indicates that some of the migrants were not satisfied with the type of care being provided for their children. They were, however, unable to control the situation either because to University of Ghana http://ugspace.ug.edu.gh 129 a large extent they did not have enough money to fund how they wanted the children to be raised, and to lesser extent the caregivers were older, such as parents or older siblings of the migrants, and would not allow themselves to be ordered about or told what to do. When migrants pay for everything concerning a child’s care and, in addition, either pay or give regular allowances to the caregiver, it increases their level of involvement and gives them more ‘control’ over the children’s upbringing. I found out that the migrants had expectations of how they wanted their children to be raised. One way to meet that expectation was for them to prove their ability to fund it. For instance, one grandmother mentioned that her son did not want his daughters to do household chores, so she asked him to bring more money to enable her hire a house help. ‘He did so his girls do not touch anything in the house with the exception of their food.’(Maa Fosua, Kumasi June 2012). In Ophelia’s case she was asked not to leave the children by themselves or with any other person. Where it became necessary for her to do so, whoever she decided to leave them with had to be somebody known to the migrant and whom the migrant approved of. In line with that, she was asked to let her boyfriend if she met one, move in with them to avoid the incidence of spending time with him at the expense of attending to the children. When a migrant can fully provide for his or her children but the caregiver is not dependent on the migrant, there is bound to be a power clash. The situation of the teacher and Opoku, the migrant whose child was in her care is a case in point. When we met Opoku, we inquired about the care arrangement with the teacher, his son’s caregiver. We wanted to know why he had moved his son from the previous caregiver to the teacher and why he offered to pay the teacher. He explained that his son was not doing well academically and he needed someone who could pay more attention to the child’s education, thus he preferred a teacher. He was also University of Ghana http://ugspace.ug.edu.gh 130 very emphatic that the reason he proposed to pay the caregiver was so that he could instruct her on how she should raise his son. “I offered to pay her so that I can tell her what to do” (Opoku NL, February 2012). By the end of my field research, Opoku had withdrawn the child from the teacher’s care. The teacher had refused to be paid, and thus she hardly followed the migrant's instructions as he had intended. That is not surprising because Opoku wanted control, and the teacher told me in an interview that she emphatically asked Opoku to allow her to raise the child without him interfering. She said, 'I told him that if he [had] trusted me enough to bring the child to me then he should allow me to do my own thing' (Maggi, July 2012). The two adults could not agree on a parenting style and both of them wanted to exercise as much control as possible. When I interviewed Opoku’s mother, Maa Fosua, in whose care he had left his two older daughters, she confirmed what the teacher said about his tendency to tell you how to raise his children. She said that they always argued over the ‘proper’ way to raise the children. He wanted to exercise the same level of control over his mother, but he did not always succeed because she is his mother, she has her own business and a good source of income. These circumstances made it difficult for him to order her around. She told me the only time they had arguments over the children was when her son would try to tell her how to raise the children. She said she sometimes gave in, but most of the time she just ignored him. There are other factors that allowed one to have more say in a care relationship, such as one’s expertise or hierarchical relationships (mother-son) as in Opoku’s mother cited above. However, when the migrant has the material resources, he is able to realize almost all of his expectations for his children, and this could sometimes be frustrating for the caregivers who felt restricted in raising the children in the way they deemed appropriate. Caregivers who have the University of Ghana http://ugspace.ug.edu.gh 131 financial capacity are also able to act independently and not follow all the ideas that migrants suggest without the fear of losing their source of livelihood (migrant’s remittances). 5.4 Summary Consistent with other studies, this study also found that a majority of the caregivers were kin and women, with grandparents as the most preferred caregivers. Additionally, the majority of the caregivers engaged in income-generating activities and were, therefore, not dependent on the migrants’ remittances for their own upkeep or that of their children. There were some new interesting findings, namely the emerging phenomenon of paid care-giving, and males entering the female sphere of care-giving. From the discussions of the characteristics of caregivers I made two significant observations, which add to the literature. The first is the inequality and power relationships that exist in the care arrangements. I observed migrant parents’ access to wealth in the care arrangement gave them greater power which has to a large extent reversed the order of power relations in child fosterage where the foster parents (caregivers) wielded more power in the relationship. The second significant observation was men entering typical ‘female’ spheres, some of whom were not assisted by female caregivers. This observation leads to the question of whether international migration engenders changes in gender roles and relationships. I cannot conclude from the findings of my study or the body of relevant literature that international migration changes gender roles and relationships because migration is seen as temporary and so are the roles that are engendered by it. The most plausible conclusion to make from this study and the University of Ghana http://ugspace.ug.edu.gh 132 body of relevant literature is that relationships and roles among men and women are prone to change over time (Oppong et al, 2006). They are also fluid, temporary and change over the life course (Baldassar, 2007). Thus the roles and relationships are malleable and negotiable based on circumstances, one of which happens to be international migration. University of Ghana http://ugspace.ug.edu.gh 133 CHAPTER SIX: TRANSNATIONAL CAREGIVERS: PATHWAYS AND EXPECTATIONS 6.0 Introduction This chapter highlights two significant aspects of caregivers lives in TCRAs, namely how they came into the care arrangement, which I refer to as pathways, and their expectations from the migrant parents and the children for whom they provide care. These features have largely been overlooked in the few studies which focus on caregivers (Moran Taylor, 2008; Mejivar and Agadjanian, 2007; Dreby, 2010; Coe, 2011; Menjivar, 2012). The pathways that caregivers followed into the care arrangement had a significant influence on their expectations of the migrant, the child and the care arrangement as a whole. The analysis of the pathways further revealed that caregivers were informed by specific factors, namely normative obligations, generalized reciprocity and the capacity to provide care. The chapter is organized as follows: In the first section, I look at the variety of pathways caregivers followed into the care arrangements and the factors that informed their choices. This is followed by a discussion of the caregivers' expectations of the care arrangement. Here I show that their motivations as caregivers are not always as altruistic as they may have one believe. Instead, as Mauss (1954: 3) put it, 'their services are given and reciprocated obligatorily'. 6.1 Pathways into care By pathways into care, I refer to how caregivers come into their caregiving roles. In this section, I examine the process of decision-making and discussions (if any) that took place before they accepted to take care of a migrant’s child. The data present two broad categories of pathways namely the ‘involved and ‘uninvolved’ which have been developed on the basis of the caregivers’ conscious involvement in the decision to become caregivers. Their level of University of Ghana http://ugspace.ug.edu.gh 134 involvement in the decision-making, however, would come to bear greatly on their experiences, especially their expectations of the care arrangement. The ‘involved’ Pathway Caregivers who followed the 'involved' pathway were consciously involved in the decision to become the caregiver of a migrant relative or friend. There were three ways in which they did this. First, there were caregivers who were actively involved in the decision-making process to migrate and, consequently, the care arrangement. Second, there were caregivers whom the migrants approached about caring for their children either before they travelled or in the course of travel to which the caregivers accepted voluntarily. Finally, there were those who became caregivers because they felt sympathetic towards the children of the migrant, whose current caregiving arrangement was not healthy and they asked the migrants to let them take care of the children. Caregivers in all three categories can be said to have made an informed and voluntary decision to raise the children of a migrant after carefully weighing their options, considering potential benefits and perhaps assessing their capacity to provide care. The caregivers who actively encouraged their relatives to migrate were usually grandmothers and mothers of the children left behind who encouraged the migrants to travel to secure a better future for the family. In most cases, these caregivers were also the direct beneficiaries of the migration decision. They were often involved in the decision-making from the time the idea to migrate was conceived to the migrant's departure. In fact, for these caregivers the acceptance to take care of a child was another way to express their support of the migration decision. Some of the grandmothers in this study admitted their daughters’ migration was partly their idea as observed by Dreby (2010) in her study of Mexican women who left their University of Ghana http://ugspace.ug.edu.gh 135 children in the care of grandmothers. The migrant women in her study indicated that their mothers were the ones who encouraged them to take the final step of leaving. Some of the grandmothers in this study were compelled to encourage their daughters to travel because of their low level of education and thus low economic status. In Ghana, those with little education have few prospects of climbing up the economic ladder. With the poor social security system and income insecurity, the majority of Ghanaians encourage their children or spouses to emigrate for better economic prospects. Such a phenomenon is compatible with migration research which posits that migration is a means by which many African households migrate out of poverty to improve their lives (IOM, 2009). It must be noted here that most of the daughters (migrants) in question were either divorced or separated from their husbands. The actions of the grandmothers in my study, to a large extent, also support the New Economic Labour Migration (NELM) theory which claims migration decisions are not made by isolated individuals but by larger units of related people – typically families or households who act collectively to maximize expected income and to minimize risks and loosen constraints associated with variety of market failures (Lucas and Stark, 1985; Stark and Bloom, 1895; Stark 1991). One of the grandmothers, Grace, in my study intimated to me that she was the one who convinced her older sons abroad to help their younger sister travel abroad to enable her provide for her three children. She told me that her daughter’s business collapsed in 2004, and, as a result, she had to depend on the meager remittances from her sons to feed both her daughter and her grandchildren. In order to avoid further hardships, she explained: 'I convinced her brothers to help her travel. When they agreed to do that, I gladly assured her not to worry about the children because I will take care of them' (Kumasi, November 2011). Mankpa, a grandmother also remarked: University of Ghana http://ugspace.ug.edu.gh 136 My daughter’s marriage was not leading anywhere and she was struggling to make ends meet, as her husband was also not gainfully employed. One day she told me about the decision of her brother (my eldest son) to arrange for her to join him in the United States. I assured her of my full support and before she could even ask who will take care of the children, I offered to take care of her four children while she was away. She has been away for almost seven years now (Accra, January 2012). Similarly, some of the wives in my study also encouraged their husbands to travel abroad in the hope of improving their standard of living. In their cases, the children were already living with them, so there was no discussion as to who would take care of the children when the husbands migrated. However, they fit in this pathway because of their active involvement in their husbands’ decision to migrate. The following cases of two of the wives in my study illustrate this: Pomaa described her situation as follows: My husband had earlier on abandoned his teaching profession to run a transport business for his senior brother who was in the Italy. His brother promised that he would arrange to send him over to Italy after some time. I was very much against him quitting his teaching job for his brother, but he had already made up his mind. The transport business went very well and out of it he built two houses for his brother and then started putting up his own house (where we live now). Sometime later, the transport business began to recede and he started pressing his brother to fulfill his promise of sending him abroad. Some misunderstanding ensued between them and his brother quit running the business with him and also told him that he could not send him to Italy as promised because he did not have the funds to do so. Desperate to get out of the country because there was nothing for him to do, somebody arranged for him to go to Amsterdam. I was all in favour of it because I knew he would earn some income abroad. (Kumasi, November 2011) Akua, another wife also said her husband discussed his intention to migrate with her. She mentioned that before he travelled to the Netherlands, he was already travelling between Ghana and other African countries to trade. He did this for some time, but eventually decided that the business was not profitable enough and thought of travelling to Europe. She gave him her full blessings. Akua has been raising their four children aged between 9 and 14 years in Ghana since he travelled eight years ago. University of Ghana http://ugspace.ug.edu.gh 137 Like the grandparents in my study, the two wives cited above gave their consent because they believed that the move would be an economic venture that would yield some good returns, even if not immediately. The caregivers survived the initial periods of migration by depending on income from their own businesses and financial support from others especially their older children. Most of the respondents in the study admitted that even if the migrants’ circumstance in the host country may not be what they expected, they still considered it a better option than living in Ghana. The second category under the ‘involved’ pathway includes caregivers who were usually not part of the decision to migrate, but had accepted the migrants’ request to take care of the children while they were away. For this group of caregivers, the migrants’ decision to migrate did not provide direct benefits for them; at least, not at the time the migration took place. They either took over the care of the child at the time the parents travelled, or they did so during the course of the parents’ migration. The latter situation was more prevalent in this study and was induced by a number of circumstances. One was when migrants who were already living in the host country with their children wanted to send them to Ghana to be raised there. This practice is quite common among West African migrants abroad (Bledsoe and Sow 2011; Asima 2010). The story of Kessewa below is a case in point. My sister called me from abroad and told me how overwhelmed she was with handling her four boys in the middle of an ugly divorce case, and one of the boys was constantly falling ill. This was going on for some time. One day after she had finished lamenting, she added, 'I am bringing them over to you' and I said Ok. I didn’t even have to think about it or ask her to call later for a reply. (Kesewaa, Kumasi, July 2012) Maame Ama had a similar experience. She had been taking care of her older brother’s daughter for the past two years. She told me that it was the second time the child had been put in her care. The first time was some 10 years ago when her brother and his wife were travelling for University of Ghana http://ugspace.ug.edu.gh 138 the first time. They left their three children in her care because they could not travel with them. A few years later they arranged to send all three children over to the host country after being granted the right to do so. After almost eight years, her brother called her to say he wanted to send the last girl back to live with her in Ghana, and this is how Maame Ama came to care for the child a second time. Other circumstances prevalent in this category of caregivers include situations where migrants experienced challenges with the initial care giving arrangements, for which reason it became necessary for the migrants to re-arrange care for the children which could be very challenging for migrant parents and sometimes the new caregivers as well as others have observed elsewhere (Akesson et al. 2012). For example, Cecilia, who was raising five children of her migrant niece, told me that she was not their initial caregiver. The initial plan was for Cecilia to take care of the last child, who was six months old at the time her niece travelled. The other four (4, 11, 15, 17) were left in the care of their father. About a year later, Cecilia had a call from her niece pleading with her to allow the other four children to move in with her. Her niece said the rearrangement had become necessary because of a heated argument she had had with her husband that made her fear he would abandon the children. That was how she came into her care arrangement. Additional circumstances that led to a change of caregivers in the course of migration were the parents’ desire for better living environments and educational prospects. Takyi explained 'I got a call from my sister and her husband that they wanted Tony to attend school in Kumasi and wanted him to live with me (Kumasi, November 2011). They wanted to move the child from the village where he was under the care of his father’s relatives. Takyi was the only University of Ghana http://ugspace.ug.edu.gh 139 one they knew living in Kumasi, so the boy was sent to him. Georgina also explained how her grandson was moved from his maternal grandparents to live with her: He was previously with his mother and maternal grandmother in a compound house that had many other relatives of his mother’s. My son (the migrant) was worried that the child was not receiving proper attention and didn’t have the best of care because he had to share a room with his mother and grandmother. Also, they shared everything he sent for the son, including remittances, with the other members of the family. We, therefore, decided that he should come here and be with me. He has more space here, a room to himself where he can study quietly. There are only three of us here. This was difficult for the child, his mother and her folks, but under the circumstance it is the best decision for him ( Kumasi, August 2011). The third category of caregiver who followed the ‘involved’ pathway was a friend of a migrant who asked to take over the care of her children from another caregiver. Koo explained that he had to persuade the migrant to let him move into her house to help with the care of the children. The children who were 15, 12 and 8 had been left in the care of their mother’s friend. As Koo explained: That woman (the previous caregiver) didn't have time at all for the children. She ran a provisions (grocery) store and that was all she cared about. She left the house very early and took a break during the day to cook the children’s dinner, after which she left for the shop again. She returned late in the night. I called their mother who had become my friend because I met her regularly at the caregiver’s shop when she visited Ghana. I suggested to her to let me take care of the children because I thought her friend was not doing a good job with them. Initially she was hesitant because she said it was going to be too much work for me. After further persuasion she reluctantly agreed for me to move into the house and help her friend with the children ( Accra, January, 2013). Although Koo was not involved in his friend’s decision to migrate, he consciously decided to be a care giver by offering to take care of the children. The ´Uninvolved’ Pathway The ‘uninvolved’ pathway is the opposite of the ‘involved’ pathway where a caregiver does not consciously decide to be a caregiver. There was only one out of the 34 caregivers, Naki, who University of Ghana http://ugspace.ug.edu.gh 140 followed this pathway. She is one of the three wives in the study. According to her, she was not involved in her husband’s decision to migrate. She explained: We did not discuss his plans to travel. It was sudden for me. I found out only three days before the trip. We had had some misunderstanding earlier, so we were not on speaking terms; there was no communication flow between us. Three days before his departure, I saw a new suitcase he had purchased, so I asked why a family man as he is would buy such a small suitcase. That was when he told me that the leaders (he was one of them) of their church were travelling for a conference. I did not even suspect that he was going abroad. I asked about this conference that he claimed to be attending, and then he said they were going to Spain. I was shocked and sad that it took so long for him to tell me about a trip as important as this. That was all we said about the trip. I didn’t ask any further questions. He later told me that the conference would last for two weeks. I helped him pack and get ready, but we never discussed anything. Our quarrel was still on so I didn’t ask any questions. He never mentioned his plans to remain there after the conference. In my mind he was going to be back after two weeks. I didn’t know that it would be prolonged to 14 years” (Naki, Accra March 2012) Following Naki for two years, I observed that her ‘uninvolved’ pathway was more of a function of the tension in their relationship prior to her husband’s travel and not the norm. The pathways presented are thus asymmetrical as an overwhelming majority of the caregivers followed one pathway. I had to create the second pathway for Naki because she did not fit in any of the subcategories under the involved pathway. Secondly, I reckon the creation of the second pathway also opens up the possibility for a wider study to capture caregivers who may follow such a pathway. For instance, Poeze et al (forthcoming) found in their study on migrant- caregiver relationships, a migrant who left her children with her mother without informing her that she was travelling for fear that she might refuse to take of the children. It will be useful to explore such situations and the circumstances that lead to such especially in the cultural context of child fosterage. The lack of balance in the presentation of the pathways notwithstanding, I observed some differences in caregivers who followed the two pathways. This was particularly evident between University of Ghana http://ugspace.ug.edu.gh 141 the wives who followed the ‘involved’ pathway and Naki in terms of preparedness for their role and how they handled some of the challenges the faced at the initial stages of their husbands’ migration. Men’s migration meant more work, stress and anxiety for wives who stayed behind because of the additional responsibilities placed on them as observed by Asima (2010) and Coe (2011b) among some Ghanaian wives. Mckenzie and Menjivar (2010) made a similar observation in their study of Honduran wives. Thus when the process is not well thought out as was the case of Naki in the present study, it leads to increased stress for the women (de Snyder 1993), and this stress also shapes their views of the men’s migration. The three wives in the study faced similar challenges of poor communication, insufficient and the weight of raising the children alone in the initial stages but I noticed that Akua and Pomaa in the involved pathway were more decorous in their description of the situation. Naki on the other hand intimated that she was bitter for several days and still found it very difficult to forgive him for what she suffered at the time. She said: “I was so overwhelmed with everything – financial difficulty, poor communication, loneliness and I felt very helpless when the boys were constantly misbehaving and I had to discipline them which was something he used to do before he travelled”. It appeared as though Akua and Pomaa had anticipated these difficulties and accepted it as part of the risk they took when supported their husband’s decision to travel. Naki in the other hand did not have the luxury of weighing any options. The subsequent discussions on pathways will therefore focus on the 33 caregivers who followed the involved pathway. 6.1.1 Factors that informed the decision to be a caregiver The discussion on the two distinct pathways to becoming a caregiver, especially the involved pathway, revealed that the caregivers’ motivations and decisions to raise the children University of Ghana http://ugspace.ug.edu.gh 142 were informed by one or more of the following factors: normative obligation, generalized reciprocity and capacity to provide care. In the subsequent sections I define and discuss how the three concepts apply to my study. Normative Obligations Finch and Mason (1993) used the concept of normative obligation to describe “care that is given based on notions of duty and responsibility, supposedly at the core of family relations defined by policy makers as part of the moral fabric of society that everyone should adhere to” This concept describes caring for each other in the family as the 'proper thing to do' in fulfilling kinship commitments (Williams 2004:55). In Ghana, norms of shared parenting largely inform caregivers’ decision to take part in transnational child raising arrangements. As mentioned earlier, even though some caregivers were not involved in the decision to migrate, when confronted with the situation of caring for the migrant’s child, they chose to do what was proper and acceptable. After all, in most African communities the child is perceived as being born to an entire family. This supports one key point that the growing literature on transnational families has raised--that caregiving across borders cannot be fully understood without considering the cultural context of the people being studied (Baldassar et al. 2007; Mazzucato and Schans 2011). All the caregivers in this study mentioned that their motivation for agreeing to raise the children of migrants was to help them or their child/ren. Maggi, the teacher who was approached by the migrant to care for his son said: “The man was desperate and I considered this a favour he was asking of me. The discussion we had gave me the impression that there was a problem with the previous care giving arrangements”. (Kumasi, June 2012). Takyi also mentioned that he agreed to because he saw them as his sons too. ‘They are my sister’s step sons and therefore they are my nephews and I will be happy to see them succeed in future’ (Kumasi, 2003). University of Ghana http://ugspace.ug.edu.gh 143 Almost all the caregivers mentioned reasons similar to Maggi’s and Takyi’s for accepting the children. Some of the caregivers were also already involved in the care of the children prior to their parents’ migration. Mama K. in the opening vignette and Awurakua are examples of such Awurakua said: My sister (migrant) invited me to come and live with her to help her take care of her children after her husband travelled. She had twin girls and she was pregnant with Agyeiwaa (TCRA child). I was with her when Agyeiwaa was born and I took care of her right from infancy. When Agyeiwaa was two years old, my sister told me about her intentions to travel to join her husband. She arranged and sent the twins to their paternal grandmother. Agyeiwaa was ‘automatically’ left with me as I had already started taking care of her (Accra, May 2012). Their motivation for accepting the children was to help lessen the burden on the parent and to help raise the child to become a responsible adult. The burden identified in most fosterage situations was the financial ability to provide for a child's material needs. However, in transnational migration, material provision became secondary to the caregiver’s ability to provide what the parents could not because of distance and time, such as day-to-day care of the child. The responses of the caregivers, therefore, echoed that they were doing what was the culturally acceptable thing to do. Negotiated Commitments/ Generalized Reciprocity Finch and Mason (1991,1993) take this point further by arguing that the support a child receives is not always what results from a 'normative obligation' or 'the proper thing to do', but of just 'working it out'. They referred to the act of “working it out” as a 'negotiated commitment'. In my study, this concept can best be likened to and adapted as generalized reciprocity. The authors described this as consisting of the long histories of relationships and exchange of care and support that cumulatively inform each decision to provide or withhold a certain form of care University of Ghana http://ugspace.ug.edu.gh 144 required by a particular person at a specific time and that is what reciprocity is about. There were some caregivers in my study who mentioned that they shared some past history of relationships and exchange of care with the migrants. Therefore, even though some of them were sometimes not involved in the migration process, they accepted taking care of the children first, because it was the right thing to do, and secondly, because of a past favour received from the migrants or expectation that this favour will be reciprocated in the future. The following example illustrates this point. Kwesi, agreed to take care of his uncle’s son as a way of reciprocating a past act of kindness from his uncle. He went on to describe how his uncle helped place him in an apprenticeship when he finished senior high school (SHS). My father died shortly after I completed SHS, and there was nobody to help sponsor me to learn a trade. Then one day my uncle (the migrant) visited us in the village when he returned from Holland and decided that I should come to Accra so he would sponsor my apprenticeship. He also set me up (gave me money to acquire this workshop) when I completed my apprenticeship and allowed me to live in his house until I was ready to move out on my own. He believed in me and made me what I am today (Kwesi, Accra, December 2011). He continued: Sometimes the boy gives me a hard time … disregards my advice, and I am always called upon to resolve one crisis or another at the place where he has his apprenticeship. But I don’t mind doing all that. I do it because of his father, and I also want him to get a good future too. Sometimes the remittances are delayed and I spend my money on him until we receive something from his father. I cannot pile up debt for him to come and pay. We can just say that the boy is enjoying the returns on the investment his father made in me (Kwesi, Accra, December 2011). Kwesi was taking care of his uncle’s son who was almost 17 years old when he came to Accra. His uncle instructed him to bring the boy to Accra from the village where he lived to learn a trade. Kwesi’s role was to supervise the boy and manage the remittances that his uncle sent for him. University of Ghana http://ugspace.ug.edu.gh 145 Caregivers who had not received any kind gesture in the past from the migrant envisaged their care giving role as an act that will be reciprocated in the future either by the migrant, the child or both. This is discussed in the section under caregivers’ expectations. Capacity to Provide Care Baldassar et al. (2007) adapted Finch and Mason’s (year?) normative obligation and negotiated commitments for the transnational context and added that capacity to provide care also plays an important role in care support in a transnational context. The authors define 'capacity' as the myriad of issues that encompass an individual’s opportunity and ability to engage in practices of transnational care giving. They further describe capacity as the availability and affordability of resources including money, time and mobility. Although their work discusses migrants who were adult children caring for left-behind parents across borders and, for this reason some of the things that constitute capacity for them may not apply to the caregivers in this study, the concept of capacity generally is applicable. For instance, resources including money, time and ability (i.e., physical and mental health) were required for caregivers to perform their duties. In addition, the capacity to care largely determines whether a caregiver needed a large network of people to assist with care (Chapter 8) and, to a large extent, how caregivers managed the challenges they faced (chapter 7). It was thus imperative for caregivers to give careful consideration to their own capacity to provide care before they agree to be part of the care arrangement. However, assessment of one’s capacity was largely neglected when caregivers were taking the decision to provide care. This neglect ended up posing a lot of inconvenience for them sometimes. University of Ghana http://ugspace.ug.edu.gh 146 The disregard for assessing one’s capacity can be partly attributed to the circumstances under which care is arranged. Sometimes, especially in cases where migrants have to re-arrange care from abroad, child care was arranged under such desperate situations that the caregivers’ capacity or a lack of it did not make much difference as were the cases of Takyi and Cecilia. They were reluctant to take care of the migrants’ children because they reckoned they did not have the capacity to do so. Takyi told the caregiver he did not have enough room in his house to accommodate the children and also he could not afford to provide material needs for the two boys from his income. The migrant insisted that he accept the children because he had no other option. He therefore accepted to take the children in on the condition that the migrant would provide money for their upkeep and rent a bigger apartment to accommodate the children. As at the time I completed the fieldwork in 2013, the migrant had not been able to fulfil the second condition and thus Takyi and the boys were still living in the small room. The other case was Cecilia, mentioned earlier who had five children of her niece. She said she was very hesitant to take in her niece’s four children who joined later because she did not have the space to accommodate them and the time to take care of all of them. Her niece pleaded and insisted in spite of her concerns because she was desperate and had no other option. In such desperate situations the caregivers’ assessment of her/his capacity became inconsequential. Secondly, the disregard for assessing their capacity can also be explained by the fact that sometimes caregivers do not foresee all the unexpected future challenges to help them make an informed decision at the time they accept the children. For instance, the wives whose husbands were arrested and jailed for several months did not foresee this. Some caregivers told me that they did not anticipate that it would take so long for their migrant relatives to get legal resident permits in the host countries. Thus, caregivers may accept the children based on their present condition and on University of Ghana http://ugspace.ug.edu.gh 147 the assumption that the migrants will settle down shortly after they travelled and relieved them of the responsibilities only to realize that things do not always work out as planned. 6.2 Caregivers’ Expectations The so to speak voluntary character of these total services, apparently free and disinterested but nevertheless constrained and self-interested... Almost always such services have taken the form of the gift, the present generously given when, in the gesture accompanying the transaction there is only one polite fiction. Formality, social deceit and where really there is only one obligation and economic self interest… (Mauss 1954:4) This quotation by the anthropologist Marcel Mauss in 1954 sums up the essential point I raise in this section: that caregivers’ motivations to care for migrants' children are not always as altruistic as they make them appear to be, but instead are laden with expectations. In this section, I discuss the caregivers' expectations of the migrants and the children in their care and how they experienced met and unmet expectations. The expectations of caregivers in TCRAs are embedded in the norms of reciprocity, and, as expressed in the excerpt above from Mauss, most of these acts are made with some self-interest. Some of the caregivers initially mentioned that they had no expectation of rewards from the migrants and their children, and that they were doing the migrant and the child a favour. However, my ongoing interaction with them showed that the cultural and religious motivations they attributed to their actions were only one of the reasons they took on the responsibility of caring for a migrant's children. Of course, some had higher and more explicit expectations. This was especially the case for most of the caregivers in the involved pathway, where the migration of their relative was viewed as a necessity. Some took on the responsibility to reciprocate a University of Ghana http://ugspace.ug.edu.gh 148 migrant's past acts of kindness, possibly with the hope that it would pave the way for future acts of kindness in return. 6.2.1 Features of Caregivers’ Expectations First, the expectations of both the caregivers and migrants in the care arrangement were hardly discussed at the time care was being arranged. Migrants did not tell caregivers how they expected their children to be raised, since that would almost amount to an indictment of the caregivers’ ability to raise a child. Caregivers, on the other hand, did not tell the migrants what they expected from them. Kin and non-kin caregivers alike hardly even mentioned the remittances for the child’s upkeep because this would have implied they were asking for money to take care of a child that is their child too which was to some extent culturally unacceptable The only exception was Takyi, who made it a condition for accepting to care for the children. Another common feature found in these arrangements was that some of the expectations of the caregivers were known to both parties even though they were hardly ever communicated verbally. An example of such expectation is the remittances that the migrant sends for a child’s upkeep and by extension, for the caregiver if s/he happened to be a grandparent or mother of the child. There were also some expectations that were not communicated to the migrant because they would personally benefit the caregiver. By voicing them, the caregiver might come across as openly asking for rewards for taking care of the migrants’ children, which is frowned upon in Ghanaian society. These expectations included the material support caregivers required from the migrant during a crisis as well as ordinary acts of kindness. Additionally, I observed that caregivers also had some expectations of the children that were left in their care. University of Ghana http://ugspace.ug.edu.gh 149 As mentioned in the beginning, caregivers’ pathway into care largely informed their expectations of the care arrangement. For instance, the caregivers who were actively involved in the decision to migrate and thus the decision to care for the children saw migration as a livelihood strategy for their families. They therefore had the expectations of receiving remittances for their (including the children) general upkeep, better education and health care and possibly money to start a business or build a house. These expectations were explicitly expressed when I enquired, and like Mama K., they were also happy when the expectations were met- ‘ye di di ho,’ or ‘we benefit from it,’ she told me. Figure 6.1 shows the features of caregivers ’expectations and their actual expectations. Each rectangle represented a type of caregivers’ expectation. These are discussed in detail below. Figure 6.1 Features of Caregivers’ (CG) Expectations Of migrant parent Key *CGs’ Expectations Remittances for child’s upkeep Material support (present and old age) Companionship Help with domestic chores Non- verbal Of Child Known to CG & migrant Known to CG only University of Ghana http://ugspace.ug.edu.gh 150 Remittances for Children’s Upkeep Receiving remittances for a child's upkeep is an expectation that is more in line with the purpose of migration--that is, to work and improve the family's standard of living, especially that of the children. All the caregivers in this study expressed the expectation that migrants would send remittances for their children’s upkeep. Caregivers usually spent the remittances migrants sent on education, health, feeding and clothing the children in their care. Both caregivers and migrants were aware that the migrant would provide for the material care of the child, whereas the caregiver provided proximate care or ‘direct mothering’ (Schmalzbauer 2004) which the migrant was not able to do because of distance. This may explain why, even though they hardly discussed this expectation, both the migrant and caregiver understood each other in this regard. The expectation of remittances for the children’s upkeep was generally met by the majority of the migrants, albeit they were often insufficient. The degree to which the expectation was met varied from one caregiver to another. In situations where the expectations were fully met, the caregivers admitted that the remittances helped them a great deal. For elderly grandparents who were not working, remittances from adult children who worked domestically and in particular abroad were a significant source of income. What is more, the continuous stay of the grandchildren in their care was one of the surest ways to be constantly provided for by their migrant child. Gyeiwaa, a maternal grandmother told me the following story to illustrate her point that even though adult children had the moral obligation to take care of their parents in old age, this was not always guaranteed, especially in the case of migrant children. However, the continuous presence of their children in their care ensured that this happened. I am not working, and I rely entirely on remittances from my migrant daughter and migrant sister. My sister who used to send me money regularly stopped when my daughter migrated. I have my daughter’s four children in my care. I have been taking care of them since she travelled some 14 years ago. My daughter and her husband send University of Ghana http://ugspace.ug.edu.gh 151 me regular monthly remittances with which I take care of myself and my grandchildren. Even when things are very tough over there, they still try to send us something. I am not sure if they would bother so much to send me money when they are in dire circumstances if the children were not with me. (Gyeiwaa, Accra, January 2013). The story of Gyeiwaa and other grandmothers suggest that taking care of migrants’ children is a way to guarantee financial support for grandmothers and corroborates Abanihe’s (1984) finding among grandmothers care-giving of their grandchildren in fostering that takes place nationally: [W]ealth-flow from parents to grandparents takes place in the absence of fosters but their presence guarantees regularity of such visits and the size and assortment of goods that are brought with it (Abanihe 1984: 9). Hence, grandparents taking care of their grandchildren functions as an informal support system and the occasional visits and gifts of money and food items form part of the resource accumulation for the grandparent as has also been found among transnational families elsewhere (Dreby, 2010). Caregivers who had other sources of income also found remittances very useful. To these, a remittance meant they did not have to take money from their business for the children’s upkeep. Their own money, therefore, served as a buffer when these caregivers did not receive remittances. Some of them expressed the strain that a cut in the flow or reduction in the amount received put on their own resources since they could not watch the migrants' children--whom they usually refer to as their own children--suffer in the absence of remittances. These caregivers were, therefore, also happy that the migrants endeavoured to send remittances to take care of the children. In situations where the expectation of remittances was not fully met, whether in terms of the insufficiency of the amount or when it has delayed posed very severe situations for the caregivers as we shall see in Chapter 7. University of Ghana http://ugspace.ug.edu.gh 152 Having discussed the expectation common to all caregivers, I move on to more specific expectations. Expectations for material support In Ghana, people in both urban and rural areas can hardly rely on formal or state-initiated social provisions. This is why they usually rely on each other for support, especially in times of crisis. In addition, migration and development studies usually point to migrants as one of the main providers of security for their relatives and sometimes friends in the home country (Kabki 2007; Smith 2007). It is also known from transnational migration literature that those left behind usually ran errands, supervised projects or cared for children of the migrants in order to earn their support (Smith 2007; Mazzucato, 2010). Even though some of the caregivers in my study mentioned that they did not have any expectations especially of financial support for taking on the role, it is possible to argue that the fact that migrants were a better source of material support partly informed their decision to take care of their children. The expectations for material support for the caregivers have been grouped under two categories. First there is material support which goes into taking care of present needs such business capital, children’s education and health-related expenses. The second category is material support for caregivers in their old age. The former is expected of the migrant parents and the latter largely of the children in the caregivers' care. The difference between this expectation and general expectation of remittances is that the expectation for material support is mainly for the caregivers’ personal needs whereas the latter is purposely for the children’s upkeep. Expectation for material support was also not as fixed or permanent; they were expected to be met as and when it became necessary unlike remittances for daily upkeep of the children. University of Ghana http://ugspace.ug.edu.gh 153 The expectation for material support can be described as latent since the caregivers in my study did not make them explicit. However, I gathered the caregivers had these expectations from the excitement or disappointment they expressed when they talked to me about material support from the migrants. The experience of Awurakua, a maternal aunt who was raising her sister’s daughter demonstrates this: 'I appreciate the fact that they have been paying for my son’s school fees for some time now. It helps a lot, even though I have made other provisions for him' (Awurakua, Accra 2011). Awurakua explained that she did not ask them to do this and was quite pleasantly surprised that they decided to do so. Awurakua owned a vibrant business, and even though she was not expecting this gesture from the migrants, she still acknowledged that it was very helpful. Another example is the case of Maame Ama who said she was very happy that her migrant brother always sent some extra money to cover the expenses of the child and added an extra amount for her, especially on the rare occasions on which she needed help paying her children’s school fees. She said to me ‘nsesa ka dabiara', meaning there is always surplus. There were also cases in which caregivers were quite displeased with the way the migrant treated these ‘unspoken’ terms of the exchange In some cases, caregivers did not receive the expected material support from the migrant when they really needed it and that was how the latent expectations became visible. Kessewa’s situation illustrates this: At the time the first child was brought to me, I was running a shop in Kumasi. I was selling bags of rice, vegetable oil and spices. Business was good then. The boy was ill all the time. In fact, it started while he was with his mother abroad. She told me he had been sick since childhood. Somebody advised me to send him to a prayer camp in Accra. I discussed this with my sister and she agreed. When I had to leave for the prayer camp with her child, I left the shop in the care of somebody, and by the time I got back almost a year later the shop was closed. All the items were gone and the girl was nowhere to be found. I had lost my entire capital and the shop too since I could not continue paying for it I told my sister about the state of my shop upon my return and I expected that she would suggest sending me some capital to restart the business, but she did no such thing (Kumasi, July 2012). University of Ghana http://ugspace.ug.edu.gh 154 According to Kesewaa, she relied on friends and secured a loan from a micro-finance institution to restart her business. She says her sister has not even asked her how she managed to get the business started again. She admitted to being very bitter about her sister’s attitude. One might argue that unexpressed expectations could never be met. But this argument can be resolved by referring to the cultural norms of reciprocity. An Akan proverb which captures this aptly says, 'Benkum dware nifa na nifa nso edware benkum’, or 'the left hand bathes the right and the right also bathes the left'. With this understanding of one's role regarding the other in a migrant-caregiver relationship, it was not necessary for the caregivers to explicitly express their desire for the migrants to reciprocate the favour they do them. Another common feature of this practice of reciprocity I found was that the frustrations of the offended party might never be disclosed to the other because it might be misunderstood as openly seeking a reward or an offer of kindness in return for what was given. This proved to be the case for some of my interviewees who told me that they would never ask the migrant to do 'what they should know’ because the migrant might think that they accepted the arrangement for money or personal gain. One caregiver, Kumiwaa, mentioned that she thought the migrants (her sister and husband) would send her money to start a business since the children were now grown up, and she was idle most of the time. The migrants, however, had not mentioned anything of the kind to her. When I asked if she had mentioned this to them before, she said yes but admitted that she had said it only in passing in the hope that they would understand and act accordingly. This way of presenting one’s expectation to the migrant was common as seen in the cases of Koo and Kesewaa. They emphasized that they did not voice their expectations of financial support as they explained the migrants ‘knew what to do’. Hence, while, they expected the migrants to help University of Ghana http://ugspace.ug.edu.gh 155 them in return for their ‘good deed’, they did not make this explicit. This is a reflection of the cultural norms of reciprocity. Caregivers also held out the hope of some form of social security or being cared for in old age when the children under their care had succeeded in life. The act of providing material support for older people especially those who contributed to your upbringing is one of the cherished cultural practices in Ghana and there are many adages and folklores in almost every Ghanaian language to emphasize its significance (see chapter 3). Some caregivers expected intergenerational reciprocity. As one caregiver put it: If one of these children even remembers to send me a handkerchief 6 when they leave I will still know that my labour was not in vain (Cecilia, Accra, January 2013). Kwesi illustrated this expectation of intergenerational receiprocity by quoting an Akan proverb, 'obi nnim nea obehwe obi daakye meaning no one knows who will take care of someone in the future. Many Ghanaians, especially, those who are parents, use this proverb when talking about child-raising. After quoting the proverb, he added that: 'the boy (his charge) is my uncle’s son, so if I do anything for him I would not have wasted my time. Moreover, If he does well in future, I may also benefit because I have a very young family, and he may one day remember what I have done for him and decide to help my children too in case I am not able to do so myself'. Other caregivers also expressed the desire to be supported in the future by the children they cared for, but also added that they were careful not to depend solely on this prospect since human 6 Handkerchief is used figuratively here to signify the children giving back to her in return for her taking care of them up and they can only do this when they are adults and are successful. This use of handherchief is this context is common in Ghana. Another item that is most commonly used is an ‘office pin’ so it is very common to hear expressions like, I have sacrificed so much to take care of this child and now that he is successful he has not even given me a ‘pin’ for my efforts. But what they mean in reality is not a pin or a handkerchief. University of Ghana http://ugspace.ug.edu.gh 156 beings could not be completely trusted. All these expectations conform to the norms of reciprocity which is very pronounced in many Ghanaian societies. Expectation of Companionship The expectation of companionship was present primarily among grandparents and other elderly caregivers whose own children had grown up and left home. In the case of the grandparents, the children’s presence not only constituted a guarantee for resource accumulation, but also for companionship. One grandmother told me how the children had filled the vacuum that her son’s migration had left: I love my son, and we were very close. Believe me, when he was travelling, even though I knew he would be visiting, I could not stand the separation. I used to miss him a lot. I was, therefore, very happy when he started sending the children over to Ghana when they were born. First it was the girls, then much later the boy. I don’t miss my son that much anymore. They make me so happy (Fosua, Kumasi, January 2012). Some caregivers also talked fondly about how the children, especially the young ones made them laugh by some of the things they made or talked to them about. Cecilia said one of the five children in her care, who was six years old, returned from school and realized that she was still cooking. She said 'he looked at me and said aaah, grandma what have you been doing all day that you still haven’t finished cooking. I am very hungry now, what do I do’? She recalled that she was so surprised and amused by his reaction that she called his mother to tell her about it, and they both laughed over it. She added how the two youngest children in her care would usually fight over who would sit on her lap or who she should bathe first. Other caregivers laughed with pleasure as they recounted episodes they had had with the children in their care, such as a child promising to buy a red car for the grandmother or another University of Ghana http://ugspace.ug.edu.gh 157 child telling his grandmother that her house was not nice, so he would build a nicer house for her when he grew up. Sometimes, as a researcher, I could tell that taking these children from the grandparents would be quite a painful experience for them. They had gotten used to having them around. Fortunately for some of these grandparents, the likelihood of the children being reunified with their parents was very low because of the parents’ situation in the host country (i.e., they are undocumented or have low paying jobs or both). As companionship was important in grandmothers’ experiences with caregiving it could lead to loneliness when the children were taken away especially if the children had been with them for a long time as was the case of Owontaa. The first time I met Owontaa she was quite sick. She had suffered a stroke and also had diabetes. Her hearing was failing, and she told me she was worried that the children (19 and 12 years) were not receiving enough care because of her state. She was more concerned about the younger child. She complained about not being able to cook ‘proper’ meals for them. She had been taking care of the children since their parents brought them back to Ghana when they were babies. She repeatedly said throughout the interview that the children’s parents should let the children join them in the host country. Inasmuch as she would have loved for the children to stay with her, she felt she could no longer take very good care of them, especially the girl. She believed her granddaughter needed more attention than what she could offer her in her state. A few months later, the parents succeeded in bringing the children to the host country to live with them. Almost a year later, I went back to visit Owontaa and she was alone in the house. She told me how lonely she was, and how much she missed the children, especially the girl with whom she used to share the same bed and room. She could not stop talking about her granddaughter. Owontaa told me that the girl liked to play soccer and she loved to watch her play and also University of Ghana http://ugspace.ug.edu.gh 158 enjoyed compliments from admirers who thought she had done a good job of taking care of the children. Expectation of help with domestic chores When caregivers had older children in their care, they often helped them with chores in the household. The grandparents in the study often talked about how helpful--or otherwise--the children in their care were. Children providing domestic help to parents, especially in a fosterage context, were more prevalent in traditional fosterage situations where fostered children provided small services and did chores in exchange for the care they received from foster parents. In TCRA situations, such help by children is, however, minimal and, in some cases, absent. The reason for this is that in TCRAs, the migrant sends remittances for their children’s upkeep, whereas in traditional fosterage, the caregivers do. Migrants, therefore, are usually unhappy when their children engaged in domestic chores. According to the caregivers, migrants often told them that they did not want the children to engage in domestic chores at the expense of their education. A grandmother, Maa Fosua told me that her migrant son had asked her not to allow the children to do household chores. For this reason, her son began sending enough money to pay for the services of a house-help. The migrants’ concerns, notwithstanding, some caregivers, especially grandmothers, relied on their older grandchildren for help. During one of my interviews with a grandmother, Mankpa, her oldest grandchild (17) promptly administered her medication on two occasions within a period of two hours. Mankpa had been taking care of this girl and her three younger siblings since their mother travelled almost eight years earlier (as of the time of the interview). University of Ghana http://ugspace.ug.edu.gh 159 Such household services were not only useful for grandmothers, especially convalescent grandmothers, but even the fairly younger and more active caregivers had the expectation of receiving help in the home once the children grew older. This was the case of Awurakua who had been raising her sister’s daughter for over 12 years. The child had been with Awurakua since she was two years old. At the time of the interview, she was 15 and helped Awurakua with household activities such as cooking which was one of Awurakua’s principal responsibilities in the house. She could afford to stay longer at the market because her niece could prepare what Awurakua described as less complex food for her siblings when she came home from school. She could also participate in other social activities, because together with her husband, her niece could take care of the other children in the house. The reverse situation when the expectation of providing domestic services was not met was quite frustrating for the caregivers, especially the grandmothers. In the few instances where I encountered such situations, the children in question were boys. This is not too surprising in the Ghanaian context since boys are usually not brought up to be useful in a domestic setting. Georgina, expressed how frustrated she was by the fact that her grandson could not cook rice for them to eat for lunch. She complained that [t]he only problem I have with this boy is he does not know how to do any domestic chore. Today, for instance, I am not feeling too well and the house-help has left for school, and we are both hungry. So I asked him to cook rice for us to eat, but he said he couldn’t and that he did not know how to. I am also not ready to go and cook for him in my state, so we have all found something else to manage the hunger until the girl returns from school. Meanwhile, if he were a girl, at the age of 14 she would have cooked this rice for us to eat (Kumasi, August 2011). What Georgina did not mention explicitly was whether she had taught the boy to do household chores. She did not express surprise at the boy’s inability to help she was only University of Ghana http://ugspace.ug.edu.gh 160 frustrated because she need that help badly. Which raises the questions as to whether she was not perpetuating this old cultural practice herself? Mama K. had similar experiences with her grandsons. She told me that her grandsons were rarely around to help her with anything. She became quite upset as she explained that she had been sick for some time and the boys were never available to take care of her. She complained bitterly to me that they were very mean and sometimes refused to run errands for her. She also said they were more cooperative when they were younger. From these, I observed that whereas girls tended to become more responsible and helpful as they grew older, it was quite the opposite with the boys. I often found this to be the case in a lot of the situations I encountered. Perhaps both boys and girls became more conscious of their gender roles as they grew older, so girls realised they had to be more ‘domestic’ and boys found out that the domestic sphere was not their space. As a result, caregivers who had older girls benefitted more than those with older boys. 6.3 Summary In this chapter, I examined some aspects of caregivers’ experiences in TCRAs, namely their pathway into care and the expectations they had of the care arrangement. In my discussion, I came up with two pathways which caregivers followed into their care arrangement: ‘involved’ and ‘uninvolved’. First, there were caregivers who were consciously involved in the decision to become caregivers, because they had a vested interest in the decision to migrate, had weighed their options and saw that taking on the role would be beneficial or considered it the proper thing to do for the migrants and their children. These caregivers followed the ‘involved’ pathway into their care arrangement. I called the second route to caregiving the ‘uninvolved’ pathway which University of Ghana http://ugspace.ug.edu.gh 161 was followed by one caregiver. I acknowledge that the two pathways are very lopsided as only one person followed the uninvolved pathway. This chapter also argued that caregivers’ pathways were informed by an array of factors, such as normative obligations, where caregivers chose to take care of children because it was the proper and acceptable thing to do. Caregivers' pathways were also informed by the past histories of their relationship with the migrant. The third factor identified in the chapter, which had the least influence on the caregivers’ pathway was the capacity to care. Subsequent chapters will show that the neglect of the latter factor by both parties created inconveniences for caregivers. The difference in pathways notwithstanding, all the caregivers had some form of expectations of the care arrangement. However, one’s pathway, which represents the level of involvement in the decision to be caregivers, influenced how they considered and to whom they expressed their expectations. Whereas those who had a vested interest in the migration decision had more concrete expectations and openly talked about them when I asked, those who saw their roles as altruistic, devoid of any expectations and the proper thing to do could not admit to the migrant and to me that they had expectations from the migrants and their children. Their expectations could best be described as hidden and it took cases of unmet expectations to make these latent expectations visible. University of Ghana http://ugspace.ug.edu.gh 162 CHAPTER SEVEN: ‘W’A YE FINE’ – ‘PROVIDING CARE’ AND ‘MANAGING CARE’ IN TCRAS 7.0 Introduction In this chapter, I examine what goes into caregiving in TCRAs. I discuss two aspects of caregiving which I have termed ‘providing care’ and ‘managing care’. In her work on transnational families in Guatemala, Moran-Taylor (2008) noted that caregivers said migrant parents were most interested in their children being well fed, well cared for, and kept out of trouble. What Moran-Taylor's and others' work on caregivers did not address is the question of what precisely is involved in making sure the children are well fed, well cared for and kept out of trouble. This is the subject of Chapter 7. 7.1 The concept of care in TCRAs In this study, the concept of care comprises caring activities that includes feeding, bathing, taking care of the children when they are sick, ensuring they attend and do well in school and making sure they are well-behaved. I present the tangible caring activities which I refer to as ‘providing care’ that caregivers undertake which echo the concept of care that considers caregiving as an outward show of the caregiver's concern for the recipient of care (Dressel and Clark, 1990). I argue that there is an entire array of invisible or unobservable practices that take place which ensures that caregivers are able to engage in caring activities for the children left with them as well as enable them to keep the TCRA functioning well. Dressel and Clark (1990) also observed that sometimes care involve activities that are not observable. These include caring feelings (Folbre and Nelson, 2000), such as those of concern or affection for the cared-for. These unobservable aspects of caregiving are said to assure the effectiveness University of Ghana http://ugspace.ug.edu.gh 163 of care provided to the recipients (ibid). In this study, I describe another aspect of unobservable caring that goes beyond the caregivers’ feelings for the cared-for to include invisible activities caregivers engage in to ensure the effectiveness of the provision of care in the TCRAs as seen by others. These aspects of care are invisible because they mostly go unnoticed by the migrants, the children and people outside the TCRAs who are usually the ones who appraise a caregiver's performance. I refer to this invisible care practice as ‘managing care’. 'Managing care' entails managing resources and relationships with others in order to provide care effectively. They comprise activities such as managing insufficient remittances, communicating within the TCRAs (with a greater emphasis on what information to withhold), dealing with the assumptions of outsiders, handling a change of lifestyle and managing the unforeseen in TCRAs. To have a well functioning TCRA, in which the children are properly taken care of, there must be a cordial relationship among the members in the triad, devoid of tension and mistrust. In other words, there must be a balanced interplay between providing care and managing care. I conclude that the success of the former is largely dependent on how the latter is carried out. 7.1 Providing Care Providing care entails daily care practices such as feeding, bathing and caring for young children, taking care of them when they are sick, ensuring they go to school, helping them with schoolwork and making sure they behave in a socially acceptable manner. In most cases, the caregiver and the migrant in this study did not discuss these kinds of activities before the care arrangement was finalized. But these tasks are the basic kinds of things that every migrant expects a caregiver to do for his/her children. As mentioned in Chapter 6, most of the caregivers University of Ghana http://ugspace.ug.edu.gh 164 were aware that these kinds of tasks would need to be done daily, and, in some cases, they had already been providing this kind of care before the parent's migration. Caregivers were ‘appraised’ on their ability to provide care. This was the most tangible way of judging whether they had done well and deserved to be rewarded, or had failed in the duties expected of them. The migrant parents could tell that a caregiver had done well when a child looked well fed, was properly groomed in both manners and appearance, did very well in school (attendance and performance) and possessed the peace of mind to concentrate on his or her schoolwork or apprenticeship. The remark I often heard when everything looked good was 'w’aye fine', meaning looking well or looking good. Caregivers often told me this was how the parents described their children when they saw them or their photos. The parents were very grateful to the caregivers because the children looked so well. The parents I met in the host country also often used the expression ‘w’aye fine’ after receiving the pictures I had sent them from the caregivers and children. In addition, the caregivers often told me that when migrant parents received good reports about their children’s academic performance (both written reports at the end of the school term and oral feedback from teachers), they felt the caregivers were taking very good care of them. The day-to-day activities caregivers carried out were largely dependent on the age of the children, and they evolved during the course of the care arrangement. Some caregivers received children at a very tender age and, because the children had grown up in their care, had already gone from the nurturant to the supervisory stage of caregiving. At the nurturant stage, a child needs more time and attention because it is totally dependent on the caregiver. Some caregivers were responsible for bathing, feeding, toilet training and getting the children ready for school everyday as well as finding schools for them to attend. One caregiver, a maternal aunt of a University of Ghana http://ugspace.ug.edu.gh 165 migrant who left all five of her children in her care, described her daily routine to me. Two of the five were in their late teens and were boarders at a senior high school. The youngest two were a three-year-old toddler and six-month-old baby. At the time I began visiting them, the youngest child had turned two and had started day care. The caregiver, Cecilia, recounted her daily routine: When I wake up in the morning, I bathe both children and prepare breakfast for them and the eleven-year-old and get them ready for school. The 11-year-old attends the same school as the three-year-old so they leave together and I send the youngest child to day care. I start to clean after them when they leave. The room is always messy after they leave in the morning. It is my room as well and, as such, I cannot live in the mess, so I have to clean it every day. I also have to wash clothes everyday because of the two younger children. It is almost midday by the time I am done. Then I have to start preparing their evening meal as they eat lunch in school. By the time I am done cooking and taking care of some stuff for my aged mother, they will be returning from school. When they get back, I get back into full gear minding them, feeding and bathing them and preparing them for bed. I have been doing this everyday since the children were sent to me (Cecilia, Accra January 2012). On one of my visits to another caregiver’s house, the boys in her care had just returned from school. She had already prepared their meal and had to stop during our interview to attend to them. I watched as she served them their meals and the youngest child (nine) asked her to mix the rice and stew for her. At some point she had to pause during the interview to urge them to finish their food because one of them had left the food to get his soccer ball. She cleared the table after they were done eating and told me that she usually must also check their room to see if they have hung their clothes properly. This caregiver had had only a primary education and could not help the children with their schoolwork. Thus, she had to consult their parents to arrange private tuition for them at home. Another grandmother, Maa Asafo, who was raising eight grandchildren from her three daughters who were abroad, also mentioned that she woke up at 4:30 a.m. everyday. She took University of Ghana http://ugspace.ug.edu.gh 166 her bath first and then woke the children up. She said the most difficult task in the morning was getting all the children to wake up; sometimes she had to threaten them with a cane. The older ones took their baths and then helped bathe the younger children while she prepared breakfast for them. She succeeds to get them all to leave the house by 7:30 a.m. She also had to walk the youngest ones to their schools. Like Cecilia, she has a regular routine of cleaning up after the children leave for school and then preparing meals for them before they return home. Once they are back, she helps those she can with their schoolwork, while the others are given extra lessons by a teacher she hired. Maa Asafo said that she was busy everyday of the week, and that their care got more tiring on weekends because they were home all day and she had to prepare three square meals for them in combination with shouting and yelling when they misbehaved. But on weekends, the older children were at home and helped her with the younger ones. She told me how she went into a PTA meeting to criticise a proposition that the children in junior high school should have extra classes on Saturdays. She explained: I told them straight at the meeting that I am not in favour [of this measure] and none of my grandchildren will attend. I added that, 'you people keep the children from 7:30 a.m. till after 4 p.m. every day. I hardly get them to help with any chore during the week, and it's only on Saturdays that I get some help from the children. You are proposing to take that away too? They did not heed this and still went ahead. I don’t allow the children to attend. There’s a lot of work around the house on the weekends. (Maa Asafo, Kumasi, May 2012) When caregivers were left with young children, as was the case for some of my respondents, they depended on their own older children or on those living in the same compound (who were usually family members) to help with nurturant care for the young children. In the absence of these helpers, some of the caregivers who were engaged in active commercial activities had to stop to take care of the children because they could not manage both. Cecilia, said that she had to close her shop for a few months after receiving five children to care for, but University of Ghana http://ugspace.ug.edu.gh 167 later shut it down completely because she could not pay enough attention to it because of the demands of her new caregiving task, which were both overwhelming and unexpected. Other caregivers who were actively engaged in commerce and did not want to abandon that activity because it was their main source of income got house-helps to assist them with the provision of day-to day-care for the children. The task of caretaking was less strenuous and less time consuming for caregivers who were left with older children or when the younger ones in their care grew older and were able to take care of themselves in terms of personal grooming. Some of these children were even able to cook as well as provide some small services for the caregiver. In cases in which the caregiver had other young children to care for, these children were also able to help with their care. Most older children required only supervision and guidance. While caregiving for these children was less tiring, there could be some challenging moments, especially during the transition from junior to senior high school. During my research, there were a few such moments. This period of transition could be quite hectic and time-consuming for caregivers. For example, Cecilia had to find a senior high school that would admit the two teenagers in her care. During one of my visits, she told me that she had spent almost every day of that week chasing schools outside of Accra for the children. She finally succeeded in getting them admitted into two separate schools. Another caregiver I spoke with had gone through the same ordeal for the boys in his care. He is self-employed and had to lock up his business for an entire week in order to go to all the different regions looking for a school for them. Another situation that demanded the caregivers' full attention and energy regardless of the children's age was when they fell ill. Caregivers spent a lot of their time taking care of children when they were ill. During my research, caregivers usually told me that they were very University of Ghana http://ugspace.ug.edu.gh 168 grateful that the children in their care seldom fell ill. They did not have to deal with the challenges or issues that came with ill health, except for the occasional fever, which was usually treated at home without seeing a doctor. However, there were some caregivers who had had experiences with taking care of their charges when they fell ill. One of the grandmothers mentioned that her grandson who was sent to her from abroad was very frail and was almost always sick when he was young. He easily caught infections which made him sick for many days. She explained that she did not allow anyone near the child for fear that he would catch an infection. This also meant that she had to suspend going to the market in order to pay more attention to the child because she could not allow other people to come near him. Kessewa, a maternal aunt in Kumasi also described how she had to leave her house and her shop at the market to spend over a year at a prayer camp with her sister’s son who was seriously ill. Her sister, who was living abroad, sent the child to her because of his constant ill-health. Her sister sent him back to Ghana so they could try other means of curing him than the conventional medical channels they had used in the host country. Someone had told Kessewa about a prayer camp in Accra where prayer was used for healing the ill. With the approval of her sister and husband, she left for the prayer camp with the boy. They spent an entire year there. This meant, however, that she had to leave her business in the care of someone else, and when she returned it had collapsed. During the year she spent at the prayer camp, she could not take part in any of her usual social activities (e.g., going to church meetings and funerals, visiting or communicating with her family and friends) because of how far it was from her home and its strict rules which required spending a lot of time in seclusion to concentrate on the spiritual exercise. This cut her off from her family and friends. She was, however, grateful that the spiritual exercise yielded results because the boy has since been well. But every once in a while he falls ill and has to be University of Ghana http://ugspace.ug.edu.gh 169 rushed to the hospital. 'I can’t tell you the number of times I have had to sleep in the hospitals both before and after the prayer camp. Only after the prayer camp, the hospital admissions have not been frequent and we spend fewer days in the hospital (Kessewa, Kumasi, June 2012). Ophelia, the paid caregiver, also recounted the story of how all three of the boys in her care fell ill in succession. The first boy was admitted to the hospital three times in one month. She told me '[t]he boy was almost gone, and I was so scared partly because I was losing the boy and was also thinking of how I would explain this to the parents'. She had been keeping them apprised of the boy’s ill health, however, and they recommended seeing a doctor. But she was still thinking of how everything would turn out if the boy died, which at the time appeared very likely. She left the other two children in the care of her friend and the migrant’s younger brother. Describing the boy's illness as a spiritual attack, she told me that she had gone through hell during the period he was ill. The experience had taken a big toll on her. She had lost weight from the emotional agony and the continuous fasting that she and her pastors did as a means of averting the spiritual attack that was causing the boy’s ailment, while the doctors continued to battle the boy's illness in the hospital. All the while, she was trying to calm the boy's migrant parents and asking for her pastor to call regularly and pray with them. Having described the tasks that the caregivers routinely did for the children in their care, I now move to cases in which they were not able to accomplish a task. There were some instances, albeit only a few, in which caregivers admitted being too busy to undertake activities such as attending PTA meetings or visiting children in boarding schools, which both parents and the children deemed very important. Some of the grandmothers in my study who felt too old to attend PTA meetings asked others to attend on their behalf. But in other instances, caregivers had not bothered to find someone to go in their stead. I also met a grandmother who told me that University of Ghana http://ugspace.ug.edu.gh 170 because her grandson had become very stubborn and did not run any errands for her, she had lost interest in paying attention to him. She admitted that she did not really care what he ate and had stopped giving him pocket money for school. She said: Can you believe that I share the same room with the boy, yet he expects me to clean up the room? Even when I am sick, he does not show any concern. All he does is stay out with friends. I won’t take money out of my purse to give to him or worry about whether he eats or not. I even told his mother about it. She is not happy about it, but that is what it is. I am tired of this boy (Mama K, Kumasi June 2012). Mama K has been raising the boy for 14 years now. He was only two years old when he was sent to her from abroad. Having looked at what providing care entails does not imply that these tasks are exclusive to caregivers in transnational child raising arrangements. One may argue that parents perform these tasks for their children regardless of first, whether they are in fosterage or not and second whether one or both parents is an international migrant. That is a valid argument. However, this study argues that in the settings of TCRAs, providing care is more challenging because of the unique circumstances under which child care is arranged – child fosterage in the context of transnational migration. Providing care under this arrangement is thus shaped by the convergence of child fostering norms and transnational migration. In the traditional fosterage situation that does not involve an international migrant, the caregivers are usually the sole financiers of the children's care and the ones responsible for making major decisions concerning their lives, as has been noted in the literature on child fosterage and is known from caregivers’ own histories about their upbringing. In TCRAs, the situation is almost the reverse. In these arrangements, caregivers share the rights and responsibilities to raise the children with the migrant parents, who, through remittances and communication technologies, try to parent from afar. In some cases, the migrant parents are so involved that caregivers feel they are being stifled University of Ghana http://ugspace.ug.edu.gh 171 in their efforts to provide effective care. Caregivers in TCRAs are also faced with the pressures and anxieties induced by societal expectations about how such caregiving should be adequately carried out as well as the implications of transnational migration on their role as caregivers. To negotiate their rights and responsibilities in these caregiving arrangements and still succeed at providing adequate care for the children, it becomes expedient for them to ‘manage care’ 7.2 Managing Care In this section, I turn my attention to managing care. What goes into a caregiving situation to make it 'w’aye fine'? What are the things that caregivers have to juggle on a daily basis to earn this complimentary remark? One word that caregivers frequently used, especially when asked about how they dealt with the task of providing care was ‘manage’. I heard this word very often, for instance, 'the money is never enough, but I manage', 'the children are stubborn but I manage', or as they said in the Twi langauge 'me managee' (I am managing), or 'me ye no nkakrakakra' (I do it little by little), which also implies managing a situation. I thus came to the conclusion that caregivers are very good managers, and that a lot of management must take place daily to arrive at a state deserving of the expression 'w’aye fine'. What goes into managing care may never come up in the discussions (if there are any) that precede the care arrangement. I have identified five broad categories of situations that caregivers managed in their daily experiences as caregivers. These are managing remittances, managing communication within and outside the TCRAs, managing the assumption of outsiders, managing a change in lifestyle and managing the unforeseen in TCRAs. Caregivers always had a fair, if not a concrete, idea of what went into providing care. However, when it came to managing care, it was not the same. University of Ghana http://ugspace.ug.edu.gh 172 In fact, most of the time, caregivers only realised the import of managing care in the process of providing care. 7.2.1 Managing remittances Since most parents travelled in order to provide a better standard of living for their children, it was an unwritten code in the care arrangement that parents would send remittances to provide for the upkeep of the children they left behind. As already mentioned, the caregiver and the migrant hardly ever discussed the amount and frequency of remittances. Instead, the terms of financial support were usually left to the migrant's discretion and depended on the conditions of the migrant in the host country. Caregivers made me understand that they knew that when people travelled, it would take them a while to get settled in the host country. So for the first few months and, in some cases, years, the caregivers had to bear the cost of caring for the children. During the initial years following travel, some migrants did send money from time to time to help pay for school fees. But the cost of many other things, such as feeding, clothing, and the healthcare of the children were borne by the caregiver until the migrant was considered to have settled in the host country. Even when migrant parents were able to remit more than they had during the initial stage of migration, caregivers often told me that there was never enough money to take care of the needs of the children. Three out of the 34 caregivers in the study admitted that the money they received from the migrant parents was sufficient for taking care of the children, and there was sometimes surplus to keep for themselves. All the remaining caregivers had issues with remittances. They complained about the insufficient amount and irregular flow of the remittances. Below I present the case of Cecilia to illustrate this point because she was the only caregiver who mentioned how much she received in remittances for the upkeep of the children. University of Ghana http://ugspace.ug.edu.gh 173 Knowing the amount enabled us (Cecilia and myself) to tally the amount of her expenditures against what she received from the migrant parent. Ghanaians are not comfortable discussing money (salary, allowances and remittances), and this made it quite difficult for me as a Ghanaian to even ask the caregivers about their remittances. Nevertheless, I found ways of asking about how much they received, and most of the responses were ‘enso hwee’ or ‘esua’, both of which mean that the remittances were not enough. Two of the three caregivers who said they received enough in remittances used the phrase ‘oo eye’ (it is enough), while the other said 'nsesa ka da biara' (there is always surplus). I rarely ever got an exact figure of how much was sent to the caregivers. Cecilia mentioned the amount even before I asked her because on the first day of our interview she vented her frustration about the remittances and in the process mentioned the amount she received monthly for the children. Below I present her case on insufficient remittances. Cecilia cared for the five children of her migrant niece. At the initial stages of my fieldwork (2011), she was receiving 200 GHS (75 Euros) each month to cover the daily expenses of caring for the children. At the end of the data collection in July 2013 the amount remained 200 GHS. According to Cecilia, her niece said because two of the children had left for Boarding schools, the amount should be enough for the remaining three children. She, however, did not increase the amount when the two children returned from boarding school. The log sheet below is her expenditure when all five children were at home. This amount did not cover school fees. The 200 GHS was meant to last a month. However, according to Cecilia, the money barely lasted 10 days. The table below captures Cecelia’s regular weekly expenditures for the children. Table 7.1: Cecilia’s weekly expenditures (in GHS) Item Unit cost Quantity Total cost Remarks University of Ghana http://ugspace.ug.edu.gh 174 Milo 7.50/400g 1 7.5 Usually served three times/week Sugar 1.5/200g 1 1.5 Milk 1.40/tin 5 7.0 Used together with Milo and on weekends when they have rice porridge/pudding Hausa Koko 1/portion 8 8 Twice a week. Does not require milk. The two youngest children shared one portion Bread 2/loaf 5 10 Every day with the exception of weekends when they had rice porridge /pudding Rice 3.5/kilo 5 17.5 This includes the amount used for rice porridge. They had other rice dishes twice a week Oil 4/litre 1 4 Meat 8/kilo 2 16 Fish 25 each 1 25 This includes both frozen and smoked fish, which is usually not sold in kilos Tuber crop (yam) 3 each 2 6 Twice a week Vegetables 20 1 20 The amount is a lump sum because of the low cost of basic vegetables (tomatoes, pepper, onions, okro) Corn dough 4 1 4 To prepare banku thrice a week. Cecilia says this was very economical, and there was usually some left over for lunch the next day. Toilet soap 2/cake 1 2 University of Ghana http://ugspace.ug.edu.gh 175 Toilet paper 1/roll 2 2 Laundry Soap 2 1 2 Half of a full bar (long bar) that cost GHC 4 Feeding fees 1 5 5 The youngest child in daycare Snacks for school 0.7 5 3.5 For the youngest child to send to school. It was compulsory. Total 141 *(47 Euros) * The exchange rate at the time (July 2013) was GHC 3: 1 Euro. These items constitute the barest minimum Cecilia could buy with the amount she received monthly from her migrant niece. She told me that until we calculated the costs together, she had not realized she was spending so much. In addition, many of the costs in the calculation were estimated. Some of these may have been overestimated, but it was more likely that most fell below the actual cost. In addition to this weekly expenditure, she also incurred monthly costs, such as for having the children's hair cut at a cost of GHS 1 for each. She added that at some point her niece insisted that she have the hair of the third girl relaxed. 'I tried for a while, and then I stopped because I could not afford the cost of the relaxer and maintaining her hair, and her mother was not happy about that'. She also had to buy three packs of sanitary towels at GHS 2.50 each for the three teenage girls every month. In addition, there was an increase in the cost of utilities, especially electricity. All three girls owned cell phones which meant regular battery charging. Besides this, almost every electrical gadget was active when they were at home. Once when I visited their house, I witnessed first-hand what Cecilia had complained about. The children were at home because of a school break. They watched television during the two hours University of Ghana http://ugspace.ug.edu.gh 176 I spent with her. She told me that they watched television throughout the day, switching from one soap opera to the next. The television was on all day as was the standing fan. The cost of buying electricity units, she said, had increased from GHS 11 per month before the children came home on holiday to almost GHS 50 per month. Cecilia added that there was also the occasional special treat on birthdays that demanded she cook a special meal and buy soft drinks. She told me her migrant niece promised to send a small remittance for the children's birthdays because she insisted that she treat the children to something special on that day. She sent money for the first two birthdays, but then stopped. But because her migrant niece had made her start the practice, Cecilia had to continue it for the other children. However, she only spent what she could afford on the other children's birthdays. She also mentioned that since the children left for boarding school, occasionally they would call her saying they had run out of food or pocket money, which Cecilia replenished even when their mother had not sent money for this purpose. I witnessed one such call on my last visit to the house. Meanwhile, Cecilia's migrant niece insisted that since the two girls had left for boarding schools, the money (GHS 200 per month) should be more than enough because there were two fewer mouths to feed. How Cecilia managed her insufficient remittances At the initial stages of the care arrangement, the remittances Cecilia received from her migrant niece were supplemented by a monthly allowance her two adult daughters gave her after they depleted the revenue she had earned with her shop. Later, however, the daughters threatened to stop paying her the allowance because they claimed she spent all the money on the children in her care and was always in need of more. To avert this situation, she devised a University of Ghana http://ugspace.ug.edu.gh 177 different strategy. She offered to do her daughters’ weekly grocery shopping at the market for them because they were too busy to do this. She then added their money to the amount she allotted for shopping and bought most of what she needed in bulk. She shared what she bought with her adult daughters according to how much she thought they needed. She obviously took the largest portion for herself because her daughters had very small families and did not need much. She therefore got more from the small amount of money she had contributed to the shopping. She took this a step further by offering to cook meals for her daughters. On my last visit, I joined her in the middle of the compound in which she lived, where I found her cooking different dishes for her daughters. She explained that at the time she was ready to go for the next grocery shopping, her daughters told her that because they did not have time to cook, they still had some groceries from the last shopping. This was a huge loss to Cecilia because it meant she could not buy in bulk at a lower cost, and the reason she decided to start cooking for them after the subsequent grocery shopping. Cooking for them served two purposes: first, it ensured that the groceries she bought for them were used up so she could maintain her weekly shopping schedule, and, second, the three young children in her care could have part of the meals she prepared for her daughters for dinner, leaving a surplus for lunch the next day--all without her having to touch her own money or groceries. This was her new strategy for coping with the limited resources she had available to her. Cecilia was not the only one who complained about insufficient remittances or who developed strategies to deal with them. Caregivers created a number of managing strategies, including using their own resources to supplement remittances, borrowing from friends and their adult children, negotiating with school authorities when remittances for school fees were delayed University of Ghana http://ugspace.ug.edu.gh 178 or not sent at all or, like the grandmother discussed below, skipping meals to save money to care for the children. Gyeiwaa explained: The child was four months old when my daughter left. She couldn’t send me money for the first three years that she was gone. She said she first had to work to pay back the expenses the person who worked on her migration documents incurred and must send the little that was left to her husband to take care of the three other children. At the same, time she was arranging for her husband to join her. I was living on the remittances my sister sent and that had been drastically reduced because my daughter had travelled. Many times I had to buy infant formulae on credit to feed the baby. There were times I had to starve so I could save the little money I had for the sake of the children. I could not work because I was still recovering from surgery at the time my daughter left. That was the most difficult time for me during these 14 years of taking care of my grandchildren. It was only when her husband joined her three years after her migration that they started sending money for our upkeep. Even this was sometimes not enough, and to this day I sometimes still prefer to skip breakfast and lunch just to make sure I have enough money for the children to take to school. ( Accra, January 2013) Another caregiver, Takyi, who was taking care of two of his brother-in-law’s sons, said his major challenge was delayed remittances when school reopens for the year. After trying several times to raise money to pay the fees himself, he decided to approach the headmasters of the two schools that the boys attended for help. He intimated: Sometimes, remittances especially for school fees are delayed. I understand this. My brother-in-law tells me that things are really tough in the host country. Therefore, anytime remittances are delayed I have to find money, either my own or borrow it to pay the fees until they send me the money. There are times when I do not have anyone to turn to--to borrow money--and that put me under a lot of stress. At some point I decided to go to the headmasters who are aware that the children’s parents are abroad, and plead with them to bear with me for some time and that I will pay the fees as soon as I receive the money from the parents or as soon as I am able to raise the money. The headmasters in both schools have been very cooperative in that respect. (Kumasi, November 2011) Not all the caregivers were as good as these at managing insufficient resources. There was one caregiver, Kumiwaa, who told me that she did not have anybody to borrow from, so all three University of Ghana http://ugspace.ug.edu.gh 179 children often had to stay at home for days until their parents were able to send them money to go to school, especially at the beginning of the school term. She said 'I am always indoors with the children. I don’t have friends, and most of our relatives here in Accra are not in the position to help us. So we usually wait till they send us money' (Accra May 2012). She is one of the few caregivers who did not have any source of income. She said she used to be a hairdresser, but quit when her sister left the children in her care when she went abroad. The children were very young, and she needed to be at home to take care of them. Caregivers’ inability to manage delayed and insufficient remittances, like the case of Kumiwaa, had some negative consequences for the care arrangements. In her case, the children were not the only ones affected. She told me the parents were also stressed from knowing that their children were missing out on their school work and this placed a lot pressure on them to find money for the children. 7.2.2 Managing Communication Communication within TCRAs is also an area which caregivers had to manage so as to build and maintain trust with other members in the TCRAs and to reduce tensions that arise as a result of mistrust. Communication is one of the transnational practices that hold transnational families together. And members of TCRAs thrive on it. Caregivers and parents communicate, parents communicate with their children, caregivers communicate with the children and all members in the triad communicate with each other to ensure that the TCRAs function well. Poor communication flow, especially between the adults (caregivers and parents) in the triad, affects the TCRAs as a whole. Communication within TCRAs is usually via telephone, email or other social media platforms, mostly Facebook and Whatsapp. University of Ghana http://ugspace.ug.edu.gh 180 Managing communication within TCRAs presented itself in two forms. First, there is the communication caregivers had control over: they decided what to tell and what not to tell the migrants and the children to ensure that their TCRAs worked well. The other form was the communication that caregivers had little or no control over and, for this reason, had to conduct themselves in such a manner so as to avoid tension or trouble from arising. The latter form includes direct communication between the children in their care and their parents as well as communication from people outside the TCRAs, such as family, friends or even neighbours who voluntarily supply information to the migrants. What ‘not’ to say in TCRAs Throughout the data collection phase, I came across caregivers who withheld a lot of information from those it might concern--especially the migrant parents--for the sake of protecting the relationships in the TCRAs, including both the triad relationship among caregiver, parent and child and the dyadic relationships, for instance between the parent and child. The information they withheld usually concerned insufficient remittances, rebellious children or unmet expectations. There were also instances in which caregivers withheld information about the children from the parent or vice versa, thereby managing the relationship between the two. Caregivers had various reasons for withholding information in the TCRAs, especially from the migrants. One of the reasons was to avoid overwhelming the migrants with disturbing news from home (Ghana). Caregivers did not usually tell migrants about the challenges they faced due to insufficient remittances or how they managed them. They also did not let the children know their parents were not sending enough money. But there were exceptions to this. Two caregivers in the study mentioned that at one point they had to let the children know about the problems they were experiencing because the children were acting as if they were keeping University of Ghana http://ugspace.ug.edu.gh 181 the money their parents sent for themselves. Cecilia told me that one time when her niece sent money, she asked the oldest child (18) to accompany her to pick it up. She deliberately made the girl check the money at the counter to make sure it was the right amount. Cecilia told me: ‘The girl then asked me "grandma, will this money be enough"? I smiled and didn’t answer, but I was so happy that she noticed that the money was small'. (Cecilia, Accra August 2013). Takyi also recounted a similar story about bringing one of the boys in his care to the bank to obtain the money his father had sent. Apart from these two caregivers, however, all the others in the study said that they did not tell the children about problems with remittances, and, on very rare occasions, such as when they were very hard pressed for money, did they discuss insufficient remittances with the migrants. They chose instead to tell the migrants how well their children were doing, and how they had everything under control. Dreby (2010) made a similar observation in her study on caregivers in Mexican transnational families. These caregivers chose to let the migrants believe that everything was working well with the care arrangements in order to avoid overwhelming them with worry. Another reason the caregivers did not share challenges with the migrant parents was that the migrants had told them that life was difficult in the host country. One of the caregivers mentioned that her migrant niece had called to tell her that she had lost her job, and that she worked only a few hours a week. She told me how much she sympathized with her niece, but only after she had complained about her 'entangling [herself] in a web' and being 'in hot water' because she had not received money for the children's upkeep. Takyi also said his brother-in- law, whose children were in his care, had told him that things were really difficult for him. He understood this because his brother-in-law had three children in the Netherlands and another daughter in his village in addition to the two sons in his care. His brother-in-law, who had been University of Ghana http://ugspace.ug.edu.gh 182 away for almost 16 years at the time of the interview and who was still undocumented, had told him that he was earning very little because he did not have ‘papers’. Some caregivers also mentioned that they did not tell the migrants about insufficient remittances or ask to be reimbursed because the children were theirs too. I heard phrases like 'They are my grandchildren. Why should I expect to be reimbursed when I spend on them, it just happens that sometimes things are a bit tough for me, otherwise I should be able to provide for these children and not expect to be reimbursed'. (B. Manu, Kumasi June 2012) Other caregivers told me that the migrants did not do anything when they complained about receiving too little money because they thought the children were also theirs and had therefore decided not to complain any more. Thess remarked: I complained a few times to my sister about the remittances I received for the upkeep of her four sons. I only complained because things were really tough for me financially. When I complained to my sister that the money was not enough for her boys, she told me that that is all she has, so I should make do with that. After all they are my children too. (Thess, Kumasi July 2012) Another reason caregivers decided to keep things to themselves was because sometimes the migrants accused them of being wasteful and mismanaging the remittances. One of the stay- behind wives, Naki, told me that anytime she asked her husband for more money, he accused her of not being economical enough. Yet she knew she was doing her best with the money he gave her. Another issue that is closely related to being accused of being wasteful was the fear of being seen as an opportunist who had agreed to the care arrangement because of the migrants’ money. Takyi told me that he did not want to ask the migrants (his sister and brother-in-law) for more money because he did not want them to think he was after their money. This was the response I received from some caregivers, especially those who were neither wives nor parents University of Ghana http://ugspace.ug.edu.gh 183 of the migrants. Their concern about this was especially strong because they did not keep good track of the things they spent their own money on or keep receipts to back up their claims. Some migrants were thus reluctant to accept their claims for reimbursement. One of the caregivers who faced this sort of situation mentioned that she started keeping track of her expenditures for caring for her granddaughters in a notebook, and this enabled her to claim reimbursements from her son. Maa Asafo also told me she kept track of her expenditures just to have a good idea of how much she spent overall and how much of her own resources she invested in the children. I recommended this approach to some of the caregivers, suggesting they do this just to keep track of their expenses and to know how they managed with so little in remittances, rather than for the purpose of getting a reimbursement from the migrant. Unmet expectations for material support were one of the situations caregivers chose not to tell migrants about. In Chapter 6, I presented the cases of caregivers who were unhappy about the care arrangements because their personal expectations, which were usually unknown to the migrant, were not met. These expectations included capital to boost their businesses and to help pay for their own children’s school fees or medical bills when there was the need. Most of those who complained of such unmet expectations were unhappy and some were bitter. One example is Kessewaa, who had lost her business capital as a result of relocating to a prayer camp in Accra for a year to pray for her sister’s son to be healed. These caregivers, however, had not attempted to tell the migrants how they felt about their unmet expectations because of the fear of being branded as entering the care arrangement for their own material gain, or telling them would create tension between the migrants and themselves. Moreover, because the caregivers had not mentioned their concerns to the migrants, they also did not have to act in a way that left an impression with the children or the migrant parent that they were not happy because of unmet University of Ghana http://ugspace.ug.edu.gh 184 expectations. Thus, these feelings of bitterness and dissatisfaction did not lead to tensions in the TCRAs or in the children being neglected. Managing communication to manage parent-child relationship On many occasions, some of the caregivers in my study had to withhold their thoughts or say the right things in order to promote a cordial parent-child relationship especially the situations with rebellious children. Kwesi, a caregiver in Accra who was taking care of his uncle's son, told me that the boy was engaging in 'galamsay' (illegal mining activities), and that was why he was not focusing on the apprenticeship his father had placed him in. I asked if he had informed the boy's father about it and this was his response: No I haven’t. I just told the father that he has not returned from the village. I don’t want his father to be angry with him and decide not to help him anymore. I am always advising him. I guess he will change in his own time. ( Accra, November 2011) Abla, caregiver, said this about her sister’s son in her care: My sister bought a plasma TV for us to sell and save the money for her building project. The boy sold the TV and didn’t bring me the proceeds of the sale. He couldn't even explain what he used the money for, but the money is all gone. My sister called once to check if we had sold the TV, and I had to cover up for the boy because, if I tell her what has happened, this will be the last we will hear from her. I told her we hadn’t sold it yet, and thankfully she seems to have forgotten about it. She has not asked again. This happened about two years ago. (Abla, Accra, July 2013) In managing the relationship between migrants and their children, caregivers did not just have to withhold information from them, but also had to know the right things to say. There were instances in which caregivers had to plead on behalf of the children so that the parents would not lose faith in them. Abla, the caregiver whose experience with the TV, I recounted above, told me how she had to plead with the migrant to pay for the child to take a course he was University of Ghana http://ugspace.ug.edu.gh 185 interested in rather than the one his mother had proposed. The boy’s mother complained about having wasted money on the previous course he took and, for this reason, was not ready to waste any more money on his education. There were also situations involving the communication between the parents and the children in their care. For example, because of the distance and length of separation involved in most of the TCRAs, sometimes the children got the idea that their parents had abandoned them. In these situations, the caregiver had to talk with them to keep them from developing an aversion to their parents. Gyeiwaa described one such instance: My youngest grandchild, who has been with me since she was four-months-old, told me that she is not sure that she is the daughter of her mother (migrant). That was not the first time she was saying that, so this time I decided to ask her why. When I asked why, she explained that she has never seen her, and her mother has not come to Ghana to see her, not even once. Moreover she does not call her often. She added that if her mother does not want her, she will adopt her Sunday school teacher as her mother. I had to convince her that her mother loved her, and that she has not come to see her because she does not have papers, and so as soon she does she would come and see her. I quickly called my daughter and told her about the incident to prompt her that she might be losing her daughter if she does not start calling often to assure her of her love. (January 2013) Awurakua, whose sister’s child had been with her since she was two years old, and had only seen her mother for the first time after a twelve-year absence, said: The child is very close to me. For a long time she thought I was her mother. She only got to know that her mother was abroad when my sister started calling to tell her that she was her mother. The child was about seven-years-old then. I know she has a strong affinity for me and her mother has noticed it. I am not sure how she reacts to it, but I try as much as possible to get the child to be close to her [mother]. For instance, the first time she came to Ghana, she was living at Kasoa and requested the child join them. The child wanted to make a return trip, but I convinced her to spend a few days with her University of Ghana http://ugspace.ug.edu.gh 186 mother and her sisters. Two days later my sister called and asked me to speak to the child because she had refused to eat and was not cheerful. So I had to talk to her and ask her to enjoy the time with them and assured her that she would return to me soon. She said she wanted to come home (to me). I felt very bad for my sister. I know she might be thinking that she has lost her daughter to me. But I have been talking to the girl and also encouraging my sister to keep drawing close to her. (Awurakua, Accra, May 2012) The caregivers’ role in maintaining a cordial parent-child relationship in TCRAs is an important one, and this has been documented in other studies on TCRAs (see Dreby, 2010, Moran-Taylor, 2008; Menjivar and Agadjanian, 2007). Dreby calls the caregivers in her study ‘middle’ women largely because of this important role they play between the migrant parents and their children. Managing communication with members in and outside the TCRAs The caregivers’ attempts to control information flow within the TCRA by concealing some information or divulging only the ‘right’ information has not always been easy. As noted by Madiannou and Miller (2012), it is even more difficult with the proliferation of new communication technologies. This makes it difficult to control who communicates with whom among the members of the TCRAs. There is direct communication between migrant parents and children who have been left behind due to the wide accessibility and affordability of mobile telephony in Ghana. Caregivers are unable to control information flow between the parents and their children, and this is another situation they have to manage in their day-to-day care for the children. Most of the caregivers in the study cared for children who owned cell phones bought by their parents. This made it possible for the children to communicate directly with their parents out of sight of the caregiver's watchful eyes. In this way, the children sometimes served as their parents’ main source of information in the TCRAs. Similarly, Dreby (2010) observed University of Ghana http://ugspace.ug.edu.gh 187 that parents relied on their children for information about how the remittances sent to the caregivers were used. Some caregivers found this very frustrating. What was more disturbing to them was that the children never reported the entire situation surrounding an incident to their parents. Instead, they only presented the part they found convenient to report. To manage this situation, the caregivers adopted what can be described as the practice of ‘careful parenting’, where they claim to be very careful around the children of migrant. For instance, they allowed the children in their care to get away with conduct that their own children would not have. Different methods of disciplining a child which involved administering physical punishments or withholding privileges are all part of character development in children (Gyekye, 1996; Nukunya, 2003; Twum-Danso, 2010). Parents would spank their own children, insult them or deny them food in order to correct unacceptable behaviour and develop good character traits in them. Many caregivers mentioned that this was how they were brought up by both of their biological and foster parents. However, if they extended the same methods of child upbringing to the migrants’ children, it would be taken in a negative light by the parents, usually because the children told them about it. Cecilia recounted how one of the five children in her care actually called her mother abroad to tell her that she was starving her. According to Cecilia, the girl in question had refused to eat the food she left for her on two occasions, which she had to throw in the bin because it went bad, and this was the reason she decided not to reserve any food for her if she was not at home for supper. I also witnessed an incident when visiting Cecilia’s house. Cecilia had asked one of the children to wash some clothes. After washing them, the child went to sleep without hanging them out to dry. When I was interviewing the oldest girl, she turned to look at Cecilia University of Ghana http://ugspace.ug.edu.gh 188 drying the clothes and asked if her sister had not done it. Cecilia nodded and added that she had gone to sleep. Later Cecilia told me that her own child would not have gotten away with this. She continued by adding but with these children, when you touch them, they will call their mother and tell them a completely different story, making you appear like a bad person before their mother and, these migrants, when anything like that happens, they take their children’s side. (Accra, November 2011) Koo reported a similar incident where the oldest child under his care (15) had called his mother to tell her that Koo was calling them animals. Koo said their mother called him to gently inquire why he had called them animals. He then had to explain to her that he was quite overwhelmed by their behaviour and so he asked them if he was living with animals or human beings. What surprised Koo was the fact that the boy who made the call did not explain the whole incident to his mother, making it appear as though he was deliberately abusing them. Gyeiwaa also explained why she was careful about spanking her grandchildren. There was a time that my son beat the third child (the son of his sister), and he reported this to his mother. She was so upset and talked so angrily, asking whether her brother wanted to kill her son and accused me of watching her brother trying to ‘kill’ her son. I have therefore decided that I will not use the cane on them. First of all, it is not my style and, secondly, I do not want to have issues with my daughter. ( Accra, January 2013) Caregivers were not happy raising the children that way i.e. being very careful with them and sometimes leaving them to their own devices. While this conflicted with their own ideals of proper child upbringing, they resorted to this in order to avoid tensions in the relationships in the TCRAs. University of Ghana http://ugspace.ug.edu.gh 189 Closely related to the problem posed by direct communication between the children and their parents is the communication between people outside of the TCRAs and the migrants. Caregivers have no control over communication between neighbours, friends, members of the extended family in the home country and the migrant parents. One example of this problem involved a stay-behind wife whose neighbours called her migrant husband to tell him she was having an affair with one of her benefactors. This led the benefactor to stop assisting them. However, fortunately for the wife, the migrant husband did not believe the neighbours' story and no conflict developed between them as a result. Another stay-behind wife, Akua, blamed her husband’s lack of communication with her for more than two years on what she suspected was gossip originating from her husband’s family and her neighbours. Akua lives in the same village as her husband’s relative. Mckenzie and Menjivar (2010) noted this problem to be the greatest challenge for stay-behind wives, which also led these wives to always be careful about what they did because the neighbours or members of their husband’s families might be monitoring their activities. The story of Florence, a caregiver who was taking care of her younger sister also shows the difficulty caregivers have controlling communication outside the TCRAs. She told me about an incident in which communication by outsiders had benefitted the TCRA, but only at the expense of exposing a previous caregiver. She said that her younger sister who was now in her care was initially left in the care of their maternal aunt at the time her mother was travelling. The neighbours of their maternal aunt called her parents in Amsterdam to inform them about the mistreatment and physical abuse that her sister was going through at the hand of their aunt. They further warned her parents that if they did not take the child away from the caregiver, she would die from the physical abuse. This information led the parents to ask Florence to take her sister University of Ghana http://ugspace.ug.edu.gh 190 away from the caregiver and into her care. Neighbours and outsiders do not always report half- truths and false information. But the fact that the caregivers were aware they were being watched and their actions or inaction could be reported to the migrant by anybody at any time put them in an undesirable position, which led to them treading cautiously around the children of the migrants. 7.2.3 Managing the Assumptions of Outsiders The caregivers were also faced with the challenge of managing the assumptions that people had about them because they were taking care of the children of migrants. These assumptions were held mostly by people outside of the TRCAs such as neighbours, friends and members of the migrants’ family in the home country. Because of how Ghanaians perceived emigrating abroad, especially to the West, caregivers sometimes had to deal with persons who assumed that they had accepted the care arrangement because of the remittances they would receive from the migrant parents. Additionally, people assumed that they were enriching themselves with the remittances the migrant parents sent for their children or that the migrant parents were showering them with gifts and money because they were taking care of their children. The caregivers who were faced with this challenge told me that these assumptions were sometimes expressed in the form of gossip, and, in other cases, communicated directly to them. Some of the caregivers in my study told me how they believed people envied them because they were taking care of a migrant’s child. Caregivers told me about instances in which people asked them to lend them money and were very surprised to learn that the caregivers did not have the money to lend them. There were a few particularly bold ones who told the University of Ghana http://ugspace.ug.edu.gh 191 caregivers directly that after all the dollars and pounds they received from abroad, how could they claim to not have money. Some caregivers also mentioned how people considered any nice thing they had or wore as something they had received from abroad. Takyi told me that he was spotted taking pictures of a building in his village with a huge camera. A few days later, he heard from one of his friends that the family members of his brother in-law (migrant) had been speculating about how he must have gotten the camera from their brother and that he was getting lots of gifts and money from him as well. Takyi intimated that he was really upset because they had no idea what he had to go through when the migrant was not able to send money for the upkeep of his two sons in his care. Caregivers did not only have to deal with the envy of outsiders, but also the assumptions people held about them made it difficult for them to ask for financial assistance from their friends, neighbours and sometimes their own family members when they faced financial difficulties. They usually hesitate to ask for fear that they would be ridiculed and/or seen as spending the remittances they receive on themselves. They also could not ask for assistance from outsiders because they were expected to have more resources from the remittances they received. Cecilia mentioned that her niece had not been able to send her money for some time, and every effort to get the father of the children who lived in Accra to send them money had proven futile. He kept giving her the excuse that he did not have money. But she did not believe his excuses. She said 'he thinks that my niece is sending me bags of money, and that is why he is behaving that way. If the children were living with him, would he have watched them starve to death? (Accra January 2013). The caregivers who found themselves in this situation had to manage these assumptions and not allow the comments that people made to affect the way they took care of the children. Caregivers were, however, not pleased with the assumption that they University of Ghana http://ugspace.ug.edu.gh 192 were enriching themselves with the remittances of the migrants, when, in fact, what was going on in most cases was that they were having to spend their own money when the remittances were delayed or insufficient. Some of the caregivers said that they kept the financial challenges to themselves or, if they had to borrow, they did so from people they trusted. This, however, limited the number of people from whom they could seek help. Not asking for financial assistance from outsiders was a way of avoiding strengthening the assumptions that people held about them. And, it was a way to protect the image of the migrants by not disclosing that they were facing challenges in the host country. Living up to Society’s Expectation of a Good Caregiver The caregivers also had to live up to society’s expectations of a good caregiver. Caregivers sought to manage all these different kinds of situations in part to preserve the relationships in the TCRAs to enable them to function properly and in part to live up to society’s expectations. In Ghanaian society, foster parents and, for purposes of this study, caregivers are expected to raise the children under their charge as they would their own according to child fostering norms. There are many adages that support and encourage caregivers to take very good care of the children in their care. One example of this is the Twi adage 'wo ye obi die yie a na woso die be ye yie' (if you treat what belongs to someone well, then your own will also turn out well). In addition, because of the general belief that a child is born to everybody, caregivers were expected to raise them without, for instance, demanding money, even when the parents had the financial resources. Most of the caregivers mentioned above did not want to ask the migrant parents for more remittances or for reimbursement for money they spent on the children because they did not want to be seen as being in the care arrangements for money. They were afraid of University of Ghana http://ugspace.ug.edu.gh 193 being labelled as greedy and wanting to collect money to raise another person’s child. However, against this ideal position which society holds when it comes to child care is the reality that caregivers sometimes find it difficult to look after the children with their meagre resources. The caregivers, however, had to manage the expectations of society; even the wives in the TCRAs had to meet society’s own expectations of who a good wife of a migrant is. Caregivers often strove at the expense of their own comfort to live up to society’s expectations of them. They were criticized, sometimes quite harshly, when they failed to meet these expectations. Ophelia, whose charge was sick and at the point of death, mentioned that in the midst of all the confusion, the only thing that kept running through her mind was what the family members and other people might think of her if the child died. What made this tougher for the caregivers was the strain they felt on their role due to the attributes of transnational migration, especially the communicative environment it entailed and the role of migrant remittances in the care arrangement. Closely related to the problem of being criticized by society is what the caregivers thought the migrant parents might think about them should anything go wrong with their children. Some caregivers gave reasons why they believed the migrant parents would not consider their personal sacrifices in the event that something happened to their children. Stella, a grandmother taking care of her migrant daughter’s child told me: We (Stella and her husband) sent her pictures of the child while she was away, and she told us how nice the child was looking. However, a few days before she returned, the child fell ill and she lost some weight as a result. The child was picked up by her father to meet her mother at the airport. The arrangement I had with her over the phone while she was still abroad was that even when she returned, the child would stay with us during the weekdays to attend school and spend the weekends with them. We were therefore expecting the child to be brought to us a few days before the reopening of the University of Ghana http://ugspace.ug.edu.gh 194 school term, but her parents did not bring her over. They (my daughter and her husband) later came to tell us that they would like the child to be with them longer. I only got to hear from my younger daughter that the child’s mother was not happy with the state in which she found the child. Since she is my daughter, I called and gave her a piece of my mind. She apologized and here we are; she has brought the child to me again. (Stella, Kumasi May 2012) Cecilia also mentioned that she knew that should anything happen to the children in her care, her niece would forget about all of her efforts and blame her. She added, 'I always pray for the children that nothing happens to them so that their mother can come for them in peace. I will be so elated when they [the children] finally leave'. (The migrant is working on family reunification in the host country.) 7.2.4 Managing a Change in Lifestyle Caregivers, especially those who care for very young children, have sometimes had to change their entire schedules and lifestyles to suit the care arrangement because of the amount of attention children at this age require. One grandmother said that she had to cut down on attending social functions, especially funerals and women’s fellowship meetings, unless it was one that she must attend. Attending funerals and other social functions are very important in Ghana, especially for the elderly, because, apart from the company it provided, it was a way of ensuring that they would receive the needed support when they were bereaved or in any kind of trouble. This grandmother recounted how she had to line up all of the eight children belonging to her three migrant daughters on Sundays to go to church. She also had to wake up at 4:30 a.m. every day to get them ready for school. Her schedule became a bit more relaxed when the first few children got to the age where they could assist her in the home. University of Ghana http://ugspace.ug.edu.gh 195 Ophelia also had to take all three of the boys in her care along with her if she had to attend a function or arrange for somebody (known to the migrant) to stay with the children for the few hours that she had to be away. This did not occur too often. Even though no one mentioned this to her, Ophelia knew she should not make such outings a habit. Thus, her life revolved around the children. There was no vacation period for her, not even when the children’s parents returned from abroad. She had to cut down on her social activities including weddings, funerals and church activities. Yet, the importance of these activities for a single woman of her age cannot be over-emphasized. There was another caregiver, Kumiwaa, who had to be separated from her husband. Her husband was not willing to live with her and the children in the house the migrant had rented for them. In addition, there was not enough room for him. Kumiwaa got married a few years after she had assumed responsibility for the children, but she and her husband have been living apart, Kissiwaa in Accra and her husband in Kumasi, because she had to take care of the caregivers' children. This meant that she had to move between Kumasi and Accra to keep her marriage intact. At the time of the research, the children, who were 15, 17, 19, were old enough to take care of themselves, so she could leave and spend longer periods in Kumasi, much to the chagrin of the migrant parents, who thought she was leaving the children by themselves for far too long. 7.2.5. Managing the Unforeseen in TCRAs Migration is an uncertain and at times risky investment, and immigration policies often add considerable insecurity and powerlessness (Akesson et al., 2012). There were unforeseen situations that befell the caregivers or migrants, but they still had to carry on as if nothing had happened or deal with them as best as they could because there was very little each could do University of Ghana http://ugspace.ug.edu.gh 196 about it. In the TCRAs I studied, some caregivers had to deal with the migrants' imprisonment in the host country and the consequent hardships which followed from this. In other instances, the migrants lost their jobs or the caregivers suddenly fell ill and took very long to recover. Two out of the three stay-behind wives had to manage the unforeseen imprisonment of their husbands. On different occasions, the migrants were arrested and jailed for living illegally in the host country. In the case of Akua, the migrant husband was arrested a year prior to my research. His incarceration lasted ten months. During that time, Akua had to take care of all their five children on her own. Thankfully, the last three were quite young and not aware of what was going on. It was the second child who kept asking questions about his father because he used to call to speak with the children regularly before his arrest, so the child could sense there was something wrong. Yet, Akua could not let him know the truth. Their oldest daughter was the only one who could speak with him because he only spoke English when he called to avoid betraying his nationality. It was a very difficult period for the mother-caregiver who had little to no assistance. This was due partly to her not wanting anybody to know about her husband's imprisonment. 'My daughter and I kept it to ourselves; not even my landlady of over 20 years knew about it'. The caregivers whose husbands were incarcerated did not only have to manage the financial difficulties they almost always faced, they also had to manage the situation so that their neighbours did not disgrace them. As a result, in most cases, no one should know about the situation, as the example of Akua illustrates. Pomaa, the other stay-behind wife who faced a similar situation, said the biggest challenge she was confronted with was keeping the information from her youngest daughter (16) who was then in boarding school. Pomaa believed the child would have been devastated: 'So for eight months I had to be telling one lie after another to University of Ghana http://ugspace.ug.edu.gh 197 explain to her why her father had not been calling. I am so glad it’s all over now, and she still doesn’t know about it'. 7.3 Summary In this chapter, I have discussed the practice of care in TCRAs. The few existing studies on transnational parenting that have focused on caregivers have hardly looked at what their tasks as caregivers of children of transnational migrants entail. What I have done in this chapter is to unveil what goes into the day-to-day care activities of the caregivers in my study. In performing their activities, which I divided into two main categories, namely 'providing care' and 'managing care', I have shown that caregivers dedicated their time and energy not just to the children in their care, but stretched these efforts to protect the migrants from anxiety and to keep the TCRAs functioning properly as a whole. Managing care together with practicing care constituted all that went into the tasks of the caregivers and was also what earned them the attribute of ‘w’aye fine’, which deemed caregivers to have performed their role well. I have also argued in this chapter that the success of providing care is largely dependent on how care is managed. I also show in this chapter how management of care in TCRAs is made difficult because of the cultural ideals of child fosterage and demand it places on caregivers to live up to societal expectations. University of Ghana http://ugspace.ug.edu.gh 198 CHAPTER EIGHT: CAREGIVERS’ NETWORKS OF CARE IN TCRAS 8.0 Introduction In many African families, large households which include extended kin, child fosterage, the practice of polygyny and the existence of co-wives, make it possible for child care to be shared (Goody, 1982; Isuigo-Abanihe, 1984; 1985; Bledsoe, 1990; Clark, 1999; Verhoef, 2005; Notermans, 2008). In the case of transnational families, migrants who have children in the home country also depend mainly on kin to help with the care of their children. Schmalzbauer (2004) describes caregiving in this context as a cross-border division of labour in which production occurs in the host country and reproduction in the home country. These primary caregivers in TCRAs are also known to actively engage both local and international actors to carry out their responsibilities to the migrant and the children they leave behind. My data shows that these actors were mainly people in the migrant’s network. Bernardi (2011) established that people in the network of migrants, both at home and abroad, are providers of affective environment, practical support and identity. It is, therefore, a useful exercise to consider primary caregivers and the actors they work with in studying TCRAs. Focusing on the networks of migrants and the caregivers of their children stresses the notion that individuals do not act in isolation, but are embedded in a network of social relations (Granovetter, 1985). Additionally, from the literature on African families, it is realised that there is a large support network for child care yet how this works out in a transnational context is not known. The aim of this chapter is to establish two main points: First, that primary caregivers usually receive support in providing care for the children of the migrants. And second, that, contrary to what has been suggested by the literature on support for childcare in Africa, and in Ghana in particular, the care networks of the caregivers University of Ghana http://ugspace.ug.edu.gh 199 are small in size (the number of persons that a caregiver had and used in taking care of the children). The first section looks at the characteristics of this network. In the second section, I discuss small networks where I focus mainly on what I mean by small networks, factors which could account for the small size of caregivers' care networks and the consequences that small networks have on some of the caregivers and their TCRAs. 8.1 Characteristics of the Care Networks 8.1.1 Size and Composition There was an average of 3.2 people in each caregiver’s network. Two caregivers had the highest number of persons in their network with six people, and three had the fewest number with 1 person each. Of the 48 members in the caregivers’ network, 44, or 91.6 per cent, were the caregiver's kin as well as the migrant because all 15 caregivers were kin of the migrants and the children. The remaining 4 (8.4per cent) were non-kin. These comprised 3 housemaids and a friend of a migrant parent. There were 25 females (53.1 per cent) in the caregiver networks, including 8 migrant mothers and 23 males (47.9 per cent) of which 10 were migrant fathers. Eighteen out of the 48 members lived in Amsterdam, and 1 lived in The Hague, the two main research sites in the host nation. There were some TRCAs in which both parents lived in the host country. This is the reason I have 19 parents in the Netherlands rather than 15 in consonance with the number of TCRAs. Of the remaining members, Seven lived in Accra, most of which belonged to the network of caregivers interviewed there. Eighteen lived in Kumasi, and again most of these belonged to the network of caregivers who lived and were interviewed in Kumasi. One of these, however, lived in the UK and another in another part of Ghana. Tables 8.1 and 8.2 sum up the size and composition of caregiver networks University of Ghana http://ugspace.ug.edu.gh 200 Table 8.1 Caregiver’s Care Network: Size and Composition CG Location Size Sex Male Female Relation Kin Non-Kin* Naki Accra 3 2 1 3 Pomaa Kumasi 4 2 2 3 1 Kwesi Accra 1 1 1 Takyi Kumasi 4 2 2 4 Akua Kumasi 1 1 1 Awurakua Accra 3 2 1 3 Jennifer Kumasi 5 2 3 5 Victor Accra 1 1 1 Cecilia Accra 3 1 2 3 Goergina Kumasi 2 1 1 2 Owonta Kumasi 4 1 3 3 1 Kissiwaa Accra 3 1 2 3 Mama K Kumasi 4 3 1 4 Christie Kumasi 3 1 2 1 2 Fosua Kumsai 6 2 5 4 2 *Non-Kin members presented included 3 housemaids and a friend of a migrant parent Table 8.2 Caregiver’s Network: Spatial Dispersion CG Location Size HH* SC** Outside CG’s City*** Naki Accra 3 1 1 1 Pomaa Kumasi 4 3 1 Kwesi Accra 1 1 Takyi Kumasi 4 3 1 Akua Kumasi 1 1 Awurakua Accra 3 1 2 Jenifer Kumasi 5 1 2 2 Victor Accra 1 1 Cecilia Accra 3 2 1 Georgina Kumasi 2 1 1 Owonta Kumasi 4 1 2 Kissiwa Accra 3 1 2 Mama K. Kumasi 4 3 1 Christie Kumasi 3 2 1 Fosua Kumsai 6 4 2 *HH – Same household as CG and child/ren of migrant. **SC – Same City as the CG and the child/ren of migrant *** -- All but two in this column lived in the Netherlands. Of these two, one lived in the UK and the other in the Eastern Region of Ghana. University of Ghana http://ugspace.ug.edu.gh 201 8.1.2 Caregivers’ Network Members and Support Functions – repetitive Network members offered a broad range of support to caregivers. In my study, the members that were mentioned played different roles with respect to the care of the migrants’ children. In most cases they ended up playing many roles. Their roles depended primarily on their availability, access to resources and their expertise in a particular area (e,g education). I gathered from the caregivers that there were members from whom they requested for assistance and there were also others, especially parents, who offered assistance out of a sense of obligation to their children. And yet another set of members offered to help voluntarily because of their past relationships with the migrant. In order to catalogue the support functions provided for the caregivers, different sets of questions were asked from the name generator questionnaire and further elaborated on during in-depth interviews. Table 8.3 Network of Caregivers by Support Functions Support function(s) Network partners Remark Who pays? (education, medical bills, feeding and clothing, etc.) Migrants parents, caregiver’s siblings of caregiver, friend of migrant parent Siblings and friends are called upon when caregiver is in dire straits. Proximate care (daily care). Who does the caregiver trusts most to leave child with when absent? House helps, caregivers’ adult children, caregiver’s mother, caregiver’s husband Live in same household as caregiver and child. Who educates child to behave (discipline and advice)? Migrant parents, caregiver’s husband, caregiver’s mother, caregiver’s brother, caregiver’s adult children With the exception of migrant parents, all the others live in the same house as caregiver. Who checks on child’s progress at school or an apprenticeship? Caregiver’s brothers, caregiver’s adult child, caregiver’s husband, some migrant parents With the exception of the caregiver’s husband, others do not live in the same household as the caregiver and the child. This role does not require this. Who decides on matters Migrant parents Major decisions are made University of Ghana http://ugspace.ug.edu.gh 202 regarding the child’s education, health together with the migrant parents. However, when it comes to communication with the parent, migrant parents make the decision. Who do caregivers have arguments with about the child's upbringing? Migrant parents Arguments are usually about differences in how the child should be treated and raised. Who do caregivers seek advice from about the care of the child? Migrant parents, caregiver’s adult children Caregivers usually sought advice when the child/ren became rebellious or troublesome. 8.1.3 Incidental Helpers in Caregivers’ Network In addition to the active members of a caregivers’ network (Table 8.3), there were some people that were missed during the mapping exercise. I referred to these as incidental helpers in the caregivers’ network. These helpers provided support at some point in the care arrangement and stopped for a variety of reasons such as relocation of the helper of the caregiver and the children. Since the name generator exercise focused more on people currently providing help, incidental helpers were not mentioned. However, caregivers’ narration during in-depth interviews indicated that the support of these people helped a great deal at the time it was offered. The support provided by incidental helpers could also be social, emotional and mainly material. Caregivers who had such support described them as ‘timely interventions’ the following example by Naki, a stay behind wife illustrates the role of one of such incidental helpers: I remember a particular day when my boys and I had to eat gari7 for supper. I had no money on me, and all I had was gari and palm oil (red oil). I soaked the gari until it was 7 A starchy food made from a tuber called cassava. It is advised to eat it with protein because it contains mostly carbohydrates. University of Ghana http://ugspace.ug.edu.gh 203 very soft and then I mixed it with the palm oil and we ate together. I didn’t know where our next meal for the next day was going to come from. Then something miraculous happened the next day. I received a visitor in the morning. He happened to be the driver of the then National Chairman of one of the large churches in Ghana. The driver said that the chairman had sent for us – the children and I. You can imagine the shock on my face. I had no idea why he sent for us, but I knew we might be fed there at least. When we got to the house, he and his wife welcomed us warmly, and there was a buffet table set for us. There was so much food that I could not eat. I was both happy and sad at the same time. After eating, he asked the children to go out and play, and he told me that they had met my husband in Holland a week earlier and he told them about us. They even brought pictures they took with him for us (these are the latest pictures we have from him). The man told me that my husband was very kind to him when he came to Ghana after the war in Liberia. He said my husband was the first person to meet him at the gate of churches HeadQuarters. He gave him a big hug and told him how happy the entire church was that he was safe in Ghana. He also said that he believed that God spared his life for a purpose, and that God will make him a big person in future. His words were so kind, and that was how he and my husband became friends. He was very nice to him all the time. He added that if he had known about my husband’s intention to travel, he would never have encouraged him to go. He and his wife promised him when they met in Holland to look out for us when they got back and that was why they invited us over. We had a nice time and late in the day, the driver was asked to drop us off at home. We were sent back with a car full of food stuffs and money. He continued to send us food and some money from time to time until his tenure as chairman was over and he relocated. That source was cut from then. (Accra, November 2011) Another example involved an incidental helper that came into the life of a caregiver in the course of the research. Shortly after my first meeting with Pomaa in 2011, her husband (migrant) was arrested in the host country for being an illegal migrant. During my visits at the time of the migrant’s incarceration, the caregiver mentioned one person who was her husband’s friend in the host country. She had not known about him before her husband’s arrest. He was the one who had called her to break the news about the arrest. He was very supportive during the entire period of her husband's incarceration. He sent her money to help pay for the last child’s school fees and called regularly to check on them. He gave her information from her husband and vice versa. On the two occasions when I visited during her husband’s incarceration, she told me how good this man had been to them. Her husband was released eight months later, and I visited again to do a post-prison interview with her. I asked about this friend whom she claimed University of Ghana http://ugspace.ug.edu.gh 204 had been very supportive, and she said the last time they heard from him was a few days after her husband’s release. Her husband had told her that the man had come to Ghana, and therefore she was expecting him to call them so they could keep in touch while he was in Ghana, but they had not heard from him yet. I cited the above examples to illustrate that caregivers had what I would call regular or active members in their networks as well as incidental helpers 8.2.4 Caregivers/Migrants Network Another interesting observation I made was that almost all the people who were named in the caregivers’ network were drawn from the migrants’ network. They were people who were directly linked to the migrant but not to the caregiver per se. The majority of these members were kin and since all the caregivers happen also to be kin, the caregivers were related to them as well. However, these people performed their stated roles more out of obligation to the migrant as kin or a friend rather than because of the caregiver. They also sometimes offered their support as a way of reciprocating past or ongoing acts of kindness or as an ongoing reciprocal relationship with the migrant. There were two caregivers who had members in their personal network who provided support because of their ties with the caregiver and not the migrants. These were siblings of two of the wives in my sample. They usually called on them for support when they found themselves in severe financial straits. 8.2 Small Networks A combination of the active network members which gave an average of 3.2 people per caregiver, and the incidental helpers notwithstanding, demonstrates that the networks of the University of Ghana http://ugspace.ug.edu.gh 205 caregivers are small in size as compared to what the literature and common knowledge assert. Studies on transnational parenting have hardly paid attention to the network of people that help care for the children that migrants leave behind. Consequently, there is very scant (if any) information on the size of these networks. Unfortunately, as much as this claim suggests that a large number of people are involved, few of them provide a sense of how large the numbers are. Thus, I am not able to juxtapose the size of the networks in my study against those in other studies. I hope that subsequent studies on the networks available to children of migrants or child care in general will begin to pay more attention to the actual sizes (number of people) of one's care networks. My argument that the network I discovered was small is thus based on the use of terms like ‘large numbers’, ‘wide network’ and ‘extended kin’ that punctuate the literature which are usually mentioned in comparison to the Western family model consisting of a couple and their biological children. In the subsequent sections, I discuss some of the factors I have drawn from my data which might account for the small size of these networks. 8.2.1 Mobility in Urban Centres Awurakua – close yet so faraway Awurakua takes care of her migrant sister’s daughter. She hails from Akyem-Oda in Ghana's Eastern region. (Her family belongs to the Akan speaking group). She lives in Dansoman, a surburb of Accra with her husband and son, three children of her husband and the child of her migrant sister. Awurakua works as a trader at one of the market complexes in Accra. She leaves home before 6 a.m. to avoid traffic as well as to catch the ‘early birds’ at the market. Early mornings and late afternoons are the best times for her business owing to the kinds of wares she sells. She stays at the market all day and returns around 7 p.m. Awurakua has two brothers, an older brother who comes after their migrant sister and a younger brother who is the last child. Both brothers live and work in Accra. The older brother, who was named as a member of Awurakua’s network, lives in another surburb of Accra and has to endure thick traffic jams to get to Awurakua’s house. He is a self-employed auto-mechanic with a workshop located in a part of Accra noted for unpredictable and high levels of traffic any time of the day, making journeys University of Ghana http://ugspace.ug.edu.gh 206 which in moving traffic can be made in 20 minutes take no less than two hours. His house and workshop are located at two extreme ends of Accra, both of which are quite far from where Awurakua lives. The second brother lives in a town bordering the Greater Accra region and the Central region. It is a very densely populated area with most of its inhabitants travelling daily to work in Accra's centre. Until recently (barely a year earlier when I was conducting my research), this road was clogged with thick traffic jams throughout the day, because of long-term construction. He is a taxi cab driver and, therefore, his job takes him round various parts of Accra during the day meaning he must grapple with his share of traffic jams. The location and work schedules of Awurakua and her brothers do not allow for much face-to-face interaction among them. The best time for them to see each other is during the weekend, especially Sundays as all three of them work Monday to Saturday. Saturdays are also very busy days for residents of big cities such as Accra. Saturday is a day that is usually reserved for social activities – a very important part of the lives of urban dwellers. All of this it makes it almost impossible for Awurakua to receive any support (proximate care and socialization) from her relatives. I used the vignette above to illustrate how urbanization and the increasing dispersion of the members of a family contribute to reduced kin support for child-care. I conducted my study in the two largest urban centres in Ghana. The literature on the changing structure of the family has posited that migration, both internal and international, weakens extended family bonds and reduces the interdependency that family members shared. Nukunya argues that when lineages were more localized, cooperation was much easier as was the regular expression of cooperate unity (Nukunya, 2003). Urbanization is also said to be associated with the movement towards monogamy and the strengthening of conjugal bonds above all else (Zeitling and Megawangi, 1995; Badasu 2004). It appears that when people migrate outside of their region, they do so as individuals, not in droves of extended family members, and they begin to live in self-contained individual houses instead of in compound or family houses. Some caregivers agreed to the increasing nucleation of the family in the urban centres. One of them remarked: 'we are trading the nice compound house system in favour of self-contained houses where each for himself, God for us all'. (Maria, Accra, Janaury, 2012). Furthermore, life in urban centres poses great challenges to providing each other with the needed support even for family members who live University of Ghana http://ugspace.ug.edu.gh 207 within the same urban area. The effect of internal migration into the cities is having its toll on the infrastructure, which has not grown proportionately to the increase in population (Spio, 2011). This includes roads and transport services (ibid). Distances in the cities are, therefore, expanding with the increasing population and pressure on transport services. As illustrated by the story above, movement within these cities (Accra and Kumasi) is quite hectic. The cost of movement and the long hours of sitting in traffic jams make it very difficult and costly to move from one place to another. What is more, the work schedules of individuals and the distance between home and the work place only exacerbated the issue as was pointed out in Awurakua’s situation. It, therefore, appears that in order to receive assistance with proximate care in an urban environment, the best situation is for the members in the care network to be living in the same compound as the caregiver and the migrant's child or in the same neighbourhood. There caregivers such as Maa Fosua who had this kind of living arrangement. She lived in Kumasi with her husband, her adult daughter and two house-helpers. All these people can be viewed as offering some support to Fosua in caring for her granddaughters. These care networks stand in contrast to those in rural or semi-urban centres where distances must be covered on foot because members in the same family live close to each other where people were not saddled with issues of vehicular transportation. 8.2.2. Economic Status of Members in Caregivers’ Network Besides migration and urbanization, another factor that has been strongly associated with the increasing nucleation of families, is the economic hardship that a large majority of Ghanaians University of Ghana http://ugspace.ug.edu.gh 208 face. This means that many a Ghanaian can barely manage to stretch their resources beyond their nuclear families. The situation is more pronounced among the low-educated and low- skilled population who live in Ghana’s urban centres. A majority of my respondents and the members in their network fall within this category of Ghanaians. The dire economic situation was even worse in urban areas where paying for basic utilities, educating one’s children and fulfilling other obligations needing to be met made life quite unbearable for those in the lower income bracket. As was discussed in chapter 7, one major challenge caregivers are constantly faced with is the inadequacy and irregular flow of remittances which means that most caregivers would need some financial assistance from members in their network. However, this assistance was not forthcoming. Apart from the migrant parents in the network of caregivers and a few instances where caregivers relied on a friend or a sibling to support them financially, most of the people named in their networks performed non-financial functions. Some caregivers mentioned that they knew that they could not count on the members of their network for financial assistance, so they hardly ever asked them for such help, as Cecilia, a caregiver who mentioned some members of her immediate family in her network explained to me. She lived in the same household as the persons she named in her network, but most of the time she could not rely on them for financial assistance because she knew their financial situation that they could not be of much to her. Hence, caregivers had the least number of people in their network that they could rely on for financial assistance for the upkeep of the children which happened to be the assistance they needed most. University of Ghana http://ugspace.ug.edu.gh 209 8.2.3 Caregivers’ and Migrants’ Capacity to Provide Care There were three caregivers who mentioned only one person in their network, and this was the migrant parent of the child in their care. They had two things in common, both caregivers and the migrants in those TCRAs had the financial resources to take care of the child/ren and second, usually the child/ren in their care were older and required minimal attention of the caregivers. Consequently caregivers in these TCRAs do not require a many people in their network of support to provide care. Even though the presence of these two factors - caregivers’ and migrants’ access to material resources and the age of the children – accounted for why caregivers provided adequate care with a small network - they are mutually exclusive. For instance, there were caregivers who had younger children but the financial capacity of the migrants and caregivers could make the TCRAs function properly without the need for many people. I present the cases of Maa Fosua and Awurakua to illustrate this point. Maa Fosua was taking care of her three grandchildren. She owned a shop at the Central Market in Kumasi where she sold different kinds of wax prints. She said business was good and therefore she lacked nothing. Additionally her migrant son sent remittances to cover the living expenses of the children and even paid for the services of a house-help. She admitted that there were times that she had to spend some of her income on her grandchildren but she got reimbursed as soon as she made the claim. Awurakua was also a trader at the market and she was taking care of her sister’s daughter. She mentioned that her sister and her husband always sent money at the beginning of every school term which means she received remittances once every quarter. The amount was to cover the cost of school fees and school supplies for the child and her son. They also added an extra amount to support with other living expenses. She said that extra amount could barely last one month; hence for the remaining months until the school University of Ghana http://ugspace.ug.edu.gh 210 term ended, she had to take care of every other expense concerning the child. She did not complain about facing financial difficulty because in her own words By God’s grace I have a regular source of income so I can afford to spend on her without asking the parents. However, on a few occasions when I am broke I ask them to send money for one purpose or their other and they immediately respond. (Accra, May 2012) There were, however, only few caregivers in the study who had experiences like Maa Fosua and Awurakua where both the caregivers and the migrants had good financial conditions because in most of the other cases either the caregiver or the migrant had difficult financial situations. I provide an additional illustration where TCRAs worked well with only one person in the caregiver’s network. In this TCRAs not only were caregivers and the migrants both financially sound but also, because the child was older and taking care of them was not time consuming and did not require the input of many people. Therefore, between the caregiver and the migrant parents, they could provide care for the children with minimum assistance. Victor (a civil servant) lived in the house of his migrant aunt and was in charge of her 18-year-old daughter, with whom he had been living with for almost two years. When I asked Victor why Ata was put in his care, he said: Okay … being a young woman of her age with her mother being abroad, she needs somebody to direct her and also correct her when she is going wrong. Also, for instance if she falls ill at some point, I believe that even before her mother gets to know it, somebody should be a first point of call for her. Apart from that, we are all humans and once in a while there is some small assistance that on my part I can continue to offer her ( Accra, November, 2011) Victor’s role in her life was to guide her in the choices she made and also to ensure that she did not, for instance, stay out too late in town or engage in activities that any girl of her age might be tempted to do. According to Victor, the migrant expected him to report on Ata's activities from time to time. He functioned almost like a spy for his migrant aunt. He said University of Ghana http://ugspace.ug.edu.gh 211 Maa expects me to report to her every single thing Ata does. It is as if she is always expecting a negative report on what Ata is doing. There are things that I try to handle on my own and don’t find it necessary to tell Maa (Accra, November 2011). Ata also communicated directly with her mother and even received remittances directly from her. Ata’s mother seemed to be doing very well caring for her across borders, and since she was old enough to do things for herself, it appeared all the caregiver needed to provide was guidance and to be a reliable ‘spy’ to ensure Ata did not fall into bad behaviour. With these cases, I have demonstrated that not every TCRA needs a many people in their network of support to function properly. A collaborative effort between the caregivers and migrants was enough to provide care. 8.2.4 Caregivers Do Not Ask For Help Another factor accounting for small networks was that caregivers sometimes did not ask for help from people. There could be many people that caregivers could rely on (family, friends and teachers) for non-financial assistance. However, I found out that some of the caregivers did not deem it a desirable practice to let people outside of their immediate family know, for instance, that a child in their care was going wayward. This sentiment was especially so with grandmothers and wives in the sample. This, to a large extent, explained why caregivers told me that the children were not problematic even when they were reported to have told the migrants about the children's hard-headedness. Such talk would be seen as 'washing one’s dirty linen in public'. Thanks to the SMS methodology, I received a summary of an interview with a migrant from my Dutch colleague. I read from the notes that his teenage son was giving his mother a difficult time because he was becoming rebellious. The migrant feared that this might be detrimental to the old woman who was growing old and could not keep the boy in check and had to step in all the time to talk to the boy. Armed with this information, in my next interview with the old woman, I asked her about her grandson and whether she was having difficulties with University of Ghana http://ugspace.ug.edu.gh 212 disciplining a teenage boy. She declared that she could adequately handle him because he was a good boy. She added that her only problem was his inability to do house chores because he was boy. As the interview progressed, the issue of parenting came up, and she remarked in Twi (Akan) ‘mogya nti yen ton wo’, a statement that implied that one’s own mother would not ‘sell’ him to outsiders if he was bad. This explained why she was not willing to tell me about the boy even though she was the same person that told her son about the rebelliousness. It could, therefore, be anticipated that people were not willing to ask for advice in situations which might make them appear as exposing their problems to the outside world. In as much as it can be argued that she was not willing to tell me because I am not kin, it is important to note that most of these caregivers live in the urban centres and are more likely to be surrounded by non-kin since members of their families are dispersed across the region or the country. Thus, the most available people that caregivers can seek assistance from may not be kin and yet they were not comfortable ‘exposing’ their problems to outsiders. Sometimes, seeking advice in child care include advice to deal with stressful situations that have direct effects on the childcare arrangement, such as the wives whose husbands were arrested and jailed for living in the host country illegally. These women mentioned that they did not only have to deal with the situation of single-handedly providing material care for the children because they were not receiving any remittances, but that they also lost the emotional support they received from their husbands. Moreover, they could not afford to tell anyone to get the needed emotional or material about their concerns for fear of being stigmatized. Wives and grandmothers were largely affected by situations like these because they refused to ask for help in order to save their family from ‘shame’. University of Ghana http://ugspace.ug.edu.gh 213 8.4 Problems with Small Networks Whereas some caregivers did not face any challenges by having a small care network, for others small networks had adverse effects on them. There were cases in my study where caregivers could use the help of more people than those they currently had in their networks to assist with material care. Kumiwaa, a caregiver who was raising three children of her migrant sister and husband, explained to me how the children had sometimes had to stay at home for several days after their school had reopened because their parents had not sent them money for their school supplies and tuition fees. She intimated that some years before, the migrant (father) used to direct her to one of his friends for money. But later the friend of the migrant had also travelled abroad and, therefore, there was no one else to ask. She did not receive any more of such directives from the migrant, and she did not have anybody else she could borrow money from to repay later when she was reimbursed. Another set of caregivers who struggled with small networks were the ageing grandparents who did not have a lot of energy to take care of the children in their care. Some grandparents had very young children in their care. In some cases, they brought in house- helps to help them with child care. But this could be expensive since it meant there was an extra mouth to feed and person to take care of from the same meager resources. In one of my cases, the caregiver was the great grandmother of the child in her care. She was frail and almost always sick and the 11-year-old child in her care did most of the house-hold chores because she had few people to help her. Her own three children lived in different parts of the area and only checked up on her once in a while. There was a couple in her network on whom she relied for assistance with the child. The first time I had an appointment with this caregiver; it was the couple who met me for the interview. The caregiver had taken ill and the couple said she was at a health University of Ghana http://ugspace.ug.edu.gh 214 facility. She had been there for a month before my visit, and the child had been left in the couple's care. The new caregivers of the child explained to me that they had a hectic schedule which gave them barely anytime to offer proper care for the child. The original caregiver was quite worried about the child, because she did not seem to have much help for herself or for the child. Rather, the child was doing for her what she was supposed to be doing for the child and this really bothered her. There was another instance in which a grandmother mentioned that she sometimes had to let one of the older children in her care skip school to accompany her to the market or help her with the care of the younger children when she really had to go somewhere. She said that besides it being quite expensive to have house-helps, she had had bad experiences with them in the past and therefore did not want to work with them anymore. The caregiver had come from another region several years before when she married her husband, and they both worked and lived in Kumasi until her husband died. Most of her relatives still lived in her hometown, and thus she had none of them with her in Kumasi to help her with the children. 8.5 Summary In this chapter, I tried to establish two points. First, I showed that caregivers did not work in isolation in taking care of the migrants’ children in their care; they received help from people who can be described as their network of care. Second, I demonstrated that contrary to the existing literature and common knowledge, caregivers in my study did not have a big care network on which to fall for help in raising the children in their care. In my discussion, the important issues that came up include the fact that the caregivers' care network was drawn mostly from the migrants’ network of people with whom they kept some ties, and that the extent University of Ghana http://ugspace.ug.edu.gh 215 to which the migrants actively engaged in these relationships with their ties back home determined whom they could fall on or ask the caregiver to fall on for help in raising their children. Another interesting issue that came up was the fact that caregivers also had small networks because they did not ask to be helped. In addition, caregivers barely mentioned their own network of people – friends, church members or adult children in the network of care. It seemed to me that these were people on whom the caregiver could count for help and, in the cases of the caregivers in my work, who were also kin to the migrants, since most of these people (siblings and adult children) were somehow related to the migrant. However, it appears they were either not called upon or were just not willing to help. This buttressed the point made in the previous chapter six (6) that a person's commitment to migrants and their left-behind children are largely based on what kind of reciprocal relationship exists between them and the migrant. University of Ghana http://ugspace.ug.edu.gh 216 CHAPTER NINE: SUMMARY OF FINDINGS, CONCLUSION AND RECOMMENDATIONS 9.0 INTRODUCTION This study has focused primarily on the experiences of caregivers in Transnational Child Raising Arrangements (TCRAs) in Ghana. Caregivers have been considered to be very important to the maintenance of relationships within the household in the event of international migration. However, very few studies have focused on the experiences of caregivers from their own perspectives. Existing studies (Hondagneu-Sotelo and Avila, 1997; Suarez-Orozco, 2002; Parrenas, 2001, 2005) on TCRAs have focused largely on the dyad between migrant parents and their children, laying particular stress on the migrants' experiences in the host country, those of the child they leave behind and how they maintain close ties across borders. Furthermore, in what has been called a geographical bias (Mazzucato, 2013), the African perspective on this subject is largely absent as most of these studies have been conducted in Latin America and South East Asia. This study, therefore, contributes to scholarly work on TCRAs by focusing mainly on the experiences and perspectives of these ‘unsung heroes’ (Moran-Taylor, 2008) in TCRAs, while also adding to the scant literature on Africa on this subject. This is an ethnographic study that followed the lives of 15 caregivers in a matched sample (parent, caregiver and child) between Ghana and the Netherlands for over two years. The record of their experiences was supplemented with semi-structured interviews of 19 caregivers in a non-matched sample. The study employed the Simultaneous Matched Sample (SMS) methodology and the caregiver was the unit of analysis in each of the TCRAs. The study set out to understand how caregivers came into their caregiving roles, whom they relied on for support University of Ghana http://ugspace.ug.edu.gh 217 in raising the migrants' children, and how in performing their daily tasks they managed to negotiate between the demands of child fostering and transnational migration. This study analyses the caregivers' experiences through the lens of a conceptual framework that incorporates different aspects of transnational child raising arrangements including transnationalism, child fosterage and transnational care practices. 9.1 Summary of empirical findings In chapter 5, I presented and discussed the characteristics of the sample I used for this study. I highlighted the characteristics that were important in the analysis of my data and also highlighted some of the important issues that emerged from the discussion of the characteristics and how they contribute to existing literature on TCRAs. In consonance with the findings of the majority of studies on child fosterage, I found that most of the caregivers were kin to one or both of the migrant parents. Thirty of the caregivers, which constituted 88 per cent of my sample, were kin to a migrant parent. The remaining four caregivers (12 per cent) that were non-kin included a friend of the migrant, two teachers, and a paid caregiver, all of whom the migrants did not know prior to the care arrangements. These caregivers were contacted by the migrants on the basis of recommendations from people the migrants knew. In Ghana, it is not uncommon for children to be sent to live and be trained by non-kin members such as friends, religious leaders, and teachers, in what is referred to as wardship or apprentice fostering (Goody, 1982; Abanihe, 1984). But in most of the cases in this study, the caregivers were known to the parents because they usually lived in the same community and were chosen because they were highly respected by the community's members. However, the cases of the three non-kin caregivers were different. According to the caregivers, the migrants needed to move the children from previous care University of Ghana http://ugspace.ug.edu.gh 218 arrangements and were quite desperate to find a new caregiver. Akesson et al. (2012) refer to this as ‘re-arranging care’ and noted that this could be very stressful for some migrants. The circumstances under which these three and others in my sample became caregivers leads me to conclude that in the context of transnational migration, care can sometimes be arranged under desperate circumstances, especially if this happens in the course of a parent’s migration. This kind of situation is akin to crisis fosterage. Other cases where care was arranged out of desperation include a male caregiver in the study who was taking care of an eight-year-old girl without the assistance of a woman, a sick and frail 80-year-old caregiver who had an 11-year-old granddaughter in her care, and a maternal aunt who had to take in her niece's five children although she did not have the resources to do so. In all these cases, the migrants were desperate and did not have anywhere to send their children. The case of the paid caregiver in the non-kin category reveals the emerging phenomenon in TCRAs in Ghana in which a caregiver is hired and asked to live in the house of the migrant without the supervision of any of the migrants' family members. I found that hiring the services of a paid, live-in caregiver was costly, and only migrants who had the financial capacity could use this option. Care giving has mostly fallen to women and this is also reflected in the large number (n=28) of women in my sample. These were mostly grandmothers, mothers, or aunts of the children that were left behind. Yet, even though the majority of the caregivers were women, I found a few men who were also engaging in caregiving. There were six men in the sample, four of whom were raising the children without the support of any female relative, which stands in sharp contrast to the norms of parenting in Ghana, especially in TCRAs. The male caregivers in most studies on transnational parenting used the support of other female relatives and these were University of Ghana http://ugspace.ug.edu.gh 219 also usually the spouses of the migrants or the fathers of the children in question (see Asima, 2010; Fresnoza-Flot, 2013). In this study, none of the men were spouses of the migrants. I found, however, that the differences among caregivers based on their relation to the migrants or their sex did not reflect in their motivation to become caregivers, the responsibilities they shouldered and the way they experienced and managed care in their TCRAs. For instance, almost all the caregivers said they accepted becoming a caregiver primarily as a favour to the migrants who needed their help. They also all had a common responsibility to provide care to the children who had been entrusted to them. The difference in the activities they performed was largely based on the age of the children in their care. Additionally, how they experienced and managed challenges in the care arrangement had very little to do with their being kin or female, since almost all the caregivers were at some point faced with the common challenge of insufficient remittances. The characteristics which mostly shaped their experiences included their economic status, their ages vis-à-vis the ages of the children in their care, and the legal and socio-economic status of the migrants whose children were in their care. With respect to economic status, to a certain extent, the findings from my study challenge Hochschild's (2000) three-tier care chain theory, which asserts that caregivers were the last in the three-tier care chain and depended on the migrants' remittances in exchange for the care they provided for their children in the home country. Of the 34 caregivers, 21 were economically active and were engaged in various income-generating activities and did not have to depend on the migrant for their personal sustenance. Of the remaining caregivers who were not actively working, some had sources of income such as support from their spouses and their adult children. University of Ghana http://ugspace.ug.edu.gh 220 In Chapter 6, I discussed the pathways caregivers followed into their care arrangements and the expectations they had of the migrants and their children. I found that the caregivers followed two pathways, which I call the ‘involved’ and ‘uninvolved’ pathways. The vast majority of the caregivers, 33 out of 34, followed the former pathway, meaning they made a conscious decision to take care of the migrants' children. The caregivers who followed this pathway did so in three different ways. First, there were caregivers who encouraged the migrant to travel abroad to seek better prospects for the family and opted to take care of the children as an expression of their support for her/his decision to migrate. I observed that these caregivers were direct beneficiaries of the migration process and were usually the migrants' wives or mothers. The second group of caregivers includes those whom the migrant approached about providing care for their children, either at the time or in the course of migration. The third consists of one caregiver who asked the migrant to let him take care of her children when he realized that the previous caregiver was neglecting them. The ‘uninvolved’ pathway was followed by only one caregiver. Her pathway was more of a function of the tension in their relationship prior to her husband’s travel. Secondly, the study of the caregivers' pathways into the care arrangement also reveals that taking care of another person’s child sometimes involved what Baldassar et al. (2007) refer to as ‘working out’, where, for instance, the caregivers have to consider the past histories of their relationships with the migrant parents and their ability to take in the children. It is usually taken for granted that members of an extended family will ‘automatically’ take care of each other’s children. These notions have been engendered by some anthropological studies on child fostering in many West African families. Even though almost all the caregivers mentioned that they were caring for the children out of a normative obligation (Finch and Mason, 1991) for their University of Ghana http://ugspace.ug.edu.gh 221 kin, some of them made the decision to do so because of the migrants’ past act of kindness. Other caregivers had to assess their capacity to perform the task. Capacity encompassed resources such as time, money, enough room to accommodate the children, and one’s own physical and emotional well-being. Two caregivers initially declined the migrants’ request to accept their children because of their lack of capacity. One of them did because she did not have enough space to accommodate the children and the other did because he did not make enough income to support the children. I must add, however, that, only a few caregivers paid attention to the issue of capacity, however, their neglect to assess their capacity sometimes presented severe consequences for them. For example, some caregivers, especially grandparents, were not physically fit to take care of the children and this became problematic. Others had to shut down their businesses because they could not combine the care of the children with running them. Another observation made was the non-verbalized nature of the care arrangement (Chapter 6). One of the research questions of the study was aimed at examining the processes and discussions that preceded the care arrangement. The study found however, that these discussions seldom happened at all because the caregivers and migrants expected each other to know what to do for the children. The only verbal discussions that appeared to have been held were primarily the request of migrants and the caregiver accepting to take care of the children. Issues such as remittances, responsibilities, the duration of the care arrangement and the expectations of the migrants and the caregivers were hardly ever discussed. Caregivers went into the care arrangement with expectations, but these were usually latent because they hardly admitted that they had them. Such expectations became apparent when they expressed disappointment about how the migrants was not helping them to either set up a business or assist them with a specific need. Some caregivers, especially those who actively persuaded their University of Ghana http://ugspace.ug.edu.gh 222 relatives to migrate, expected to be able to count on the migrants as social capital, and they explicitly expressed this when asked about their expectations. Other caregivers also had expectations of being supported by the migrant parents, but because they did not discuss these with them, when various incidents arose, the caregivers became disappointed because their expectation of support from the migrant parents was not met. Sometimes, the lack of discussion about the care arrangement resulted in a general dissatisfaction with them, especially for those caregivers who were not the direct beneficiaries of the migration process. But, some of the caregivers were unable to explicitly voice their expectations of the migrants because of the cultural norms of reciprocity that govern many relationships in Ghanaian communities. The caregivers often said that they did not voice these expectations or explicitly ask the migrants to assist them in a time of need because the migrants ‘knew what to do’. That is, the migrants knew that in exchange of taking care of their children, they should be able to assist them without the caregivers having to directly ask for it. In chapter 7, I illustrate that the responsibility of taking care of children in TCRAs can be placed on two levels. The first of these are the tangible caring activities, which include feeding, bathing, caring for young ones, nursing them when they are sick, and helping with their school work. I call this 'providing care'. The second level, which I call 'managing care', entailed the situations that the caregivers had to juggle behind the scenes in order to provide adequate care. These included an array of invisible practices which went unnoticed most of the time. This chapter argued that the success of the former (providing care) is largely dependent on how the latter (managing care) is carried out. I identified several broad types of activities which fell within managing care: managing remittances, communication within and outside of the care arrangement, managing unmet expectations and the expectations of people outside the TCRA, University of Ghana http://ugspace.ug.edu.gh 223 unforeseen circumstances and alterations in lifestyle. What made managing care harder for the caregivers was that they reckoned that if they did not manage these aspects of care graciously, all their efforts to provide care would prove futile. Caregivers noted that migrant parents rarely acknowledged the efforts they had made in the past when something went wrong in the care of the children. This finding reveals that caregivers did more work in the care arrangement than most studies on TCRAs have acknowledged. For instance, I found that caregivers invested their own financial resources, withheld information they considered disturbing from the migrant parents, sometimes at the expense of their own happiness, and they constantly strove to live up to the expectations of people outside the TCRAs who might judge them harshly should they fail. This is one reason the perspectives of caregivers need to be accorded more attention in research. In Chapter 8, I examined the caregivers’ networks of support. The importance of networks in migration studies cannot be overemphasized. However, how these networks affect transnational child-raising arrangements has been underexplored. This study found that in the execution of their tasks, caregivers involved both local and international actors. These consisted mainly of family, friends and sometimes professionals, such as the children's teachers and religious figures. Using a name generator tool, I assembled the networks of support which the caregivers possessed. I followed this with in-depth interviews and observations made during visits. The study made several observations. First, the caregivers’ networks were comprised primarily of kin, including the migrant parent. Second, the people that supported the caregivers were usually members of the migrants’ network. This meant that they offered help to the caregiver based on their commitment to and their relationship with the migrant. Another observation was that the networks are small in size. Contrary to both what the literature on child-raising in West Africa has claimed, the caregivers in my study did not possess a big University of Ghana http://ugspace.ug.edu.gh 224 network of people on whom to fall for help in raising the children in their care. I was not able to compare my findings with existing studies on caregivers’ networks because, so far, there has been hardly any attention given to the caregivers' network of support. I proposed possible explanations as to why the networks were small, and these include the role of urbanization and the increasing dispersion of the family, which contributed to reduced kin support for child care. Another explanation for the small network was that the capacity of the caregiver and the migrant parents was sufficient to provide care for the child. Thus, some caregivers do not need a large network of support. This was often the case when the caregiver cared for older children who did not require nurturant care. In such instances, the efforts of caregivers and the migrants (the latter through remittances and communication), was enough to meet the care needs of the children. 9.2 Conclusions The following are the conclusions drawn from the empirical findings and the conceptual framework. First, caregivers give expression to the assertion made by Levitt and Nyberg-Sorensen (2004) that movement is not a prerequisite to engaging in a transnational social space. Even though none moved beyond Ghana's borders, the caregivers in this study played a very active role in the transnational social space, by way of their care-giving activities which took place in more than one nation. This study has shown that although the caregivers operate in the home country and act according to the norms of local parenting practices, they commit their time and energy, and sometimes their financial resources, to meet the physical and emotional needs of both the children in their care and the migrant parents abroad. In performing their daily tasks, University of Ghana http://ugspace.ug.edu.gh 225 they actively engage people at both ends of the care arrangements, while earnestly trying to protect the relationships involved in them. Second, the study concludes that the economic status of the caregiver plays an important role in maintaining a stable care arrangement. Having access to material resources, particularly money, was one of the main characteristics of well-functioning TCRAs, while the lack of it was deemed problematic for all the actors involved. I have illustrated in this study how caregivers who had a regular source of income absorbed the expenses of the children at the initial stages of migration as the migrant parents settled down in the host country. Care givers were also able to work with small networks because of their own ability to provide for the children. Also notable, is their ability to supplement remittances when they were insufficient or delayed, without negatively affecting the children or putting undue strain on the migrants. The caregivers often mentioned that they understood that migrants faced some financial constraints in the host countries. However understanding the migrant’s situation was not enough, and they needed to be able to ensure that the children’s needs were met. It was those who had access to money that succeeded at this. The financial assistance that caregivers rendered to the migrant parents and the children in their care, supports observations made by some authors that the people to whom migrants entrust responsibilities such as businesses, housing projects and the care of their children were seldom persons in need (Mazzucato, 2003; Smith, 2007). These authors added that migrants selected such people to reduce the likelihood that the remittances that are sent are used for purposes other than that for which they were intended. The caregivers’ access to financial resources in my study did not only ensure that migrants’ remittances were spent on the children in their care, but it also relieved the migrants of stress because the caregivers were able to use University of Ghana http://ugspace.ug.edu.gh 226 their own financial resources on the children when remittances were delayed or insufficient, which was very common in the TCRAs I studied. Third, the study also concludes that the rights and responsibilities of caregivers in TCRAs are largely determined by a combination of local parenting practices and some attributes of transnational migration namely, communication and remittances, which results in some unpleasant experiences for some caregivers. When child fostering is practiced within the context of transnational migration, there are different expectations with respect to the roles of the two ‘parents’ involved. The roles of the caregiver vis-a-vis the biological parents in a transnational context are different from those when fostering is practiced locally. This difference is engendered by the role of transnational care practices, such as the sending of remittances and communication within the TCRAs. I show that child fostering norms and some of the attributes of transnational migration operate in tandem to shape the experiences of caregivers and influence how the TCRAs function. While child fostering norms largely inform and guide the caregivers’ behaviour in TCRAs, I found that transnational migration also allows for active parental involvement in the upbringing of the children, and this is viewed differently by the caregivers. Additionally, child- fostering norms make it possible for parents to leave their children behind as well as help each member in the TCRAs to settle into the new parenting arrangements that come with migration. As a result of existing child fostering norms, TCRAs in Ghana and contexts similar to Ghana’s cannot be described as inherently traumatic or pathologic because of the normality of parent- child separation which is prevalent in the Ghanaian society. I also pointed out that the norms of child fosterage guide the behaviour of caregivers in the face of unmet expectations and a general dissatisfaction so that the TCRAs are not fraught with tension. University of Ghana http://ugspace.ug.edu.gh 227 However, with transnational migration comes a form of shared parenting between the caregivers and biological parents, a practice which is rare in local child fostering. Through transnational care practices such as sending remittances, communication and, to a lesser extent, return visits, the biological parents become actively involved in the children’s lives in spite of the geographical distance (Poeze et al, in press). Major decisions concerning the children, such as their education, have to be taken jointly with the parents, and, in most cases, the parents have to give their approval before they are carried out. The level of parental involvement was met with ambivalence by the caregivers because, on the one hand, they viewed the parent’s contribution positively since it took away the burden of being the sole financier of the children's care and decision making was also joint because of new communication technologies which enabled situations to be handled more quickly. On the other hand, the caregivers considered the migrant parents' active involvement in their children's care to stifle their efforts to perform their roles effectively rather than enhance them, particularly with respect to how to discipline children. The exercise of control that caregivers’ found frustrating was more prevalent in TCRAs in which the migrants were documented, had well-paid jobs and could therefore send remittances regularly for the children’s upkeep. The caregivers' position was further circumscribed by their inability to control communication within the TCRAs. Most of the children in the care arrangements had been given mobile phones by their parents to enable them communicate directly, and caregivers believed that these children served as the ‘eye’ of their parents in the care arrangement. Parental involvement, especially via communication, has driven caregivers to act carefully which also meant leaving the children to their own devices sometimes. They struggled with their own parenting ideals as they could not raise the children in ways they deem appropriate because they do not want their actions to be misinterpreted by the children to their University of Ghana http://ugspace.ug.edu.gh 228 parents. In the caregivers’ estimation, this attitude of the parents, abetted by transnational care practices, defeated the whole purpose of child fosterage. In their view, the practice of child fosterage was aimed at raising responsible and independent adults through strict training and discipline, a practice that most of the caregivers claimed to have benefitted from immensely during their own upbringing. Thus the management of care within fosterage in TCRAs had changed qualitatively from what pertains in traditional child fosterage. Fourth, this study adds to previous scholarly work on transnational parenting in Ghana which claims that TCRAs function well with comparatively favorable results. These studies (Coe, 2008; 2011; Dankyi, 2011; Poeze and Mazzucato, 2013) studied TCRAs mainly from the perspective of the children and concluded that TCRAs in Ghana are not characterized by neglect, abandonment and cases of traumatic separations, as have been found in studies from other social contexts. In her study of Guatemalan migrants, Moran-Taylor suggested to the President of Guatemala, who had called migrants ‘heroes’ because of their contribution to the country's economy, to also praise the caregivers who made it possible for these migrants to go and work in peace (Moran-Taylor, 2008). Moran-Taylor and others (see Dreby, 2010), commended care givers for their efforts in providing day-to-day proximate care, even when remittances were not forthcoming. The findings of this study suggest the same, stressing that caregivers, as well as the parents, make these TCRAs function, but that caregivers play a greater role in how well they function.. As this study has shown, it is the caregivers' ability to manage care in spite of the demands of local parenting norms and transnational migration that is most notable. I have shown (see chapter 7) that child-fostering norms place a lot responsibility on their shoulders and set a high standard for them to meet. This is made more difficult by parental involvement and their inability to control communication within and outside of the TCRAs. They therefore have to University of Ghana http://ugspace.ug.edu.gh 229 manage a lot of situations in order for their TCRAs to function well. Dreby calls the caregivers 'middle women' (2010), while Moran-Taylor (2008) refers to them as ‘unsung heroes and heroines’. In this study, I describe the caregivers as ‘managers’, for even though they do not always succeed in handling every difficult situation they encounter, their ability to manage the majority of the situations is what holds their TCRAs together. Finally, I conclude that the findings of this study have enriched the existing studies on transnational parenting, especially those that focus on caregivers, because of the conceptual framework that was developed for the study. The framework incorporated the local parenting context and the transnational element of migration in the analysis of caregivers’ experiences in TCRAs. This study has not only discussed various aspects of the caregivers' experiences, such as their pathway into care, their expectations, their care networks and their tasks as caregivers, but it has also shown how these experiences are shaped by both local parenting norms and some attributes of transnationalism. Thus, a framework such as the one developed for this study is useful to capture a more nuanced picture of the experiences of caregivers. For instance, it becomes possible to understand why a caregiver agrees to take care of the children of migrants, even when s/he does not have the financial capacity to do so. It also becomes explicable why some caregivers remain in a care arrangement they complain bitterly about, in contrast to some of the caregivers in other social contexts who easily turn down a migrant's request to raise her child or ‘quit’ the TCRAs when migrants do not send remittances regularly for the child’s upkeep (see Dreby, 2010). A conceptual framework such as this one helps to address some important gaps in the literature on the experiences of caregivers as well as helps scholars to appreciate the differences in the experiences of caregivers in different contexts. University of Ghana http://ugspace.ug.edu.gh 230 9.3 Recommendations Based on the findings and conclusions made, I now provide some practical suggestions on how TCRAs can be made to function better for the benefit of the caregivers, the children in their care and the migrant parents. Even though my findings and conclusions are based on an analysis of data on the caregivers, the suggestions below are targeted towards both caregivers and the migrants, since the functioning of the TCRAs rest squarely on both of their shoulders. I end by echoing recommendations made by scholars on this subject for state policy makers. My first recommendation is that caregivers and migrants freely communicate the terms of the care arrangement before, and during the course of migration. Throughout the study, non- communication was found to be one of the greatest burdens that the caregivers bore. The caregivers' non-communication of their expectations concerning remittances and personal rewards was due largely to the norms of child fosterage and reciprocity. They could not discuss remittances because the children are considered to be theirs too and they feared they would be seen as being greedy, nor could they discuss their expectations of reward for fear of being considered as opportunists. They could also not decline a request to take care of children even when they did not have the capacity to do so as they thought this would make them unpopular among their kin. Taking care of each other’s children, bearing all the costs involved and not openly asking to be rewarded, are all ideals that used to govern families many years ago, at least according to earlier anthropological studies. In contrast, the reality today is that it is very costly to raise children in Ghana. Therefore, caregivers should be encouraged to discuss remittances openly and to not leave this to the discretion and condition of the migrants. Caregivers should also be encouraged to discuss rewards openly without feeling guilty since reciprocity is not guaranteed. If caregivers are able to communicate these expectations more freely, it will reduce University of Ghana http://ugspace.ug.edu.gh 231 their dissatisfaction with the care arrangement as well as help reduce the burden which the challenges of insufficient remittances place on them. While communicating their expectations does not guarantee that they will necessarily be met because of the conditions that migrants experience in the host country, the freedom to talk about them would help to dispel doubts in the minds of the caregivers’ with respect to some actions on the part of the migrants. The second recommendation also concerns communication in the TCRAs with respect to remittances, as most of the issues that bred tension and mistrust in the TCRAs were related to them. It is very important to freely discuss issues concerning remittances, such as how much the migrant should remit, and how the caregivers should manage insufficient remittances. However, to be able to this without raising doubts in the minds of the migrants and the older children in their care, I suggest that caregivers keep records of their expenditures. During the research period, I came across two caregivers who kept records of the expenses they made on the children in their care. One of these did this so she could claim reimbursement for the excess money she spent on the children over what she received in remittances. She had resorted to this practice because the migrant did not believe her when she said she spent as much as she did. The other caregiver kept the record for personal reasons. As she explained it, she just wanted to have a good idea of what her expenditures on the children were. I recommended this exercise to one caregiver who allowed me to go through her expenses with her. She was amazed by how much she spent on the children and by how much this was in excess of the remittances. Keeping records of expenditures will help caregivers in two ways: first, it will help those caregivers who want to claim reimbursement; and second, it will enable caregivers to provide the migrants and older children who maintain doubts about how the remittances are spent with a much better idea of the costs s/he incurs in caring for the children. In addition, sometimes migrants remit the University of Ghana http://ugspace.ug.edu.gh 232 same amount of money over a long period of time on the assumption that rising exchange rates as in Ghana render the unchanging level of remittance sufficient. Keeping better records of expenses, then, would help to maintain the caregivers' and migrants' peace of mind in the TCRAs. Additionally, I recommend that migrants arrange for extra help when they leave their children with their parents because, as their parents age and the children grow up, it becomes extremely difficult for them to bear the responsibility of child care alone. In addition, extra help is not only needed in the course of migration, but also before the migrants depart, because some of the grandparents are already old and physically unfit at the time the migrants emigrate. This study found that the capacity to provide care played a minimal role in caregivers' decisions to care for a migrant’s child. For reasons related to kinship and familial arrangements, caregivers agree to take care of the children without fully assessing their capacity to do so, which, as explained in Chapter 6, can sometimes have negative consequences for them. One group of caregivers where I found this to be crucial was grandmothers who were raising their grandchildren, but were not physically fit to do so. Coupled with their inability to physically help the children was the emotional burden of knowing that the children might go wayward because they were losing their grip on them. Migrants sometimes left children with the grandmothers because most of them insisted on taking care of them. In fact, so long as they are alive, grandmothers expect the children to be left in their care because of the double advantages this holds for them: the guarantee of receiving assistance from their children and companionship from their grandchildren. For this reason, I suggest to migrants that they provide help to grandmothers, either by sponsoring a relative to help them, or by hiring the services of a house-help if they can afford to do so. Sometimes, due to how the travel arrangement works out, migrants are not able University of Ghana http://ugspace.ug.edu.gh 233 to make such provision before they leave. But it could be useful if these arrangements are made as soon as they settle in the host country. It is in the interest of the grandmothers and the children the migrant parents leave behind that they do this. It has been claimed by some scholars that TCRAs which involve grandmothers creates double care situations for migrants because there is both a mother and a child left behind for whom the migrant feels equally responsible (Akesson et al., 2012). I further recommend that the concerns of caregivers be seriously considered in the formulation of childcare policies. The burden that the care of migrant children poses for their caregivers- materially, physically and sometimes emotionally- has implications for policies on care and caregiving in Ghana. Care-giving within a kinship system such as child fosterage has hardly been the subject of attention both in research and policy. However, both anecdotal and empirical evidence available on the subject indicate that child care is presently being provided under some structural constraints. These include the impact of the processes of globalization and their associated demographic and economic transformations on the customary responsibilities of marriage, kinship and parenthood in Ghana. These transformations have drawn the attention of researchers and to some extent policy makers, to the subject of childcare in Ghana. (See edited volume on child care in globalizing world: perspectives from Ghana by Oppong, Badasu and Waerness, 2012). Yet, the attention has so far largely been on how the children are being taken care of within these changing circumstances. The findings of the present study, however, raise the question of what these changes mean to the caregivers. What are the opportunities and challenges that the changes in the landscape of child care pose for the caregivers? This recommendation is even more relevant, especially with the introduction of the Care Reform Initiative by the Department of Children under the Ministry of Gender, Children and Social University of Ghana http://ugspace.ug.edu.gh 234 Protection. That initiative is aimed at providing family-based foster care for orphaned and vulnerable children who cannot live within their own family settings. But it is important that the initiative places equal emphasis on the welfare of the caregiver, as the welfare of the children is inextricably linked to that of the caregivers. In placing the children with the foster families, there is the need to consider the capacity of the families, especially the primary caregiver (foster parent), to provide care, and to prepare them adequately, both physically and emotionally, on the duration of the care arrangements and also on anticipated challenges. Policies on child care should consider these factors, the absence of which might in the long term have negative implications for the children in foster care. I recommend for care to be considered as paid work as was the case of the paid caregiver in my study, whereby caregivers can earn some income in the course of providing and to save towards their old age. Studying the perspectives of caregivers also helps to bring into the open the taken for granted notion of care that is provided by these caregivers. Considering how much caregivers’ lose in terms of time, financial resources and sometimes changes to their lifestyle, as they struggle to provide care, the question that arises is ‘who takes care of the caregiver’? Ultimately, Ghana might move towards introducing policies that formalize care by introducing paid caregiving even within ‘informal’ systems. This can include setting eligibility criteria, and building the capacity of caregivers, as well as systems of accountability and provision of social security for caregivers during the course of providing care and in their old age. This may be a challenging step as care is usually provided within the kinship system, and placing monetary value on care may not be immediately accepted, however, alternative arrangements such as urging migrants as in the case of transnational caregivers to consciously set money aside towards some form of social security for the caregivers can be initially considered. For other categories of University of Ghana http://ugspace.ug.edu.gh 235 caregivers such as those of orphaned and vulnerable children, other family members who do not have the time to provide care can contribute a fixed amount periodically towards the upkeep of the primary caregiver. Additionally, they can also provide other forms of support such as paying for extra help so that the physical burden on the caregiver can be reduced. I reiterate a recommendation made by some researchers (see Coe, 2011; Dito et al. in press; Poeze and Mazzucato, 2014) that because TCRAs function well and the material support of transnational parents is so important to children back home and even to some caregivers as well, there is a need for policies in the host countries that can improve migrants’ wages, their ability to send money and gifts back home, and the ease with which they can visit their left- behind children. Such policies would include the enforcement of labour laws so that both documented and undocumented workers would be less likely to be exploited. Opening up pathways to legalization would also help in aiding migrants to visit their families regularly or to bring family members to visit in the host country, as well as to gain access to a wider and better- compensated array of employment opportunities. Their ability to send an adequate amount of money to cover the living expenses of the children and even of the caregiver will ease the caregivers’ burden and improve the situation in the TRCAs. My next recommendation echoes a quote by Takyiwaa Manuh: “Children’s well-being impact positively on society as a whole. Ministries and schools in Ghana should therefore pay more attention to caregivers who carry a huge responsibility” (Mazzucato and Manuh, 2013) It will be very helpful if schools especially the primary, junior high and senior high levels where most of the children of migrants are found strengthen the capacity of Parent Teacher Associations (PTAs) and school counseling departments to reach out to the caregivers and children of migrants in their schools. The schools that the children attend are the best media University of Ghana http://ugspace.ug.edu.gh 236 through which caregivers of migrants’ children can be reached and offered the necessary assistance. Additionally, churches often send letters on special occasions to their members who have travelled abroad to send remittances towards one specific project or another. They also hold prayer sessions and vigils for the migrants from time to time especially if they have reason to believe the migrants are facing difficult situations in the host country. In the same vein, the churches can reach out to the caregivers and children of the migrants by organizing special programs and providing an avenue for them to draw support should the need arise. Many Ghanaians are ambivalent towards seeking support from formal institutions and would rather prefer to rely on their family network or their religious institutions for support. The churches can take advantage of this and reach out to their members including caregivers of children of the migrants who belong to their churches. These caregivers may not be necessarily members of these churches but they could be reached through the migrants or their children. The Catholic Church in the Philippines for instance, has embarked on several projects to reach out to the children of migrants and their caregivers (Asis et al, 2004). The final recommendation is to Government, Ministries and all stakeholders to appreciate the efforts of caregivers of migrants’ children who make it possible for the migrants to emigrate and leave their children behind. Their efforts provide the migrants the peace of mind to work in the host countries and send the voluminous remittances that the stakeholders love to tout. I recommend therefore that in addition to praising international migrants for their contribution to building the national economy and formulating policies for them based on findings of the several studies that have been carried out on migrants’ remittances, they should begin to pay attention to the caregivers’ of the children of migrants. Their contribution may not be monetary and thus usually goes unnoticed because there is no way to capture and turn it into profits, however, University of Ghana http://ugspace.ug.edu.gh 237 caregivers are worth some amount of recognition for their efforts and deserve some attention in policy too. A good place to start is for the relevant ministries and institutions to take particular interest in findings of research activities such as the TCRA project to have some data on the people that migrants leave behind especially children and their caregivers. This will inform the development of an appropriate policy framework that covers them. 9.4 Recommendations for future research There are some areas of work that I would recommend for future research: First, I observed that the socio-economic and legal status of the migrants whose children were being raised by the caregivers in this study contributed to how caregivers provided care. I encountered two categories of caregivers in this regard. The first category includes caregivers who cared for the children of migrants who had had only primary education or no formal education at all. A large number of these migrant parents did not have resident permits in the host country. The second category includes caregivers who cared for the children of migrants who had a regular status in the host country, had a very high level of education (up to the tertiary level) and had very well paying jobs in the host country. These two groups of caregivers presented a different set of experiences, and further research, possibly comparing the experiences of caregivers in the two categories of TCRAs I have identified, would help establish the role of class and documented status of migrants in shaping the experiences of caregivers. Second, the study used a Social Network Analysis method to examine the caregivers’ networks of support. This study is one of the first to have focused on the networks in transnational child raising arrangements or, more specifically, on the caregivers' networks. This University of Ghana http://ugspace.ug.edu.gh 238 has resulted in several insights, but more research is needed in this area, such as a study on the inter-linkages and connections among the members of individual caregivers and the establishment of the weak or strong ties among network members. This will provide a more complete view of the impact of network support on the experiences of the caregivers. In addition, more work is recommended on the historical relationships that make up the caregivers' networks of care over time in order to establish how members came into and/or fell out of networks. Such research can also help trace which factors that affect the size of caregivers’ networks can be attributed to international migration. The third recommendation is research which can measure the actual impact of TCRAs on the economic, social, physical and emotional well being of caregivers as has been done on the impact of separation on the migrant parents and the children. A good example of this is the larger TCRA project that have carried out both quantitative and qualitative studies to, among other things, measure the impact of separation on the parents in the Netherlands and their left behind children. (Mazzucato and Ceboatari, 2014; Poeze and Mazzucato, 2014; Dito et al., 2014) Such research will not only expand the literature and add to the debates on the subject, but will also give a fuller picture of the impact of TCRAs on all the actors involved in them. University of Ghana http://ugspace.ug.edu.gh 239 References Aboderin, I. 2004. Decline in material family support for older people in urban Ghana, Africa: Understanding processes and causes of change. Journal of Gerontology, 59B (3): S128- S137. Adaawen S. and Owusu B. 2013. North-South Migration and Remittances in Ghana. African Review of Economics and Finance, 5(1): 29-45 Adamchak, D. J. 1995. Pensions and Household structure of older Persons in Namibia. Southern African Journal of Gerontology, 4(2):11-15. Adamu-Issah, M., Elden L., Forson, M., Schrofer T. 2007. Achieving Universal primary education in Ghana by 2015: A reality or a dream? A UNICEF working paper http://www.unicef.org/french/videoaudio/PDFs/Achieving_Universal_Primary_Educatio n_in_Ghana_by_2015.pdf retrieved on 28/11/2013 Addo, N.O. 1981. Population, Migration and Employment: The Case of Ghana. Unpublished, University of Ghana, Legon. Adepoju, A. 2010. International Migration within, to and from Africa in the Globalised world. Accra: Sub-Saharan Publishers. Adepoju A. 2007. Migration in sub-Saharan Africa. A background Paper commissioned by the Nordic African Institute for the Swedish Government white Paper on Africa. Adepoju A. 2005. Patterns of Migration in West Africa. In: T. Manuh (Ed). “At Home in the World”. International Migration and Development in Contemporary Ghana, West Africa. Accra: Sub-Saharan Publishers. 24-55 Adu Boahen, A. 1986. Topics in West African History. Essex: Longman. Afoakwa E. O. 2007. Home grown school feeding programme-the Ghanaian model as icon for Africa. http://www.gcnf.org/library/Ghana-School-Feeding-Programme-Overview-and- Progress.pdf retrieved on 28/11/2015 Ȧkesson, L., Carling, J. and Drotbohm, H. 2012. Mobility, moralities and motherhood: Navigating the contingencies of Cape Verdean lives. Journal of Ethnic and Migration Studies, 38(2): 237-260. Alicea, M. 1997. “A Chambered Nautilus”: The contradictory nature of Puerto Rican women’s role in the social construction of a transnational community. Gender and Society, 11(5): 597-626. Anarfi, J. K. and Fayorsey, C. K. 1999. Male participants in the ‘commercialization’ of aspects of the female life cycle in Ghana. The Continuing African HIV/AIDS Epidemic, 109-12. Anarfi, J., Kwankye, S., Ababio, O-M., Tiemoko, R. 2003. Migration from and to Ghana: A Background paper. University of Sussex: DRC on Migration, Globalization & Poverty. Appadurai, A. 1996. Modernity at Large: Cultural dimensions of Globalization. Minneapolis: University of Minnesota Press. Asima D. 2010. Continuities and Discontinuities in Gender Ideologies and Relations: Ghanaian Migrants in London. Doctoral Thesis, University of Sussex. Retrieved from http://sro.sussex.ac.uk/6268/1/Asima,_Prosper_Price_Delali.pdf on 31/01/2012 Asis, M., Haung, S. and Yeoh, B. 2004. When the light of the home is abroad: Unskilled female migration and the Filipino family. Singapore Journal of Tropical Geography, 25(2): 198- 215. Assimeng J.M. 1999. Social Structure of Ghana: A study of persistence and change. Accra: Ghana Publishing Corporation University of Ghana http://ugspace.ug.edu.gh 240 Awumbila, M. and Ardayfio-Schandorf, E. 2008. Gendered poverty, migration and livelihood strategies of female porters in Accra, Ghana. Norwegian Journal of Geography, 62 (3): 171–179. Babbie E. 1992. The practice of social research. California: Wadsworth Badasu, D. 2004. Child Care among Ewe migrants in the city of Accra: Cases of crisis. Research Review Supplement, 16. Institute of African Studies: University of Ghana Baldassar L.V. 2007. Transnational families and aged care: The mobility of care and the migrancy of ageing. Journal of Ethnic and Migration Studies, 33(2): 275-297. Baldassar L., Baldock C., Wilding, R. 2007. Families caring across borders: Migration, aging and transnational care giving. Canada: Palgrave Macmillan Banfi, L. and Boccagni, P. 2011. Transnational family life and female migration in Italy: one or multiple patterns? In: A. Kraler, E. Kofman, M. Kohli and Schmoll C. (Eds) Gender, Generations and the Family in International Migration. IMISCOE Research: Amsterdam University Press. 287-313. Basch, L., Glick Schiller, N. and Szanton-Blanc, C. (Eds.) 1994. Nations Unbound: Transnational Projects, Postcolonial Predicaments and Deterritorialized Nation-States. Amsterdam: Gordon and Breach. Bernard, J. K., Landolt, P., & Goldring, L. 2005. Transnational, multi-local motherhood: Experiences of separation and reunification among Latin American families in Canada CERIS Working Paper no.40. Toronto: Joint Centre of Excellence for Research on Immigration and Settlement. Bernardi, L. 2011. A mixed-methods social networks study design for research on transnational families. Journal of Marriage and Family, 7: 788 – 803. Bigombe, B. and Khardiagala G. M. 2003. Major trends affecting families in Sub-Saharan Africa. New York. United Nations. Retrieved from http://www.un.org/esa/socdev/family/Publications/mtbigombe.pdf on 22/06/2012 Blanc, A. K. and Lloyd, C. B. 1990. Women’s childrearing strategies in relation to fertility and employment in Ghana. New York Population Council. Bledsoe, C & Sow, P. 2011. Back to Africa. Second chances for the children of West African Immigrants. Journal of Marriage and Family, 73(4): 747-762. Bledsoe, C. 1990. ‘No success without struggle’: Social mobility and hardship for foster children in Sierra Leone. Man, 25 (1): 70-88. Brah, A. 1998. Cartographies of Diaspora: Contesting identities. London: Routledge. Bryceson, D. F., and Vuorela, U. 2002.Transnational families in the twenty-first century. In: D.F. Bryceson, and U. Vuorela (Eds.), The transnational family: New European frontiers and global Networks. Oxford: Berg. 3-30 Bump, M. 2006. Ghana: Searching for opportunities at home and abroad. Institute for the Study of International Migration, Georgetown University. Burt, R. 1984. Network items in the General Social Survey. Social Networks, 6(4): 293-339. Byrman A. 2008. Social Research Methods. United Kingdom: Oxford University Press Campbell, K.E., and Lee, B.A. 1991. Name generators in surveys of personal networks. Social Networks, 13 (4): 203-21. Carling, J. 2008. The determinants of migrant remittances. Oxford Review of Economic Policy, 24: 581–598. University of Ghana http://ugspace.ug.edu.gh 241 Carling, J., Menjıvar, C. and Schmalzbauer, L. 2012. Central themes in the study of transnational Parenthood. Journal of Ethnic and Migration Studies, 38(2): 191-217. Cancian, F.M and Oliker, S. J. 2000. Caring and Gender. Thousand Oaks, CA: Pine Forge Castaneda, E. 2009. “Remittances.” Pp. 904-7 in the Palgrave Dictionary of Transnational History: Palgrave Macmillan. Castenada E. and Buck L. 2011. Remittances, transnational parenting, and the children left behind: Economic and psychological implications. The Latin Americanist, 55(4): 85-110 Cebotari, V., & Mazzucato, V. (2014). Education of children of migrant parents in sub-Saharan Africa. Paper presented at the Transnational families: Multi-local, multi-sited and institutional perspectives, Amsterdam, June 26-28, 2014. Clark, G. 1999. Mothering, work and gender in urban Asante ideology and practice. American Anthropologist, 101(4): 717-729. Clinton H.R. 2006. It takes a village. New York: Simon and Schuster Coe, C. 2008. The structuring of feeling in Ghanaian transnational families. City and Society, 2(20): 222-250 Coe, C. 2011a. What is love? The materiality of care in Ghanaian transnational families. International Migration 49(6): 7-24. Coe, C. 2011b. What is the impact of migration on transnational family life? Women’s comparisons of internal and international migration in a small town in Ghana. American Ethnologist, 38(1): 148-163. Cohen, R. 1997. Global Diasporas: An introduction. Seatle, University of Washington Press. Cormode, L., & Hughes, A. 1999. The economic geographer as a situated researcher of elites. Geoforum, 30: 299-300. Creswell, J.W. 2009. Research Design: Qualitative, Quantitative, and Mixed Methods Approaches. London: Sage Cresswell J.W and Miller D.L 2000. Determining validity in qualitative inquiry. Theory into Practice, 39(3): 124 -131. Dankyi E. 2011. Growing up in a transnational household: A study of children of international migrants in Accra. Ghana Studies, 14: 133-162. de Haas, H. 2010a. Migration and Development: A theoretical perspective. International Migration Review, 44 (1): 227-264 de Haas, H. 2010b. Remittances, Migration and Development: Policy Options and Policy Illusions. In: K. Hujo and N. Piper (Eds.) South–South Migration: Implications for Social grave Macmillan and UNRISD Denzin, N. K. and Lincoln, Y. S. Eds. 2005. Handbook of qualitative research. . Thousand Oaks, CA: Sage. Dewind J. and Holdaway J. 2008. ‘Internal and International migration and development: Research and policy perspectives. In: J. Dewind and J. Holdaway (Eds), Migration and development within and across borders: Research and Policy perspectives on internal and international migration. Geneva and New York: IOM and the Social Science Research Council (SSRC). 15-23 Dito, B. and Mazzucato, V. Schans, D. 2014. The effects of transnational parenting on the Subjective health and emotional well being of Ghanaian migrants in the Netherlands. Paper presented at the conference 'Transnational families: Multi-actor, multi-sited and institutional perspectives’, 26-28 June 2014, Amsterdam. University of Ghana http://ugspace.ug.edu.gh 242 Dolphyne. F.A. 1991. The emancipation of women: An African perspective. Accra: Ghana Universities Press Dooley D. 1995. Social research methods. New Jersey: Prentice Hall Dreby, J. 2010. Divided by borders: Mexican migrants and their children. Berkeley: University of California Press. Dreby, J. 2007. Children and power in Mexican transnational families. Journal of Marriage and Family, 69 (4): 1050-1064. Dreby, J. 2006. Honor and virtue. Mexican parenting in the transnational context. Gender & Society, 20(1): 32-59. Dressel, P. and Clark. A. 1990. “A critical look at family care.” Journal of Marriage and the Family, 52 (3): 769-782. Dzorgbo, D. S. 1998. Ghana in Search of Development: The Challenge of Governance, Economic Management and Institution Building. Uppsala: Uppsala University, Sweden. Emerson, R. M., Fretz, R.I., and Shaw, L. L. 2011. Writing ethnographic field notes. Chicago: University of Chicago Press England, P. 2005. Emerging theories of care work. Annual Review of Sociology, 31: 381-399. Erel, U. 2000. Reconceptualizing motherhood: Experiences of migrant women from Turkey living in Germany. In: D. Bryceson, & U. Vuorela (Eds.), the transnational family. NewEuropean frontiers and global networks. Oxford: Berg. 127-146 Ehrenreich, B. and Hochschild A. (Eds) 2002. Global Woman: nannies, maids and sex workers in the new economy. London: Granta Books. Evergeti V. and Ryan L. 2011. Negotiating transnational caring practices among migrant families. In: A. Kraler, E. Kofman., M. Kohli and C. Schmoll (Eds) Gender, generations and the family in international migration. IMISCOE Research: Amsterdam University Press. 354-373 Falzon M-A . 2009. Multi sited Ethnography: Theory, Praxis and locality on contemporary research. In: M.-A. Falzon (Ed.), Multi-sited ethnography. Theory, praxis and locality in Contemporary research. Bodmin, Cornwall: MPG Books Ltd.1-24 Fayorsey C. 1993. Commoditization of childbirth: Female Strategies towards autonomy among the Ga of Southern Accra. Cambridge Anthropology, 16 (3): 19-45 Fiawoo D. K. 1978. Some patterns of foster care in Ghana. In: Oppong C. et al (eds), Marriage and fertility and parenthood in West Africa. Changing African Family Monographs. Canberra: Australia National University Press. 273-288 Fielding N. G. 2006. Life history Interviewing. In Victor Jupp. Sage Dictionary of Social Research methods. http://srmo.sagepub.com/view/the-sage-dictionary-of-social-research- methods/n107.xml retrieved on 25/05/2014 Finch J. and Mason J. 1993. Negotiating family responsibilities. London: Routledge. Finch J. and Mason J. 1991. Obligation of kinship in contemporary Britain: Is there normative agreement? The British Journal of Sociology, 4(3): 345-367. Findley S. 1997. Migration and Family Interactions in Africa. In: Adepoju A. Ed. Family Population and Development in Africa: London: Zed Books. Fine, M. 2005. Individualization, risk and the body: Sociology of Care. Journal of Sociology. 4(1): 247-266 Fisher, B. and Tronto, J. 1990. “Towards a Feminist Theory of Care”. In: E. Abels & M. Nelson (Eds) Circles of Care. Albany N. Y. State University of New York Press. 35-63 University of Ghana http://ugspace.ug.edu.gh 243 Fontana, A and Frey, J. 2005. The interview: from neutral stance to political involvement. In: N. Denzin, and S. Lincoln (Eds). The Sage handbook of qualitative research. London: Sage. 695-727 Folbre, N. 1995. "Holding hands at midnight": the paradox of caring labor. Feminist Economist. 1(1): 73-92 Fresnoza-Flot, A. 2013. ‘Men’s care giving practices in Filipino transnational families: A case study of left-behind fathers and sons’. In L. Baldassar and L. Merla (eds.), Transnational Families, migration circulation of Care: Understanding Mobility and absence in family life. Routledge Researh in Transnationalism. 170-184 Gamburd, M. R. 2000. The kitchen spoon’s handle: transnationalism and Sri Lanka’s migrant Households, Ithaca, NY: Cornell University Press. Gardner, K. and Grillo, R. 2002 (eds) Transnational Migration and Ritual, Special Issue, Global Networks, 2(3): 179-190 Ghana Statistical Service (GSS). 2013. Ghana Living Standard Survey. Report of the sixth Round (GLSS 6), GSS, Accra. Ghana Statistical Service (GSS). 2012. 2010 Population and Housing Census, Summary Report of Final Results, May, Accra Ghana Statistical Service (GSS). 2008. Ghana Living Standard Survey. Report of the Fifth Round (GLSS 5), GSS, Accra. GHANADISTRICTS.COM Accra Metropolitan Assembly. 2013. Retrieved 10/09/2013 from: http://www.ghanadistricts.com/districts. GHANADISTRICTS.COM Kumasi Metropolitan Assembly. 2013. Retrieved 10/09/2013 from: http://www.ghanadistricts.com/districts. GHANADISTRICTS.COM Offinso South Municipality . 2013. Retrieved 10/09/2013 from: http://www.ghanadistricts.com/districts. Glasgow, G. and Gouse-Sheese, J. 1995. Themes of rejection and abandonment in group work With Caribbean adolescents. Social Work with Groups, 17(4): 3-27. Glick Schiller, N., Basch, L., & Szanton-Blanc, C. 1992. Towards a Transnational Perspective on Migration: Race, Class, Ethnicity, and Nationalism Reconsidered. New York: New York Academy of Sciences. Glick Schiller N., Basch L., and Szanton-Blanc C. 1995. From immigrant to transmigrant: Theorizing transnational migration. Anthropological Quarterly 68(1).:48-63. Glick Schiller, N. and Fouron, G. 2001. Georges Woke Up Laughing: Long-Distance Nationalism and the Search for Home. Durham, NC: Duke University Press. Goody E. 1982. Parenthood and Social Reproduction: Fostering and occupational roles in West Africa. Cambridge: Cambridge University Press. Goody J. and Goody E. 1967. The Circulation of Women and Children in Northern Ghana. Man, New series, 2(2): 226-248. Graham, H. 1983 ‘Caring, a Labour of Love’. In: J. Finch and D. Groves (Eds) A Labour of Love: Women, Work and Caring. London: Routledge and Kegan Paul. 13-30 Graham, E. & Jordan, P. 2011. Migrant parents and the psychological well-being of left-behind children in Southeast Asia. Journal of Marriage and Family, 73: 763-787. Granovetter, M. S. 1985. Economic action and social structure: The problem of embeddedness. American Journal of Sociology, 91: 481-510. Grillo, R. 1998. Transnational. Paper presented to the Social Anthropology Graduate/Faculty Seminar, University of Sussex University of Ghana http://ugspace.ug.edu.gh 244 Gaurnizo L.E. 2003. The economics of transnational living. International Migration Review, 37(3): 666-699 Guarnizo, L. and Smith, M. 1998. “The locations of transnationalism”. In: M.P. Smith and L. E. Guarnizo (Eds) Transnationalism from Below. New Brunswich, NJ: Transaction Publishers. 3-34 Gyekye K. 1996. African Cultural Values. Accra. Sankofa Publishing Company Hadi, A. 2001. International Migration and the Change of Women’s Position among the Left- Behind in Rural Bangladesh. International Journal of Population Geography, 7(1):53-61. Heyl, B. S. 2001 “Ethnographic interviewing,” in P. Atkinson, A. Coffey, S. Delamont, J. Lofland, and L. Lofland (Eds) Handbook of ethnography. London: Sage Hochschild, A. R., 2000. Global Care chains and Emotional Surplus Value. In: A. Giddens and W. Huttons (Eds) On the Edge: Globalization in the New Millennium, London: Sage Publications. 130-146 Hondagneu-Sotelo, P and Avila E. 1997. “I am here, But I am There’: The meaning of Latina Transnational motherhood”. Gender and Society, 11(5):548-71 Horst, H. 2006. The blessings and burdens of communication: cell phones in Jamaican Transnational social fields. Global Networks 6, (2 ): 143–159. Horton, S. 2009 ‘A mother’s heart is weighed down with stones: A phenomenological approach to the experience of transnational motherhood’, Culture Medicine and Psychiatry, 33(1): 21-40. Institute of Statistical, Social and Economic Research (ISSER) 2013. State of the Ghanaian Economy in 2012, ISSER, University of Ghana, Accra. International Organisation for Migrants (IOM). 2009. Migration in Ghana: A Country Profile, IOM, Geneva. Isuigo-Abanihe U.C. 1984. Prevalence and determinants of child fosterage in West Africa: Relevance to Democracy. African Demography Working Series, Populations Study Centre. University of Pennsylvania Scholarly Commons Isuigo-Abanihe U.C. 1985. Child Fosterage in West Africa. Population and Development Review. 11(1): 53-72 Itzigsohn, J. Cabral C.D., Medina E. H. & Vazquez O. 1999 Mapping Dominican transnationalism: narrow and broad transnational practices. Ethnic and Racial Studies, 22 (2): 316-339 Kabki M. 2007. Transnationalism, local development and social security: the functioning of support networks in Ghana. Leiden: African Studies Centre Kabki ,M., V. Mazzucato & E. Appiah (2004), ‘Wo benane a eye bebree’, The Economic Impact of Remittances of Netherlands based Ghanaian Migrants on Rural Ashanti. Population, Space and Place, 10: 85-97. Khoser, K. 2003. New African Diasporas. London: Routledge. König, R. & de Regt, M. 2010. Family dynamics in transnational African migration to Europe: An introduction. African and Black Diaspora: An International Journal, 3(1): 1–15. Laczko F. 2005. Migration and development: Current Policy Challenges. In: T. Manuh (Ed.) At home in the world? International migration and development in contemporary Ghana and West Africa. Accra: Sub-Saharran Publishers, Legon. Leinaweaver, J. 2010 ‘Outsourcing care: how Peruvian migrants meet transnational family Obligations’. Latin American Perspectives, 37(5): 67-87. University of Ghana http://ugspace.ug.edu.gh 245 Levitt, P. 1998 ‘Social Remittances: A local-level, migration-driven form of cultural diffusion’. International Migration Review, 32(4): 926–948. Levitt, P. 2001. The Transnational Villagers. Berkeley, Los Angeles: University of California Press Levitt, P. & Nyberg-Sorensen N.2004. The Transnational Turn in Migration Studies. Global Migration Perspectives, 6: 2-13 Levitt, P., & Glick Schiller, N. 2004. Conceptualizing simultaneity: a transnational social field perspective on society. International Migration Review, 38 (3): 1002-1039 Levitt, P., & Jaworsky, B. N. 2007. Transnational migration studies: Past developments and future trends. Annual Review of Sociology, 33: 129-156 Lincoln, Y. E. G. (1985). Establishing Trustworthiness. Naturalist Inquiry. Newbury Park, CA: Sage. Loon V. J. 2001 “Ethnography: a critical turn in cultural studies”. In: P. Atkinson, A. Coffey, S. Delamont, J. Lofland, and L. Lofland (eds) Handbook of ethnography. London: Sage. 273-284 Lucas, R.E.B. and Stark, O. 1985. Motivations to Remit: Evidence from Botswana, Journal of Political Economy, 93(5): 901-918. Madiannou, M. and Miller, D. 2012. Migration and New Media: Transnational Families and Polymedia. London: Routledge. Mahler, S. J. 2001. Transnational relationships: The struggle to communicate across Borders, Identities Vol. 7 (4) : 583-619 Manuh T. 1999. “This Place is not Ghana”: Gender and Rights discourse among Ghanaian men and women in Toronto. Ghana Studies 2 : 77-95. Manuh, T. 2001. ‘Ghanaian Migrants in Toronto, Canada: Care of Kin and Gender Relations’. Research Review 17 New Series: 17-26. Massey, D. S., Arango, J., Hugo, G., Kouaouci, A., Pellegrino, A., & Taylor, J. E. 1998. Worlds in Motion. Understanding International Migration at the End of the Millennium. Oxford: Clarendon Press. Mauss, M., 1954. The gift: form and reason for exchange in archaic societies. London: Routledge. Mazzucato V. 2003. Asante Transnational Relations: Historical Renegotiation or Changing Structure of Social Relationships? Paper presented at CERES Summer School, Panel: Poverty in a World of Excess: Contradictions of Transnational Flows, 23-26 June 2003, Amsterdam. Mazzucato, V. 2005. Ghanaian migrants' double engagement: A transnational view of development and integration policies’. Global Migration Perspectives, 48: 1- 17. Mazzucato, V. 2006. Migrant transnationalism: Two-way flows, changing institutions and Community development between Ghana and the Netherlands. Economic Sociology, The European electronic newsletter, 7(3): 8-16. Retrieved on February 2, 2012, from http://econsoc.mpifg.de Mazzucato, V. 2007. Simultaneity and networks in transnational Migration: Lessons learned from an SMS methodology. In: DeWind, J. and Holdaway, J. (Eds.) Migration and Development Within and Across Borders. Geneva: International Organization for Migration. 71-102 University of Ghana http://ugspace.ug.edu.gh 246 Mazzucato V. 2008a. The Double Engagement: Transnationalism and Integration. Ghanaian migrants’ Lives between Ghana and The Netherlands. Journal of Ethnic and Migration Studies. 34(2):199-216 Mazzucato, V. 2008b. Transnational reciprocity: Ghanaian migrants and the care of their parents back home. In: E. Alber, S. Van der Geest, W. Geissler and S. Whyte (eds.) Generations in Africa:Connections and Conflicts, 111-133. Münster: LIT Verlag. 91-110 Mazzucato, V. 2009a Bridging boundaries with a transnational research approach: A simultaneous matched sample methodology. In: M. A. Falzon (Ed) Multi-sited ethnography: Theory, praxis and locality in contemporary social research, Chapter 12. Hampshire: Ashgate.215-132 Mazzucato V. 2009b. “Informal insurance arrangements in Ghanaian migrants’ transnational networks: the role of reverse remittances and geographic proximity,” World Development, 37(6): 1105-1115. Mazzucato, V. 2010. Reverse remittances in the migration development-nexus: Two way flows between Ghana and The Netherlands. Population, Space and Place, 17(5): 454-468 Mazzucato, V. 2013. Transnational families. In: I. Ness (Ed) The Encyclopedia of Global Human Migration. Oxford: Blackwell Publishing Mazzucato, V., van Dijk, R., C. Horst, & P. de Vries. 2004. Transcending the nation: Explorations of transnationalism as a concept and phenomenon. In: D. Kalb, W. Pansters and H. Siebers (Eds.) Globalization and Development: Themes and Concepts in Current Research. Dordrecht : Kluwer Academic Publishers 131-162 Mazzucato, V., Kabki, M. and Smith, L. 2006. “Transnational Migration and the Economy of Funerals: Changing Practices in Ghana”, Development and Change, Vol. 37(5):1047– 1072. Mazzucato, V., & Schans, D. 2008. “Transnational families, children and the migration- development nexus.” Paper presented at the SSRC conference “Migration & development: Future directions for research and policy”, Social Science Research Council, New York, 28 .February – 1 March Mazzucato, V., van den Boom, B., Nsowah-Nuamah, N. N. N. 2008. ‘Remittances in Ghana: Origin, Destination and Issues of Measurement’, International Migration, 46 (1): 103- 122. Mazzucato, V. and Schans, D. 2011. ‘Transnational families and the well-being of children: conceptual and methodological challenges’. Journal of Marriage and Family, 73(4): 704–712. Mazzucato V. and Manuh T. 2013. Transnational Families. Policy Brief TCRA project. http://fasos-research.nl/tcra/files/2013/05/TCRA-POLICY-BRIEF_def.pdf retrieved on 20/06/2014 Mazzucato, V. Dankyi, E. Poeze, M (forthcoming) Mapping transnational networks of care from a multi-actor and multi-sited perspective. In Children of migrants, intergenerational relations, and the transition to adulthood: Exploring Methodological Issues and Innovations, edited by Bolzmann, Bernardi, le Goff. Springer. McKenzie, S. and Menjívar, C. 2010. The meanings of migration, remittances and gifts: views of Honduran women who stay’, Global Networks, 11(1): 63-81. Menjivar C. 2012. Transnational parenting and immigration law: Central Americans in the United States. Journal of Ethnic and Migration Studies, 38 (2): 301-322 Menjívar, C., and Agadjanian, V. 2007. “Men's Migration and Women's Lives: Views from Rural Armenia and Guatemala.” Social Science Quarterly 88 ( 5):1243–1262 University of Ghana http://ugspace.ug.edu.gh 247 Mensah-Bonsu, H & Dowuonan Hammond C. 1996 “The child within the Ghananain Family”. In: E. Ardayfio-Schandorf (Ed) “The Changing Family in Ghana” Ghana Universities Press 1996. Merriam, S. B., Johnson-Bailey, J., Lee, M.-Y., Kee, Y., Ntseane, G., & Muhamad, M. (2001). Power and positionality: Negotiating insider/outsider status within and across cultures. International Journal of Lifelong Education, 20(5): 405-416. Ministry of Finance and Economic Planning. 2013. Multi-donor budget support in Ghana: Questions and answers. Retrieved from www.mofep.gov.gh/mdbs.cfm on 6/11/2013 Mullings, B. 1999. Insider or outsider, both or neither: Some dilemmas of interviewing in across- cultural setting. Geoforum, 30: 337-350. Moran-Taylor, M. J. 2008. When mothers and fathers migrate North: Caretakers, children, and child rearing in Guatamala. Latin American Perspectives, 35 (4): 79-95. Nanbigme E. 2004. Helping them grow their teeth: Care and conflict among the Dagaaba of the Northern Ghana. Research Review Supplement 16. Institute of African Studies National Board for Small Scale Industries. 2014. Retrieved from http://www.nbssi.org/AboutUs.aspx?PageId=16. Accessed on 13/06/2014 North, D. 1990. Institutions, institutional change and economic performance. Cambridge: Cambridge University Press. Notermans C. 2008. The emotional world of kinship: children’s experiences of fosterage in East Cameroon. Childhood. 15(3): 355-377 Nukunya G. K. 2003. Tradition and change in Ghana: An introduction to sociology. Accra: Ghana Universities Press Ocholla-Ayayo A.B. 1997. The African family between tradition and modernity. In: A. Adepoju (Ed), Family Population and Development in Africa. London: Zed Books. 60-77 Oheneba Sakyi Y. and Takyi B. K. 2006. African families at the turn of the 21st century. Westport CT: Praeger Olwig, K.F. 1999. Narratives of the children left behind: home and identity in globalised Caribbean families ,Journal of Ethnic and Migration Studies, 25(2): 267_84. Oppong C. 1981. Middle Class African Marriages, A family Study of Ghanaian Civil Servants. London: G. Allen & Unwin Oppong C. 1982 . Family structure and women's reproductive and productive roles. In: R. Anker, M. Buvinic and N. Youssef (eds.) Women's Roles and Population Trends in the ThirdWorld, London: Croom Helm.133-150 Oppong C. 2006. “Demographic Innovation and Nutirtion Catastrophe: Change, Lack of change and difference in Ghanaian Family systems In: Therborn G. (Ed). African Families in Global Context. Nordiska AfricanInstitutet, Uppsala. Oppong, C. 2012. Some biosocial Aspects of Childcare in Ghana: Decades of Change. In: Oppong, C., Badasu D, Waerness K. (eds) Childcare of a Globalising World Perspectives from Ghana.: 38-64. BRIC, Norway Oppong, C., Oppong, M. Y. and Odotei, I.K. (2006) in Sex and Gender in an era of AIDS- Ghana at the turn of the millennium. Accra: Sub-Saharan Publishers Orozco,M., Bump,M. and Fedewa R .2005. Diasporas, development and transnational integration:Ghanaians in the US, UK, and Germany. Washington, DC: Institute for the Study of International Migration and Inter-American Dialogue. Orozco M. 2006. Transnational Families: lives on the edge, but in pursuit of change. Institute for University of Ghana http://ugspace.ug.edu.gh 248 the study of international migration. Retrieved from http://www.thedialogue.org/PublicationFiles/Transnational%20Families%20Report%20w ith%20methodology.pdf on 21/06/2012 Osei-Fosu, A. K. 2011. Evaluating the impact of the capitation grant and the school feeding programme on enrollment, attendance and retention in schools: The case of Weweso Circuit. Journal of Science and Technology, 3(1): 55-65 Owusu, T. Y. 2000. ‘The Role of Ghanaian Immigrant Associations in Toronto, Canada’. International Migration Review, 34(4): 1151-1181. Page J. and Adams R. 2005. “The Impact of International Migration and Remittances on Poverty in Remittances” in: Remittances: Development Impact and Future prospects. Washington D.C.: The World Bank. : 277-306 Parreñas, R.S. (2001) ‘Mothering from a distance: Emotions, gender, and intergenerational relations in Filipino transnational families’, Feminist Studies, 27(2): 361-390. Parrenas R.S. 2005. Children of Global Migration: Transnational families and gendered woes. Stanford California: Stanford University Press Peil, M. (1995). ‘Ghanaians Abroad’. African Affairs 94:345-367. Pribilsky J. 2004. ‘Aprendemos a convivir’: conjugal relations, co-parenting, and family life among Ecuadorian transnational migrants in New York City and the Ecuadorian Andes. Global networks4 (3): 313-334 Pittman, P., Folsom, A., Bass, E., et al. 2007. U.S.-Based International Nurse Recruitment: Structure and Practices of a Burgeoning Industry. Washington, DC: Academy Health. Poeze M. and Mazzucato V. 2013 Ghanaian children in transnational families: Understanding the experiences of left behind children through local parenting norms: In L. Baldassar and L. Merla (eds.), Transnational Families, migration circulation of Care: Understanding Mobility and absence in family life. 149-169. Routledge Research in Transnationalism.149-169 Poeze, M. and V. Mazzucato. 2012. 'Transnational fathering between the Netherlands and Ghana: Fathers’ perceptions, obstacles and opportunities', Conference 'Rethinking Care: Anthropological Perspectives on Life Courses, Kin-Work and their trans-local entanglements', 6-7 December 2012, Berlin. Poeze M. and Dankyi E. 2013. "Eén familie, twee landen, twee onderzoeksters: ervaringen met een simultaneous matched sample methodology", Kwalon 18 (3): 41-46. Poeze, M. and V. Mazzucato. 2014. 'Transnational mothering and the law: Ghanaian women's restrictions to family reunion and consequences for family life', paper presented at the conference 'Transnational families: Multi-actor, multi-sited and institutional perspectives’, 26-28 June 2014, Amsterdam. Poeze M., Dankyi E., Mazzucato V. 2013. Navigating transnational relationships: Ghanaian migrant parents and their childrens’caregivers. Paper presented at TCRAf-Eu Conference, March, 26-29, 2013, Maastricht. Pottinger, A.M. and Brown, S.W. 2006 ‘Understanding the impact of parental migration on children: implications for counseling families from the Caribbean’, American Counseling Association_VISTAS Online; available at http://www.counselingoutfitters.com/Pottinger.htm accessed 8 August 2013 Quartey P. 2006. “Impact of Migrant Remittances on Household welfare in Ghana”, AERC Research Paper 158. African Economic Research Consortium. University of Ghana http://ugspace.ug.edu.gh 249 Quinn N. and Evans T. 2010. Understanding the care experiences: examples from Ghanaian context. International Social Work (53):61-74 Ratha, D. 2003 "Workers’ Remittances: An Important and Stable Source of External Development Finance" In Global Development Finance 2003. Washington D.C.:World Bank. 157-175. Ratha, D. 2007. ‘Leveraging Remittances for Development’, In: Hollifield, J.F., Orrenius, P.M. and Osang, T. (eds.).Migration, Trade and Development. Proceedings from the 2006 conference on Migration, Trade, and Development,Federal Reserve Bank of Dallas, Dalas Texas, 173-185 Ratha D. & Xu Z. Migration and Remittances Factbook 2008. Washington DC, the World Bank. Rebar, C. Gersch, C., Macnee, C. , Mc Cabe S. 2011. Understanding nursing research: Using research in evidence based practice. Baltimore: Lippincott, Williams &Wilkins Reyes M.M. 2007. Filipino children left behind: A literature review. UNICEF Salgado de Snyder, V. Nelly. 1993. Family Life across the Border: Mexican Wives Left Behind. Hispanic Journal of Behavioral Sciences 15(3): 391–401. Schmalzbauer, L. 2004. Searching for wages and mothering from afar: The case of Honduran transnational families. Journal of Marriage and Family, 66 (5): 1317-1331. Siddiqui, T. (2003). “Migration as a livelihood strategy of the poor: the Bangladesh case”, paper presented at the Regional Conference on Migration, Development and Pro-Poor Policy Choices in Asia, Dhaka, Bangladesh, 22-24 June 2003. http://livelihoods.org/hot_topics/docs/DfIDChildren.doc, accessed 7 February 2011. Smelser, Neil J. 1998. The Rational and the Ambivalent in the Social Sciences. American Sociological Review, 63: 1-15. Smith M. P. and Guarnizo L. E. (Eds) 1998. Transnationalism from Below. Vol. 6 New Brunswick, NJ: Transaction Publishers. 3-34 Smith, L. 2007 Tied to migrants: transnational influences on the economy of Accra,Ghana. Leiden: African Studies Centre Soros Foundation. 2007. Effects of Migration: Children Left at home. Bucharest. Spio A. E. 2011. The city branding of Accra in Keith Dinnie (ed) City branding: theory and cases . palgrave Macmillan . England: United Kingdom Spradley, J.P. 1979. The Ethnographic Interview. California: Wadsworth Stacey, J 1991. Can there be a feminist ethnography?: In: S. B. Gluck and D. Patai (Eds) Women’s words: The feminist practice of oral history. New York: Routledge. Stark, O. (1991), The migration of labour. Cambridge, Oxford: Blackwell. Stark, O. and Bloom, D.E. 1985. The New Economics of Labour Migration, American Economic Review, 75(2): 173-178. Stephen L. 2007. Women and social movements in Latin America: Power from Below. Austin: Univerisity of Texas. Suárez-Orozco, C., Todorova, I. L. G., and Louie, J. 2002. “Making up for lost time: The experience of separation and reunification among immigrant families.” Family Process, 41 (4): 625-643. Tanalega N. E. 2002. Families on the move. Ugat Foundation, Atenoe de Manila. Ter Haar, G. 2005.The Religious dimension in migration and its relation to development: The case of Ghanaians in the Netherlands. in T. Manuh (Ed) At Home in the World: University of Ghana http://ugspace.ug.edu.gh 250 International Migration and Development in Contemporary Ghana and West Africa,Accra: Sub-Saharan Publisher : 55-77 Teye J.K. 2012. Benefits, challenges, and dynamism of positionalities associated with mixed methods research in developing countries: Evidence from Ghana. Journal of Mixed Methods Research , 6(4).: 379-391. Thomas-Hope, E. 2002. Caribbean Migration. Jamaica: University of the West Indies. Thompson P. and Bauer E. “Jamaican transnational families: Points of pain and sources of resilience”. Wadabagei: A Journal of Caribbean and its Diaspora, 3(2): 1 -37 Tonah, Steve. 2007. Ghanaians Abroad and Their Ties Home: Cultural and Religious Dimensions of Transnational Migration, Working Paper No. 25, Center on Migration, Citizenship and Development (COMCAD), Bielefeld. Toyota, M., Yeoh, S.A. and Nguyen, L. 2007 “Editorial introduction: bringing the ‘left behind’ back into view in Asia: a framework for understanding the‘migration-left behind nexus’” Population, Space and Place Vol. 13: 157-161. Tronto J. 1998. An ethic of care. Generations 22, 3 ProQuest Research Library Tsuda T. 2012. Whatever Happened to Simultaneity? Transnational Migration Theory and Dual Engagement in Sending and Receiving Countries. Journal of Ethnic and Migration Studies, 38(4), pp 631-649. Twum-Baah, K. 2005 “Volume and Characteristics of international Ghanaian migration,” in T. Manuh (Ed) At Home in the World: International Migration and Development in Contemporary Ghana and West Africa,Accra: Sub-Saharan Publisher. 55-77 Twum-Danso, A. 2009 Reciprocity, respect and responsibility: The 3Rs underlying parent-child relationships in Ghana and the implications for children’s rights. The International Journal of Children’s Rights, 17(3): 415-432. Van der Geest, J. D. M., Whyte, S. R., & Alber, E. 2008. Generational connections and conflicts in Africa: An introduction. In E. Alber, S.Van der Geest, and S. R. Whyte (Eds.), Generations in Africa: Connections and conflicts. New Brunswick: Transaction Publishers. 1-23 Van der Geest, S. 1997. Between respect and reciprocity: Managing old age in Rural Ghana. Netherlands Kluwer Academic Publishers Van der Wiel, A. 2005 “Foreword” in T. Manuh (ed) At home in the world? International Migration and Development in Contemporary Ghana, West Africa. Accra: Sub-Saharan Publishers Verhoef, H. 2005 “A Child Has Many Mothers”: Views of Child Fostering in Northwestern Cameroon’, Childhood 12(3): 369–90. Verhoef, H. and Morelli, G. 2007. A child is a child: Fostering experiences in Northwestern Cameroon”. Ethos 35(2): 33-64. Vertovec, S. (1999). Conceiving and researching transnationalism. Ethnic and Racial Studies, 22 (2): 445-462. Vertovec, S. 2004 ‘Cheap calls: the social glue of migrant transnationalism’, Global Networks, 4 : 219–24. Vertovec, S. 2009. “Transnationalism”, Routledge, London. Wacquant, L. 2003 “Ethnografeast: A progress report on the practice and promise of Ethnography,” Ethnography 4(1): 5-14 University of Ghana http://ugspace.ug.edu.gh 251 Waerness K. 2004. The study of care in crosses cultural, interdisciplinary perspective. Research Review Supplement 16. Institute of African Studies Wetherell C. 1998. Historical Social Network Analysis. International Review of Social History. Supplement , 125-144 Williams, F. 2004. Rethinking Families. London: Calouste Gulbenkion Foundation Wimmer, A. and Glick Schiller, N. 2003. Methodological Nationalism, the Social Sciences, and the State of Migration: An Essay in Historical Epistemeology, International Migration Review, 37(3): 576-610. Wong M. 2006. “The gendered politics of remittances in Ghanaian Transnational Families” Economic Geography (82): 355-381 Wusu, O. and Isuigo- Abanihe, U. 2006. Interconnections among changing family structure, childrearing and fertility behaviour among the Ogu, South-western Nigeria: Demographic Research, 14(8): 139-156. Yeoh S. & Lam T 2006 the costs of (im) mobility: children left behind and children who migrate with a parent. Conference paper. UNESCAP Yin, R.K. 2003. Case Study Research: Design and Methods. Thousand Oaks California. Sage Zeitlin, M.F. and R. Megawangi. 1995. Modernization, Urbanization and Nutritional Care. Food and Nutrition bulletin, 16 (4): 396-406 Zentgraf K.M. and Chinchilla N. S. 2012. Transnational Family Separation: A Framework for Analysis. Journal of Ethnic and Migration Studies, 38 (2): 345-366 University of Ghana http://ugspace.ug.edu.gh 252 APPENDICES APPENDIX A: INTERVIEW GUIDE Interview Guide (This was adapted from the original guide that was prepared for the larger TCRA project. Interviews were conducted over a series of encounters with the respondents. Each encounter, whether it was an informal visit or to conduct formal interviews was aimed addressing an aspect of the thematic areas that made up the interview guide) o The caregiver’s current household and care responsibilities o E.g. Who lives in your household? Do you have the responsibility for more than one child? Are there other persons you have responsibility for (elderly or ill people)? o Decisions regarding care and caregivers o E.g.: Was the child already living with you before the parent(s) migrated or was the decision made because he or she was about to migrate? Were you part of making this decision or of not? How do you feel about it? Why did you accept taking care of the child? What were your expectations when you accepted to take care of the child? How have your expectations changed? o The organization of care o E.g. How is the care organized in practical and material ways? Do you get help in the up-bringing of the child? If yes, how are the responsibilities split? Does this/these other people take the child to activities, help them buy clothing, take them to the doctor, help them with home work, etc? What are the roles of different persons in giving care – or who does what? Are you compensated for the care you give? With money, things or for other things in return? How is the child(ren)’s kindergarten/ education paid/ apprenticeship paid for? And food, clothes, mobile credit, things for the house? How do you pay for the child’s medical care and medicines? (How) do you communicate with the parent(s)? o (If caregiver has own children) How do you juggle the care of your own children and that of the migrant’s child? Does this result in difficulties or is it easy to combine? o Education/Apprenticeship o E.g. Does the child go to school? How is he/she/they doing at school? Does the school cost money? Are you responsible for buying clothes and paying for the things the child needs for going to school? Do you help him or her with University of Ghana http://ugspace.ug.edu.gh 253 homework? Does the child have time for doing home work? Who goes to the meetings at school? who checks on his performance as an apprentice o Satisfaction and impact o E.g. How does the care arrangement function? Are you satisfied with it? What works and does not work? Are there unintended consequences? Are the aspects or situation that creates stress for you? What are the benefits of the arrangement for you, the parent(s) and the child? o Conflicts, negotiation and solutions o E.g. Have there been conflicts between you and the parent, or you and the child? Can you describe general changes in the TCRA arrangement or changes in where the child has been living? Has there been critical point or important situations in the child’s or your life that has caused conflicts? o Reciprocity, rights and responsibilities o E.g. Can you describe the expectations you have in relation to the child and the parent(s)? Is the child in any way helping in the household? Does he or she help with child care, cleaning, looking after elderly members of the household or other tasks? How did you come to an agreement with the parent? What are your responsibilities towards the parent(s) and the child? Do you have any expectations of support that you might receive from the child when you grow old? (If caregiver has children) Do you have any expectations of support from your own children when you grow old? 2. Life-story interview In this interview, follow up on themes that can be developed from the initial interview and the mapping exercise, but that is related specifically to the respondent and his or her personal history. o The respondent’s own care history: With whom did he or she grow up? Did she or he have responsibilities to look after during childhood and youth? Ask them to describe the household and the family so you can get an understanding of the different care responsibilities and who performed them. How did this change over time? o Childhood and education o Gender, care and social relationships: If she or he were raised by others than the parent(s), what is his or her view of how it was to be raised by a parent/caregiver? What were advantages/disadvantages? What does he or she think about their parents’ decision to raise him/ her in this way, etc. Also, what are people’s expectations for how they University of Ghana http://ugspace.ug.edu.gh 254 should care for their caregivers/parents when they grow old? What are their expectations for their children regarding the respondent’s own elderly care? Are there other expectations they have of their children in the future? University of Ghana http://ugspace.ug.edu.gh 255 APPENDIX B: GIFT SENT THROUGH ERNESTINA TO CAREGIVERS/CHILDREN Parent Items Recipient Remarks Joyce Toiletries(Deodorant, perfumes, tooth paste/brush, sanitary pads), Underwear(panties) socks , confectioneries(candies, cookies) All five children She is sending to caregiver’s children because they will complain if she does not send them anything and they will end up taking what she sends the children away from them. She feels she has to and not because she wants (she didn’t say that but that is the impression I get) Phone Annette Dress Hilda (for her bday) Perfume Caregiver (Evelyn) Little Shoes Socks, tooth brush For Caregivers granddaughter Akua Phones, tablets(android), shirts Her two boys She is sending to her brothers, brother’s wife and child because she cannot send to her own children without giving them. She feels she has to and not because she wants (she didn’t say that but that is the impression I get) Phone, fabric, bag, underwear Caregiver(her mother) Tee shirts Her brother’s Dress, blouse, underwear Brother’s wife (lives in the same house with her sons) Shirts, shorts, jeans, shoes For brother’s son Samuel Toiletries (Deodorants and perfume), small solar lamp, sweets (chocolates) canned fish Phone (used) For the children. Phone for the boy Didn’t send anything to Beatrice (caregiver). I also learnt she has left the kids again for the past three weeks. Not sure if that is why they didn’t send her anything Isaac Blouses His wife (CG) University of Ghana http://ugspace.ug.edu.gh 256 Shirts and underwear, Barbering tool kits Boys Kofi Laptop His son Obaa Dresses, Samsung tablet, shoes , perfumes Their daughter Obaa and her husband didn’t complain about giving things to everybody in the household. They felt pressured to give the other children will feel bad if only Abena and the caregiver’s son were singled out to receive the gifts. They were more concerned about how the children will feel than what others will say about them or the fact that the things will be withheld from Abena Clothing For caregiver’s husband Phone Caregiver Phone and a pair of shoes For caregiver’s son Shoes and clothing For caregiver’s step children University of Ghana http://ugspace.ug.edu.gh 257 APPENDIX C: NETWORK MAPPING TOOL No. Who are the important people for the care of the migrant's child, materially, emotionally and socially? Materially = eg paying for school fees; providing food; cooking for child, etc.; Emotionally= eg helps the child when he/she is feeling down; Socially= eg educates the child on how to behave. 1.1 Name 1.2 Relationship* to the child 1.3 Location 1 2 3 4 5 6 University of Ghana http://ugspace.ug.edu.gh 258 No. Type of care (mark an X next to all names of that are mentioned) 2.1 Emotional 2.2 Material 2.3 Social 1 2 3 4 5 6 7 University of Ghana http://ugspace.ug.edu.gh 259 No. 3. Questions to get additional names (mark an X next to all names of that are mentioned) (If some new names result, please ask questions 1.1-2.3 for these new people) 3.1 Are there other important people for the migrant's child whom we did not yet mention? 3.2 Who helps you the most in taking care of the migrant's child? 3.3 Who do you have arguments with regarding the child's care? 3.4 Who do you seek advice from regarding the child's care? 3.5 Who do you trust most regarding the child's upbringing? 1 2 3 4 5 6 7 University of Ghana http://ugspace.ug.edu.gh 260 No. 4. Schooling Very important = + Somewhat important Δ Not important = Ο 4.1 Who pays for the child's school fees? 4.2 Who provides for the child's school supplies 4.3 Who checks the child's progress in school (help with homework, attend meetings in school, visits the child in boarding school, resolves problems at school)? 1 2 3 4 5 6 University of Ghana http://ugspace.ug.edu.gh 261 No. 5. Health Very important = + Somewhat important Δ Not important = Ο 5.1 Who pays for the child's health expenses, like vaccinations, going to the doctor, medicines? 5.2 If the child is ill and needs to stay home, who looks after him/her? 5.3 Who decides whether the child needs to see a doctor, be taken to a clinic/hospital? 1 2 3 4 5 6 University of Ghana http://ugspace.ug.edu.gh 262 No. 6. Who decides? A lot of say = + Some say = Δ No say = Ο 6.1 When the child and the migrant parent communicate 6.2 What school the child should attend 6.3 That the child should live with you 6.4 When the child can receive money/gifts from the migrant parent 1 2 3 4 5 6 7 University of Ghana http://ugspace.ug.edu.gh